Accelerating genetic gains in legumes for the development of prosperous smallholder agriculture: integrating genomics, phenotyping, systems modelling and agronomy

Accelerating genetic gains in legumes for the development of prosperous smallholder agriculture:... Abstract Grain legumes form an important component of the human diet, provide feed for livestock, and replenish soil fertility through biological nitrogen fixation. Globally, the demand for food legumes is increasing as they complement cereals in protein requirements and possess a high percentage of digestible protein. Climate change has enhanced the frequency and intensity of drought stress, posing serious production constraints, especially in rainfed regions where most legumes are produced. Genetic improvement of legumes, like other crops, is mostly based on pedigree and performance-based selection over the past half century. To achieve faster genetic gains in legumes in rainfed conditions, this review proposes the integration of modern genomics approaches, high throughput phenomics, and simulation modelling in support of crop improvement that leads to improved varieties that perform with appropriate agronomy. Selection intensity, generation interval, and improved operational efficiencies in breeding are expected to further enhance the genetic gain in experimental plots. Improved seed access to farmers, combined with appropriate agronomic packages in farmers’ fields, will deliver higher genetic gains. Enhanced genetic gains, including not only productivity but also nutritional and market traits, will increase the profitability of farming and the availability of affordable nutritious food especially in developing countries. Drought stress, genetic gains, genomics-assisted breeding, legumes, rainfed agriculture Introduction Legumes are the third largest group among higher plants with more than 18 000 species in 650 genera. Grain legumes belong to subfamily Papilionoideae. Economically important food and feed legumes mostly fall in four clades of Papilionoideae: (i) genistoids, (ii) aeschynomenoids/dalbergioids, (iii) Hologalegina, and (iv) phaseoloids/millettioids. Recent molecular studies placed cool-season legumes in the inverted repeat loss clade of Hologalegina, while warm season legumes were included in the phaseoloid/millettioid clade (Wojciechowski et al., 2004). Cool season legumes include chickpea (Cicer arietinum), faba bean (Vicia faba), lentil (Lens culinaris), and pea (Pisum sativum). Warm season legumes include common bean (Phaseolus vulgaris), cowpea (Vigna unguiculata), pigeonpea (Cajanus cajan), soybean (Glycine max), and groundnut (Arachis hypogaea) (Doyle and Luckow, 2003). Signature features of legumes include biological nitrogen fixation in symbiotic association with bacteria and mycorrhiza, geotropic peg and pod development in groundnut, and vernalization in chickpea, each of which is of economic and scientific interest. Legumes form an important component of the human diet, provide animal feed, and replenish soil fertility through biological nitrogen fixation. In addition, legumes can recover unavailable forms of soil phosphorus, possibly lower the emission of greenhouse gases, and are prospective assets in future cropping systems (Stagnari et al., 2017). A recent review gives a good summary of the many virtues of food legumes and sets an agenda for their improvement in the decades to come (Sinclair and Vadez, 2012; Foyer et al., 2016). According to United States Census Bureau International Database, the global population is predicted to approach 9.8 billion by 2050. Further, the Global Alliance for Improved Nutrition, which works to strengthen the availability of nutritious and affordable foods for infants and women, estimated that 2 billion people across the world are undernourished and 2.6 million children die due to malnutrition (http://www.gainhealth.org/knowledge-centre/fast-facts-malnutrition/). The majority of crop improvement programs, however, have emphasized only crop productivity traits in the past including during the green revolution. As a result, despite the attaining of global food security, a continuous increase in micronutrient malnutrition has been recorded in most of the developing countries. Under such a scenario, the grain legumes are relevant in the fight against malnutrition and are considered to provide ‘nutritious seeds for a sustainable future’. Therefore, a comprehensive strategy and dedicated effort are required to produce more and dense nutritious legume crops to ensure food and nutritional security. Climate change has significant adverse impacts on all components of crop production—area, intensity, and yield— and hence producing more food to feed the growing population is a great challenge before agriculturists and other stakeholders. The World Resources Institute predicted that by 2040, the USA, China, and India are expected to face 40–70% more water stress. Globally the demand for food legumes is increasing as legumes may complement cereals in protein requirements and possess a high percentage of digestible protein. The majority of the area under legumes is in Asia (~76%), principally soybean, beans, groundnut, and chickpea. More recent studies indicate that between 1961 and 2014, 75% of the legume area in South America and 90% of that in North America are under soybean cultivation (FAOSTAT, 2016). In Sub-Saharan Africa, food legumes are cultivated on 20 million ha, and 44% of production comes from cowpea and 31% from dry beans. Sub-Saharan Africa alone contributes 54% of global cowpea production (Akibode, 2011). High tolerance to environmental stresses of many rainfed legumes, when compared with soybean, is required for sustainable production in often harsh soil and climatic conditions in dryland agriculture (Foyer et al., 2016; Sita et al., 2017). Erratic rainfall increases the occurrence of drought stress necessitating drought-tolerant legume varieties that can produce in water-limited environments. Drought tolerance is a very complex trait and hence an in-depth understanding of the genomic control and molecular mechanism of drought tolerance is essential for effective deployment in breeding. Genetic gain can be defined as the rate of increase in yield over a given time period—pure genetic gain is estimated against potential yield, but can also be assessed under defined stress conditions. Barker et al. (2010) reported a maize genetic gain of 211 kg ha−1 year−1 in favorable environments in Woodland, CA, USA. Further, the same study reported lower genetic gains under water limited environments, of 124 and 91 kg ha−1 year−1 expressed during flowering and mid-grain filling, respectively. Recently, by using marker-assisted recurrent selection (MARS) in maize, a 3% increase in yield per cycle under water stress conditions was reported (Bankole et al., 2017). Among legumes, Koester et al., (2014) reported a genetic gain of 26.5 kg ha–1 year–1 by studying the historical data of 80 years of soybean breeding at the Crop Research and Education Center in Urbana (IL, USA). This gain in grain yield is attributed to increases in light interception, energy conversion, and partitioning efficiencies. A contemporary question in legume improvement is how to enhance the synergy among the different disciplines of genomics, phenomics, crop physiology, growth modelling, and agronomy. In this article, we discuss the technological advances in different disciplines and advocate their integrated use both to understand the genetics of traits and to deploy them in crop improvement programs of soybean, chickpea, pigeonpea, and groundnut. A special emphasis has been placed on deployment of genomic selection in breeding programs to attaining faster genetic gain. Technological advances Agricultural research has benefitted from new technologies that have helped in achieving current food sufficiency in many parts of the world. However, to realize future demands, researchers need to apply new tools, technologies, and partnerships to achieve even higher productivity and profitability of farming and better nutrition for all consumers. Emerging crop improvement technologies are outlined in the following sections. Sequencing and genotyping During the past two decades, advances in molecular marker technologies and next-generation sequencing technologies have enhanced our understanding of several traits in crop plants including food legumes (Varshney et al., 2015). Genome sequences for most of the grain legumes are now available, for instance soybean (Schmutz et al., 2010), groundnut progenitors (Bertioli et al., 2016; Chen et al., 2016), chickpea (Varshney et al., 2013d; Parween et al., 2015), pigeonpea (Varshney et al., 2012a), common bean (Schmutz et al., 2014; Yang et al., 2015), and adzuki bean (Kang et al., 2015; Yang et al., 2015) (Table 1). Efforts are underway to sequence the remaining legume genomes. The past decade has witnessed an exponential increase in availability of genomic resources and their deployment in trait discovery and breeding (see Bohra et al., 2014; Varshney et al., 2015; Pandey et al., 2016). As a result, several of these legume crops have genomic resources and associated phenotypic data to support genomics-based discovery and breeding approaches to develop superior legume varieties. Table 1. Summary of genome sequence, resequencing efforts and available high throughput genotyping platforms in important legumes Crop Genome sequence Germplasm lines resequenced High throughput genotyping platforms Soybean • 85% of Glycine max var William 82 genome (1115 Mb); 46 430 protein coding genes (Schmutz et al., 2010) • 915.4 Mb of G. soja var. IT182932 (Kim et al., 2010) • 106 (7 wild, 43 landraces and 56 elite; Valliyodan et al., 2016) • 302 (62 wild, 130 landraces and 110 elite; Zhou et al., 2015a) • 28 (commercial cultivars; Maldonado dos Santos et al., 2016) • 89 lines (Lee et al., 2015) • 286 (14 wild, 153 landraces and 119 elite; Zhou et al., 2015b) • Illumina 384 SNP VeraCode assays (Lee et al., 2015) • NJAU 355K SoySNP array (Wang et al., 2016) • Illumina Infinium SoySNP6K BeadChip (Akond et al., 2013) • SoySNP50K array (Song et al., 2013) • 384 SNP GoldenGate assay (Hyten et al., 2008) • SoyaSNP180K Axiom (Lee et al., 2015) Pigeonpea • 72.7% of Cajanus cajan var Asha genome (833.07 Mb); 48 680 protein coding genes (Varshney et al., 2012a) • 292 lines (Varshney et al., 2017) • 20 (crossing parentals of recombinant inbred lines, introgression lines, MAGIC and NAM population; Kumar et al., 2016) • 60K Axiom®Cajanus SNP array (Saxena et al., 2017 and unpublished) Chickpea • 72.12% of Cicer arietinum var CDC Frontier genome (738 Mb); 28 269 protein coding genes (Varshney et al., 2013d) • 416 Mb of C. reticulatum var PI 489777 genome; 25 680 protein coding genes (Gupta et al., 2017) • 35 (parental genotypes of mapping populations; Thudi et al., 2016a) • 129 released varieties (Thudi et al., 2016b) • 300 lines (ICRISAT, unpublished) • 3000 lines (ICRISAT, unpublished) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) • 60K Axiom®Cicer SNP array (Roorkiwal et al., 2017) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) Groundnut • 1081 Mb of Arachis duranensis V14167 and 1371 Mb of A. ipaensis K30076 (Bertioli et al., 2016) • A. duranensis var PI475845; 50 324 protein coding genes (Chen et al., 2016) • 11 genotypes including synthetics and their diploid parents (Chen et al., 2016) • 41 diverse genotypes (30 tetraploids and 11 diploids) (Clevenger et al., 2017; Pandey et al., 2017a) • 58K Axiom®Arachis SNP array (Pandey et al., 2017a) •1536 SNP GoldenGate assay (Nagy et al., 2012) Common bean • 80.57% of Phaseolus vulgaris var G19833 genome (587 Mb); 26 279 protein coding genes (Schmutz et al., 2014) • 17 varieties (Song et al., 2015) • BARCBean6K_1, BARCBean6K_2 chips, BARCBean6K_3 SNP chips (Song et al., 2013, 2015) Mung bean • 84.48% of Vigna radiata var radiata VC1973A geome (548 Mb); 22 427 predicted genes (Kang et al., 2014) • 84.48% of Vigna radiata var sublobata TC1966A geome (501 Mb); 22 834 predicted genes (Kang et al., 2014) • 81.81% of Vigna radiata var glabra V1160 geome (968 Mb); 41 484 predicted genes (Kang et al., 2014) — — Adzuki bean • 96.56% of Vigna angularis var Gyeongwon genome (612 Mb); 21 532 predicted genes (Kang et al., 2015) • 83.02% of Vigna angularis var Jingnong geome (542 Mb); 34 183 predicted genes (Yang et al., 2015) — — Cowpea • 323 Mb of Phaseolus vulgaris var IT97K genome (724 Mb) (Muñoz-Amatriaín et al., 2017) • 36 diverse accessions (Muñoz-Amatriaín et al., 2017) • Cowpea iSelect Consortium Array (Muñoz-Amatriaín et al., 2017) • 1536 SNP GoldenGate assays (Muchero et al., 2009) Pea https://www.france-genomique.org/spip/spip. php?article141&lang=fr — • GenoPea 13.2K SNP Array (Tayeh et al., 2015a) • 384 SNP GoldenGate assays (Deulvot et al., 2010) Crop Genome sequence Germplasm lines resequenced High throughput genotyping platforms Soybean • 85% of Glycine max var William 82 genome (1115 Mb); 46 430 protein coding genes (Schmutz et al., 2010) • 915.4 Mb of G. soja var. IT182932 (Kim et al., 2010) • 106 (7 wild, 43 landraces and 56 elite; Valliyodan et al., 2016) • 302 (62 wild, 130 landraces and 110 elite; Zhou et al., 2015a) • 28 (commercial cultivars; Maldonado dos Santos et al., 2016) • 89 lines (Lee et al., 2015) • 286 (14 wild, 153 landraces and 119 elite; Zhou et al., 2015b) • Illumina 384 SNP VeraCode assays (Lee et al., 2015) • NJAU 355K SoySNP array (Wang et al., 2016) • Illumina Infinium SoySNP6K BeadChip (Akond et al., 2013) • SoySNP50K array (Song et al., 2013) • 384 SNP GoldenGate assay (Hyten et al., 2008) • SoyaSNP180K Axiom (Lee et al., 2015) Pigeonpea • 72.7% of Cajanus cajan var Asha genome (833.07 Mb); 48 680 protein coding genes (Varshney et al., 2012a) • 292 lines (Varshney et al., 2017) • 20 (crossing parentals of recombinant inbred lines, introgression lines, MAGIC and NAM population; Kumar et al., 2016) • 60K Axiom®Cajanus SNP array (Saxena et al., 2017 and unpublished) Chickpea • 72.12% of Cicer arietinum var CDC Frontier genome (738 Mb); 28 269 protein coding genes (Varshney et al., 2013d) • 416 Mb of C. reticulatum var PI 489777 genome; 25 680 protein coding genes (Gupta et al., 2017) • 35 (parental genotypes of mapping populations; Thudi et al., 2016a) • 129 released varieties (Thudi et al., 2016b) • 300 lines (ICRISAT, unpublished) • 3000 lines (ICRISAT, unpublished) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) • 60K Axiom®Cicer SNP array (Roorkiwal et al., 2017) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) Groundnut • 1081 Mb of Arachis duranensis V14167 and 1371 Mb of A. ipaensis K30076 (Bertioli et al., 2016) • A. duranensis var PI475845; 50 324 protein coding genes (Chen et al., 2016) • 11 genotypes including synthetics and their diploid parents (Chen et al., 2016) • 41 diverse genotypes (30 tetraploids and 11 diploids) (Clevenger et al., 2017; Pandey et al., 2017a) • 58K Axiom®Arachis SNP array (Pandey et al., 2017a) •1536 SNP GoldenGate assay (Nagy et al., 2012) Common bean • 80.57% of Phaseolus vulgaris var G19833 genome (587 Mb); 26 279 protein coding genes (Schmutz et al., 2014) • 17 varieties (Song et al., 2015) • BARCBean6K_1, BARCBean6K_2 chips, BARCBean6K_3 SNP chips (Song et al., 2013, 2015) Mung bean • 84.48% of Vigna radiata var radiata VC1973A geome (548 Mb); 22 427 predicted genes (Kang et al., 2014) • 84.48% of Vigna radiata var sublobata TC1966A geome (501 Mb); 22 834 predicted genes (Kang et al., 2014) • 81.81% of Vigna radiata var glabra V1160 geome (968 Mb); 41 484 predicted genes (Kang et al., 2014) — — Adzuki bean • 96.56% of Vigna angularis var Gyeongwon genome (612 Mb); 21 532 predicted genes (Kang et al., 2015) • 83.02% of Vigna angularis var Jingnong geome (542 Mb); 34 183 predicted genes (Yang et al., 2015) — — Cowpea • 323 Mb of Phaseolus vulgaris var IT97K genome (724 Mb) (Muñoz-Amatriaín et al., 2017) • 36 diverse accessions (Muñoz-Amatriaín et al., 2017) • Cowpea iSelect Consortium Array (Muñoz-Amatriaín et al., 2017) • 1536 SNP GoldenGate assays (Muchero et al., 2009) Pea https://www.france-genomique.org/spip/spip. php?article141&lang=fr — • GenoPea 13.2K SNP Array (Tayeh et al., 2015a) • 384 SNP GoldenGate assays (Deulvot et al., 2010) View Large Table 1. Summary of genome sequence, resequencing efforts and available high throughput genotyping platforms in important legumes Crop Genome sequence Germplasm lines resequenced High throughput genotyping platforms Soybean • 85% of Glycine max var William 82 genome (1115 Mb); 46 430 protein coding genes (Schmutz et al., 2010) • 915.4 Mb of G. soja var. IT182932 (Kim et al., 2010) • 106 (7 wild, 43 landraces and 56 elite; Valliyodan et al., 2016) • 302 (62 wild, 130 landraces and 110 elite; Zhou et al., 2015a) • 28 (commercial cultivars; Maldonado dos Santos et al., 2016) • 89 lines (Lee et al., 2015) • 286 (14 wild, 153 landraces and 119 elite; Zhou et al., 2015b) • Illumina 384 SNP VeraCode assays (Lee et al., 2015) • NJAU 355K SoySNP array (Wang et al., 2016) • Illumina Infinium SoySNP6K BeadChip (Akond et al., 2013) • SoySNP50K array (Song et al., 2013) • 384 SNP GoldenGate assay (Hyten et al., 2008) • SoyaSNP180K Axiom (Lee et al., 2015) Pigeonpea • 72.7% of Cajanus cajan var Asha genome (833.07 Mb); 48 680 protein coding genes (Varshney et al., 2012a) • 292 lines (Varshney et al., 2017) • 20 (crossing parentals of recombinant inbred lines, introgression lines, MAGIC and NAM population; Kumar et al., 2016) • 60K Axiom®Cajanus SNP array (Saxena et al., 2017 and unpublished) Chickpea • 72.12% of Cicer arietinum var CDC Frontier genome (738 Mb); 28 269 protein coding genes (Varshney et al., 2013d) • 416 Mb of C. reticulatum var PI 489777 genome; 25 680 protein coding genes (Gupta et al., 2017) • 35 (parental genotypes of mapping populations; Thudi et al., 2016a) • 129 released varieties (Thudi et al., 2016b) • 300 lines (ICRISAT, unpublished) • 3000 lines (ICRISAT, unpublished) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) • 60K Axiom®Cicer SNP array (Roorkiwal et al., 2017) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) Groundnut • 1081 Mb of Arachis duranensis V14167 and 1371 Mb of A. ipaensis K30076 (Bertioli et al., 2016) • A. duranensis var PI475845; 50 324 protein coding genes (Chen et al., 2016) • 11 genotypes including synthetics and their diploid parents (Chen et al., 2016) • 41 diverse genotypes (30 tetraploids and 11 diploids) (Clevenger et al., 2017; Pandey et al., 2017a) • 58K Axiom®Arachis SNP array (Pandey et al., 2017a) •1536 SNP GoldenGate assay (Nagy et al., 2012) Common bean • 80.57% of Phaseolus vulgaris var G19833 genome (587 Mb); 26 279 protein coding genes (Schmutz et al., 2014) • 17 varieties (Song et al., 2015) • BARCBean6K_1, BARCBean6K_2 chips, BARCBean6K_3 SNP chips (Song et al., 2013, 2015) Mung bean • 84.48% of Vigna radiata var radiata VC1973A geome (548 Mb); 22 427 predicted genes (Kang et al., 2014) • 84.48% of Vigna radiata var sublobata TC1966A geome (501 Mb); 22 834 predicted genes (Kang et al., 2014) • 81.81% of Vigna radiata var glabra V1160 geome (968 Mb); 41 484 predicted genes (Kang et al., 2014) — — Adzuki bean • 96.56% of Vigna angularis var Gyeongwon genome (612 Mb); 21 532 predicted genes (Kang et al., 2015) • 83.02% of Vigna angularis var Jingnong geome (542 Mb); 34 183 predicted genes (Yang et al., 2015) — — Cowpea • 323 Mb of Phaseolus vulgaris var IT97K genome (724 Mb) (Muñoz-Amatriaín et al., 2017) • 36 diverse accessions (Muñoz-Amatriaín et al., 2017) • Cowpea iSelect Consortium Array (Muñoz-Amatriaín et al., 2017) • 1536 SNP GoldenGate assays (Muchero et al., 2009) Pea https://www.france-genomique.org/spip/spip. php?article141&lang=fr — • GenoPea 13.2K SNP Array (Tayeh et al., 2015a) • 384 SNP GoldenGate assays (Deulvot et al., 2010) Crop Genome sequence Germplasm lines resequenced High throughput genotyping platforms Soybean • 85% of Glycine max var William 82 genome (1115 Mb); 46 430 protein coding genes (Schmutz et al., 2010) • 915.4 Mb of G. soja var. IT182932 (Kim et al., 2010) • 106 (7 wild, 43 landraces and 56 elite; Valliyodan et al., 2016) • 302 (62 wild, 130 landraces and 110 elite; Zhou et al., 2015a) • 28 (commercial cultivars; Maldonado dos Santos et al., 2016) • 89 lines (Lee et al., 2015) • 286 (14 wild, 153 landraces and 119 elite; Zhou et al., 2015b) • Illumina 384 SNP VeraCode assays (Lee et al., 2015) • NJAU 355K SoySNP array (Wang et al., 2016) • Illumina Infinium SoySNP6K BeadChip (Akond et al., 2013) • SoySNP50K array (Song et al., 2013) • 384 SNP GoldenGate assay (Hyten et al., 2008) • SoyaSNP180K Axiom (Lee et al., 2015) Pigeonpea • 72.7% of Cajanus cajan var Asha genome (833.07 Mb); 48 680 protein coding genes (Varshney et al., 2012a) • 292 lines (Varshney et al., 2017) • 20 (crossing parentals of recombinant inbred lines, introgression lines, MAGIC and NAM population; Kumar et al., 2016) • 60K Axiom®Cajanus SNP array (Saxena et al., 2017 and unpublished) Chickpea • 72.12% of Cicer arietinum var CDC Frontier genome (738 Mb); 28 269 protein coding genes (Varshney et al., 2013d) • 416 Mb of C. reticulatum var PI 489777 genome; 25 680 protein coding genes (Gupta et al., 2017) • 35 (parental genotypes of mapping populations; Thudi et al., 2016a) • 129 released varieties (Thudi et al., 2016b) • 300 lines (ICRISAT, unpublished) • 3000 lines (ICRISAT, unpublished) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) • 60K Axiom®Cicer SNP array (Roorkiwal et al., 2017) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) Groundnut • 1081 Mb of Arachis duranensis V14167 and 1371 Mb of A. ipaensis K30076 (Bertioli et al., 2016) • A. duranensis var PI475845; 50 324 protein coding genes (Chen et al., 2016) • 11 genotypes including synthetics and their diploid parents (Chen et al., 2016) • 41 diverse genotypes (30 tetraploids and 11 diploids) (Clevenger et al., 2017; Pandey et al., 2017a) • 58K Axiom®Arachis SNP array (Pandey et al., 2017a) •1536 SNP GoldenGate assay (Nagy et al., 2012) Common bean • 80.57% of Phaseolus vulgaris var G19833 genome (587 Mb); 26 279 protein coding genes (Schmutz et al., 2014) • 17 varieties (Song et al., 2015) • BARCBean6K_1, BARCBean6K_2 chips, BARCBean6K_3 SNP chips (Song et al., 2013, 2015) Mung bean • 84.48% of Vigna radiata var radiata VC1973A geome (548 Mb); 22 427 predicted genes (Kang et al., 2014) • 84.48% of Vigna radiata var sublobata TC1966A geome (501 Mb); 22 834 predicted genes (Kang et al., 2014) • 81.81% of Vigna radiata var glabra V1160 geome (968 Mb); 41 484 predicted genes (Kang et al., 2014) — — Adzuki bean • 96.56% of Vigna angularis var Gyeongwon genome (612 Mb); 21 532 predicted genes (Kang et al., 2015) • 83.02% of Vigna angularis var Jingnong geome (542 Mb); 34 183 predicted genes (Yang et al., 2015) — — Cowpea • 323 Mb of Phaseolus vulgaris var IT97K genome (724 Mb) (Muñoz-Amatriaín et al., 2017) • 36 diverse accessions (Muñoz-Amatriaín et al., 2017) • Cowpea iSelect Consortium Array (Muñoz-Amatriaín et al., 2017) • 1536 SNP GoldenGate assays (Muchero et al., 2009) Pea https://www.france-genomique.org/spip/spip. php?article141&lang=fr — • GenoPea 13.2K SNP Array (Tayeh et al., 2015a) • 384 SNP GoldenGate assays (Deulvot et al., 2010) View Large Besides disclosing the blueprint of genomes of several legumes, different research groups also re-sequenced legume germplasm lines, which further increased our understanding of genome architecture, structural variations, genome evolution, and genome dynamics during domestication. Several million structural variations that aid in trait mapping and trait improvement were reported (Kumar et al., 2016; Thudi et al., 2016b). In addition, genome-wide single nucleotide polymorphisms (SNPs) were also used to identify significant marker trait associations for economically important traits (Zhou et al., 2015a; Varshney et al., 2017). Resequencing germplasm lines also enabled us to understand the spatial and temporal trends in diversity in released varieties of chickpea (Thudi et al., 2016a), cultivated and wild accessions of soybean (Lam et al., 2010; Zhou et al., 2015a), and a reference set of pigeonpea (Varshney et al., 2017; Table 1). Resequencing of 28 Brazilian soybean cultivars suggested that, despite the diversification of modern Brazilian cultivars, the soybean germplasm remains very narrow because of the large number of genome regions that exhibit low diversity (Maldonado dos Santos et al., 2016). In recent years, genotyping by sequencing, skim sequencing, diversity array technology (DArT)-seq and restriction site associated DNA sequencing approaches were also employed for developing high density genetic maps, refining the QTL mapping and identifying trait linked markers in legumes (Jaganathan et al., 2015; Kale et al., 2015; Contreras-Soto et al., 2017; Leamy et al., 2017; Valdisser et al., 2017). Nevertheless, the percentage of missing data points and the SNP calling rates greatly reduce the number of final SNPs for different studies and this has been a concern. To overcome these constraints several high-throughput SNP genotyping platforms, such as Veracode assays, Illumina GoldenGate assays, Infinium chips and Axiom arrays, are now available, which not only increase the precision of SNP calling but also enable genotyping of larger populations at reduced cost (Table 1). Axiom arrays with >50K SNPs with uniform genome coverage were developed and are being used for germplasm characterization, trait mapping and molecular breeding in chickpea (Roorkiwal et al., 2017), pigeonpea (ICRISAT unpublished data), groundnut (Pandey et al., 2017a) and soybean (Lee et al., 2015; Table 1). High-density and precise phenotyping Understanding the complex relationships between genotype and phenotype is a major challenge in plant sciences in view of the multi-scale nature of phenotypes and of the large phenotypic plasticity displayed by plants subjected to varying environmental conditions that results in different values of phenotypic variables and different rankings of genotypes in different experiments. It requires (i) capturing information on physiological traits and performance of large numbers of plants, together with their environment, (ii) analysing and organizing the resulting datasets, and (iii) developing models able to disentangle and simulate plant behavior in a range of scenarios (Tardieu et al., 2017). Good and relevant phenotyping starts with a detailed understanding of plant biological processes underlying tolerance/resistance to well-defined constraints. For instance, tolerance of chickpea to terminal drought in South Asia depends on plant traits that ensure the availability of water during the grain filling phase (Zaman-Allah et al., 2011; Vadez et al., 2013a; Vadez 2014). Therefore, phenotyping first needs to be well targeted and defined. Precise phenotyping is the next step and involves the design of protocols and approaches that ensure the assessment of key traits at the highest rate and the cheapest cost. The cost of phenotyping increases with increase in population size, especially in the case of deployment of genomic selection (see later), which requires a larger population for high prediction accuracy. The availability of non-destructive root phenotyping methods has, in the past, limited our understanding of traits involved in drought tolerance of legumes such as stomatal control or root system architecture. Hence, genome-wide association studies of traits associated with drought tolerance, based on large-scale germplasm re-sequencing (Varshney et al., 2017), require a suite of high-throughput and precise phenotyping technologies for capturing, analysing, and interpreting images in order to infer heritable traits that can be analysed genetically (Tardieu et al., 2017). Recognizing the importance of phenotyping, in recent years several large public, private and academic efforts have led to the establishment of state-of-the-art phenotyping facilities word-wide in an exponential phase of automated high-throughput phenotyping (Table 2). These include the German and French Plant Phenotyping Networks (http://www.dppn.de; https://www.phenome-fppn.fr/), European Plant Phenotyping Network (2020) (EPPN2020; https://eppn2020.plant-phenotyping.eu), European Plant Phenotyping Infrastructure (EMPHASIS), and International Plant Phenotyping Network (IPPN; https://www.plant-phenotyping.org/). These infrastructure networks enable access to the necessary tools for phenotyping, in particular robot-assisted image capture (Cooper et al., 2009; Fiorani and Schurr, 2013), statistical designs and models for extracting relevant physiological variables from raw data (Cabrera-Bosquet et al., 2016), and specialized information systems managing large datasets originating from phenotyping experiments (Tardieu et al., 2017). GrowScreen-PaGe is a non-invasive, high-throughput phenotyping system developed at the Institute of Biosciences and Geosciences suitable for investigating root systems and root plasticity of large sets of both dicots and monocots (Gioia et al., 2016). The LeasyScan platform developed at the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) offers a novel 3D scanning technique that capture leaf area development continuously and phenotype traits controlling plant water budget (Vadez et al., 2015). The University of Western Australia, Australia has developed a semi-hydroponic phenotyping system that examines architecture of deep root system and is amenable for phenotyping large sets of plants in a relatively small space (Chen et al., 2011). The salient features of several high throughput phenotyping platforms, such as PHENOPSIS (Bresson et al., 2015; Granier et al., 2006), PhenoArch (Cabrera-Bosquet et al., 2016), DEEPER PHENOSCOPE (Bresson et al., 2015), and RADIX (Le Marié et al., 2016), are summarized in Table 2. Table 2. A list of key high-throughput phenotyping platforms Phenotyping platform (and institute or company) Salient features Reference LeasyScan phenotyping platform (ICRISAT, India) • A novel 3D scanning technique to capture leaf area development continuously • Scanner-to-plant concept to increase imaging throughput and analytical scales to combine gravimetric transpiration measurements. • Combines 3D imaging and lysimetry for high- throughput phenotyping of traits controlling plant water budget Vadez et al. (2015) Semi-hydroponic phenotyping system (The University of Western Australia, Australia) • Permits mapping and digital measurement of dynamic growth of taproot and lateral roots • Desirable tool for examining root architecture of deep root systems and large sets of plants in a relatively small space Chen et al. (2011) DEEPER: an integrated phenotyping platform for deeper roots (PennState Colleage of Agriculture, USA) • A platform for identifying the traits of deeper rooted crops in non-destructive field phenotyping of rooting depth, root modeling, high-throughput 3D imaging of root architecture, and anatomy http://plantscience.psu.edu/research/labs/roots/projects/ deeper-an-integrated-phenotyping-platform-for-deeper-roots GLO-Roots: luminescence-based imaging system (Dinnenylab, USA) • Combines custom-made growth vessels and new image analysis algorithms to non-destructively monitor RSA development over space (2D) and time • Allows information on soil properties (e.g. moisture) to be integrated with root growth data • Makes use of luminescence imaging of roots expressing plant codon-optimize luciferase Rellán-Álvarez et al. (2015) X-Ray computed tomography (University of Nottingham, UK) • Non-destructively visualizes opaque root structures by measuring the attenuation of ionizing radiation as it passes through the root • A series of projections are acquired and combined to reconstruct a 3D image of the root system Mairhofer et al. (2012), Mooney et al. (2012) Rhizophonics (University of Liège, Belgium) • Combines hydroponics and rhizotrons • System is made of a nylon fabric supported by an aluminum frame • The set-up is immersed in a tank filled with liquid medium • Allows non-destructive, 2D imaging of root architecture while simultaneously sampling shoots Mathieu et al. (2015) Clear pot method (The University of Queensland, Australia) • Uses transparent pots filled with soil or other p otting media • Seeds are planted close to the pot wall to enable high- throughput imaging of roots along the clear pot wall • To prevent light exposure, the clear pot is placed in black pots while roots are developing Richard et al. (2015) Rhizoslides (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • The set-up consists of a Plexiglas sheet covered with moistened germination paper. Seeds are planted on the slit of the Plexiglas • Allows separation of crown roots from embryonic roots Le Marié et al. (2014) Shovelomics (PennState Colleage of Agriculture, USA) • Involves manual excavation of plants and separating roots from shoots • Washed roots are then placed on a phenotyping board for root trait quantification • New algorithms allow extraction of several root traits in a high throughput manner Trachsel et al. (2011), Bucksch et al. (2014) Soil coring (CSIRO and The University of Queensland, Australia) • Uses a tractor-mounted, hydraulic soil corer to drive steel alloy sampling tubes into the soil • When combined with novel planting configurations (e.g. hill plots), this method allows for phenotyping deep rooted crop varieties Wasson et al. (2014) Rhizo-lysimetry (E. H. Graham Centre for Agricultural Innovation, Australia) • Elaborate facility consisting of an underground corridor and concrete silos and pipes to house soil-containing soil cores for direct root observation Eberbach et al. (2013) Minirhizotrons • A transparent observation tube permanently inserted in the soil. Images of roots growing along the minirhizotron wall at particular locations in the soil profile can be captured over time Iversen et al. (2011), Maeght et al. (2013) RhizoTube (INRA, France) • Cylindrical rhizotrons that allow full 2D visualization of the root system of a single or up to six plants simultaneously • The RhizoCab is designed to take images of the entire root systems of plants growing in RhizoTubes, and also permits a focus on some parts of the root systems Jeudy et al. (2016) RADIX (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • Rhizoslide platform allowing high throughput digital image analysis of root system expansion Le Marié et al. (2016) Scanner bank (The James Hutton Institute, UK) • Low-cost, high-resolution root phenotyping platform, requiring no sophisticated equipment and adaptable to most laboratory and glasshouse environments, and applied to quantify environmental and temporal variation in root traits Adu et al. (2014) The Plant Accelerator® (The University of Adelaide, Australia) • The facility offers modern plant growth environments and state-of-the-art high-throughput automated imaging and computing technologies to monitor the performance of plants under different environmental conditions (e.g. which genotype performs best under drought stress) • Research projects facilitated by this technology vary from large scale screening of early growth, salinity tolerance to water and nutrient use efficiency http://www.plantphenomics.org.au/services/accelerator/ DroughtSpotter (The University of Adelaide, Australia) • A gravimetric platform with precision irrigation to assess transpiration dynamics of plants with a precision of up to 1 g • Integrated irrigation units allow precise and reproducible water application for drought stress or related experiments that require an accurate control of water volume to 1 ml http://www.plantphenomics.org.au/services/droughtspotter/ Crop Plant Root Module (The University of Adelaide, Australia) • Comprises destructive and non-destructive measurement of root growth, architecture, morphology, and water uptake in soil in controlled environments and in the field http://www.plantphenomics.org.au/services/croproot/ Jülich Plant Phenotyping Centre (Jülich, Germany) • Reproducibly quantifies growth and architecture of roots • Elucidates dynamic establishment of roots in space and time • Determines interaction of root responses with above ground plant part http://www.fz-juelich.de/ibg/ibg-2/EN/organisation/JPPC/ JPPC_node.html PHENOPSIS (INRA, France) • Automated platform for reproducible phenotyping of plant responses to soil water deficit Granier et al. (2006), Bresson et al. (2015) WIWAM (SMO and VIB, Belgium) • Automated phenotyping platform for automated weighing, watering, and imaging of plants and, therefore, strictly controlling the applied watering regime Skirycz et al. (2011), Clauw et al. (2015) PHENOSCOPE (INRA, France) • Automated large-scale phenotyping platform that automatically adjusts watering and is equipped with a zenithal imaging system to monitor rosette size and expansion rate during the vegetative stage, with automatic image analysis allowing manual correction Tisné et al. (2013) LemnaTec (LemnaTec, Germany) • Versatile phenotyping system and analytics platform for diverse temporal responses to water availability Chen et al. (2014), Neumann et al. (2015) GlyPh (self-construction) • Simple and low-cost automatic platform for high throughput measurement of plant water use and growth to assess drought tolerance Pereyra-Irujo et al. (2012) Phenotyping platform (and institute or company) Salient features Reference LeasyScan phenotyping platform (ICRISAT, India) • A novel 3D scanning technique to capture leaf area development continuously • Scanner-to-plant concept to increase imaging throughput and analytical scales to combine gravimetric transpiration measurements. • Combines 3D imaging and lysimetry for high- throughput phenotyping of traits controlling plant water budget Vadez et al. (2015) Semi-hydroponic phenotyping system (The University of Western Australia, Australia) • Permits mapping and digital measurement of dynamic growth of taproot and lateral roots • Desirable tool for examining root architecture of deep root systems and large sets of plants in a relatively small space Chen et al. (2011) DEEPER: an integrated phenotyping platform for deeper roots (PennState Colleage of Agriculture, USA) • A platform for identifying the traits of deeper rooted crops in non-destructive field phenotyping of rooting depth, root modeling, high-throughput 3D imaging of root architecture, and anatomy http://plantscience.psu.edu/research/labs/roots/projects/ deeper-an-integrated-phenotyping-platform-for-deeper-roots GLO-Roots: luminescence-based imaging system (Dinnenylab, USA) • Combines custom-made growth vessels and new image analysis algorithms to non-destructively monitor RSA development over space (2D) and time • Allows information on soil properties (e.g. moisture) to be integrated with root growth data • Makes use of luminescence imaging of roots expressing plant codon-optimize luciferase Rellán-Álvarez et al. (2015) X-Ray computed tomography (University of Nottingham, UK) • Non-destructively visualizes opaque root structures by measuring the attenuation of ionizing radiation as it passes through the root • A series of projections are acquired and combined to reconstruct a 3D image of the root system Mairhofer et al. (2012), Mooney et al. (2012) Rhizophonics (University of Liège, Belgium) • Combines hydroponics and rhizotrons • System is made of a nylon fabric supported by an aluminum frame • The set-up is immersed in a tank filled with liquid medium • Allows non-destructive, 2D imaging of root architecture while simultaneously sampling shoots Mathieu et al. (2015) Clear pot method (The University of Queensland, Australia) • Uses transparent pots filled with soil or other p otting media • Seeds are planted close to the pot wall to enable high- throughput imaging of roots along the clear pot wall • To prevent light exposure, the clear pot is placed in black pots while roots are developing Richard et al. (2015) Rhizoslides (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • The set-up consists of a Plexiglas sheet covered with moistened germination paper. Seeds are planted on the slit of the Plexiglas • Allows separation of crown roots from embryonic roots Le Marié et al. (2014) Shovelomics (PennState Colleage of Agriculture, USA) • Involves manual excavation of plants and separating roots from shoots • Washed roots are then placed on a phenotyping board for root trait quantification • New algorithms allow extraction of several root traits in a high throughput manner Trachsel et al. (2011), Bucksch et al. (2014) Soil coring (CSIRO and The University of Queensland, Australia) • Uses a tractor-mounted, hydraulic soil corer to drive steel alloy sampling tubes into the soil • When combined with novel planting configurations (e.g. hill plots), this method allows for phenotyping deep rooted crop varieties Wasson et al. (2014) Rhizo-lysimetry (E. H. Graham Centre for Agricultural Innovation, Australia) • Elaborate facility consisting of an underground corridor and concrete silos and pipes to house soil-containing soil cores for direct root observation Eberbach et al. (2013) Minirhizotrons • A transparent observation tube permanently inserted in the soil. Images of roots growing along the minirhizotron wall at particular locations in the soil profile can be captured over time Iversen et al. (2011), Maeght et al. (2013) RhizoTube (INRA, France) • Cylindrical rhizotrons that allow full 2D visualization of the root system of a single or up to six plants simultaneously • The RhizoCab is designed to take images of the entire root systems of plants growing in RhizoTubes, and also permits a focus on some parts of the root systems Jeudy et al. (2016) RADIX (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • Rhizoslide platform allowing high throughput digital image analysis of root system expansion Le Marié et al. (2016) Scanner bank (The James Hutton Institute, UK) • Low-cost, high-resolution root phenotyping platform, requiring no sophisticated equipment and adaptable to most laboratory and glasshouse environments, and applied to quantify environmental and temporal variation in root traits Adu et al. (2014) The Plant Accelerator® (The University of Adelaide, Australia) • The facility offers modern plant growth environments and state-of-the-art high-throughput automated imaging and computing technologies to monitor the performance of plants under different environmental conditions (e.g. which genotype performs best under drought stress) • Research projects facilitated by this technology vary from large scale screening of early growth, salinity tolerance to water and nutrient use efficiency http://www.plantphenomics.org.au/services/accelerator/ DroughtSpotter (The University of Adelaide, Australia) • A gravimetric platform with precision irrigation to assess transpiration dynamics of plants with a precision of up to 1 g • Integrated irrigation units allow precise and reproducible water application for drought stress or related experiments that require an accurate control of water volume to 1 ml http://www.plantphenomics.org.au/services/droughtspotter/ Crop Plant Root Module (The University of Adelaide, Australia) • Comprises destructive and non-destructive measurement of root growth, architecture, morphology, and water uptake in soil in controlled environments and in the field http://www.plantphenomics.org.au/services/croproot/ Jülich Plant Phenotyping Centre (Jülich, Germany) • Reproducibly quantifies growth and architecture of roots • Elucidates dynamic establishment of roots in space and time • Determines interaction of root responses with above ground plant part http://www.fz-juelich.de/ibg/ibg-2/EN/organisation/JPPC/ JPPC_node.html PHENOPSIS (INRA, France) • Automated platform for reproducible phenotyping of plant responses to soil water deficit Granier et al. (2006), Bresson et al. (2015) WIWAM (SMO and VIB, Belgium) • Automated phenotyping platform for automated weighing, watering, and imaging of plants and, therefore, strictly controlling the applied watering regime Skirycz et al. (2011), Clauw et al. (2015) PHENOSCOPE (INRA, France) • Automated large-scale phenotyping platform that automatically adjusts watering and is equipped with a zenithal imaging system to monitor rosette size and expansion rate during the vegetative stage, with automatic image analysis allowing manual correction Tisné et al. (2013) LemnaTec (LemnaTec, Germany) • Versatile phenotyping system and analytics platform for diverse temporal responses to water availability Chen et al. (2014), Neumann et al. (2015) GlyPh (self-construction) • Simple and low-cost automatic platform for high throughput measurement of plant water use and growth to assess drought tolerance Pereyra-Irujo et al. (2012) View Large Table 2. A list of key high-throughput phenotyping platforms Phenotyping platform (and institute or company) Salient features Reference LeasyScan phenotyping platform (ICRISAT, India) • A novel 3D scanning technique to capture leaf area development continuously • Scanner-to-plant concept to increase imaging throughput and analytical scales to combine gravimetric transpiration measurements. • Combines 3D imaging and lysimetry for high- throughput phenotyping of traits controlling plant water budget Vadez et al. (2015) Semi-hydroponic phenotyping system (The University of Western Australia, Australia) • Permits mapping and digital measurement of dynamic growth of taproot and lateral roots • Desirable tool for examining root architecture of deep root systems and large sets of plants in a relatively small space Chen et al. (2011) DEEPER: an integrated phenotyping platform for deeper roots (PennState Colleage of Agriculture, USA) • A platform for identifying the traits of deeper rooted crops in non-destructive field phenotyping of rooting depth, root modeling, high-throughput 3D imaging of root architecture, and anatomy http://plantscience.psu.edu/research/labs/roots/projects/ deeper-an-integrated-phenotyping-platform-for-deeper-roots GLO-Roots: luminescence-based imaging system (Dinnenylab, USA) • Combines custom-made growth vessels and new image analysis algorithms to non-destructively monitor RSA development over space (2D) and time • Allows information on soil properties (e.g. moisture) to be integrated with root growth data • Makes use of luminescence imaging of roots expressing plant codon-optimize luciferase Rellán-Álvarez et al. (2015) X-Ray computed tomography (University of Nottingham, UK) • Non-destructively visualizes opaque root structures by measuring the attenuation of ionizing radiation as it passes through the root • A series of projections are acquired and combined to reconstruct a 3D image of the root system Mairhofer et al. (2012), Mooney et al. (2012) Rhizophonics (University of Liège, Belgium) • Combines hydroponics and rhizotrons • System is made of a nylon fabric supported by an aluminum frame • The set-up is immersed in a tank filled with liquid medium • Allows non-destructive, 2D imaging of root architecture while simultaneously sampling shoots Mathieu et al. (2015) Clear pot method (The University of Queensland, Australia) • Uses transparent pots filled with soil or other p otting media • Seeds are planted close to the pot wall to enable high- throughput imaging of roots along the clear pot wall • To prevent light exposure, the clear pot is placed in black pots while roots are developing Richard et al. (2015) Rhizoslides (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • The set-up consists of a Plexiglas sheet covered with moistened germination paper. Seeds are planted on the slit of the Plexiglas • Allows separation of crown roots from embryonic roots Le Marié et al. (2014) Shovelomics (PennState Colleage of Agriculture, USA) • Involves manual excavation of plants and separating roots from shoots • Washed roots are then placed on a phenotyping board for root trait quantification • New algorithms allow extraction of several root traits in a high throughput manner Trachsel et al. (2011), Bucksch et al. (2014) Soil coring (CSIRO and The University of Queensland, Australia) • Uses a tractor-mounted, hydraulic soil corer to drive steel alloy sampling tubes into the soil • When combined with novel planting configurations (e.g. hill plots), this method allows for phenotyping deep rooted crop varieties Wasson et al. (2014) Rhizo-lysimetry (E. H. Graham Centre for Agricultural Innovation, Australia) • Elaborate facility consisting of an underground corridor and concrete silos and pipes to house soil-containing soil cores for direct root observation Eberbach et al. (2013) Minirhizotrons • A transparent observation tube permanently inserted in the soil. Images of roots growing along the minirhizotron wall at particular locations in the soil profile can be captured over time Iversen et al. (2011), Maeght et al. (2013) RhizoTube (INRA, France) • Cylindrical rhizotrons that allow full 2D visualization of the root system of a single or up to six plants simultaneously • The RhizoCab is designed to take images of the entire root systems of plants growing in RhizoTubes, and also permits a focus on some parts of the root systems Jeudy et al. (2016) RADIX (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • Rhizoslide platform allowing high throughput digital image analysis of root system expansion Le Marié et al. (2016) Scanner bank (The James Hutton Institute, UK) • Low-cost, high-resolution root phenotyping platform, requiring no sophisticated equipment and adaptable to most laboratory and glasshouse environments, and applied to quantify environmental and temporal variation in root traits Adu et al. (2014) The Plant Accelerator® (The University of Adelaide, Australia) • The facility offers modern plant growth environments and state-of-the-art high-throughput automated imaging and computing technologies to monitor the performance of plants under different environmental conditions (e.g. which genotype performs best under drought stress) • Research projects facilitated by this technology vary from large scale screening of early growth, salinity tolerance to water and nutrient use efficiency http://www.plantphenomics.org.au/services/accelerator/ DroughtSpotter (The University of Adelaide, Australia) • A gravimetric platform with precision irrigation to assess transpiration dynamics of plants with a precision of up to 1 g • Integrated irrigation units allow precise and reproducible water application for drought stress or related experiments that require an accurate control of water volume to 1 ml http://www.plantphenomics.org.au/services/droughtspotter/ Crop Plant Root Module (The University of Adelaide, Australia) • Comprises destructive and non-destructive measurement of root growth, architecture, morphology, and water uptake in soil in controlled environments and in the field http://www.plantphenomics.org.au/services/croproot/ Jülich Plant Phenotyping Centre (Jülich, Germany) • Reproducibly quantifies growth and architecture of roots • Elucidates dynamic establishment of roots in space and time • Determines interaction of root responses with above ground plant part http://www.fz-juelich.de/ibg/ibg-2/EN/organisation/JPPC/ JPPC_node.html PHENOPSIS (INRA, France) • Automated platform for reproducible phenotyping of plant responses to soil water deficit Granier et al. (2006), Bresson et al. (2015) WIWAM (SMO and VIB, Belgium) • Automated phenotyping platform for automated weighing, watering, and imaging of plants and, therefore, strictly controlling the applied watering regime Skirycz et al. (2011), Clauw et al. (2015) PHENOSCOPE (INRA, France) • Automated large-scale phenotyping platform that automatically adjusts watering and is equipped with a zenithal imaging system to monitor rosette size and expansion rate during the vegetative stage, with automatic image analysis allowing manual correction Tisné et al. (2013) LemnaTec (LemnaTec, Germany) • Versatile phenotyping system and analytics platform for diverse temporal responses to water availability Chen et al. (2014), Neumann et al. (2015) GlyPh (self-construction) • Simple and low-cost automatic platform for high throughput measurement of plant water use and growth to assess drought tolerance Pereyra-Irujo et al. (2012) Phenotyping platform (and institute or company) Salient features Reference LeasyScan phenotyping platform (ICRISAT, India) • A novel 3D scanning technique to capture leaf area development continuously • Scanner-to-plant concept to increase imaging throughput and analytical scales to combine gravimetric transpiration measurements. • Combines 3D imaging and lysimetry for high- throughput phenotyping of traits controlling plant water budget Vadez et al. (2015) Semi-hydroponic phenotyping system (The University of Western Australia, Australia) • Permits mapping and digital measurement of dynamic growth of taproot and lateral roots • Desirable tool for examining root architecture of deep root systems and large sets of plants in a relatively small space Chen et al. (2011) DEEPER: an integrated phenotyping platform for deeper roots (PennState Colleage of Agriculture, USA) • A platform for identifying the traits of deeper rooted crops in non-destructive field phenotyping of rooting depth, root modeling, high-throughput 3D imaging of root architecture, and anatomy http://plantscience.psu.edu/research/labs/roots/projects/ deeper-an-integrated-phenotyping-platform-for-deeper-roots GLO-Roots: luminescence-based imaging system (Dinnenylab, USA) • Combines custom-made growth vessels and new image analysis algorithms to non-destructively monitor RSA development over space (2D) and time • Allows information on soil properties (e.g. moisture) to be integrated with root growth data • Makes use of luminescence imaging of roots expressing plant codon-optimize luciferase Rellán-Álvarez et al. (2015) X-Ray computed tomography (University of Nottingham, UK) • Non-destructively visualizes opaque root structures by measuring the attenuation of ionizing radiation as it passes through the root • A series of projections are acquired and combined to reconstruct a 3D image of the root system Mairhofer et al. (2012), Mooney et al. (2012) Rhizophonics (University of Liège, Belgium) • Combines hydroponics and rhizotrons • System is made of a nylon fabric supported by an aluminum frame • The set-up is immersed in a tank filled with liquid medium • Allows non-destructive, 2D imaging of root architecture while simultaneously sampling shoots Mathieu et al. (2015) Clear pot method (The University of Queensland, Australia) • Uses transparent pots filled with soil or other p otting media • Seeds are planted close to the pot wall to enable high- throughput imaging of roots along the clear pot wall • To prevent light exposure, the clear pot is placed in black pots while roots are developing Richard et al. (2015) Rhizoslides (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • The set-up consists of a Plexiglas sheet covered with moistened germination paper. Seeds are planted on the slit of the Plexiglas • Allows separation of crown roots from embryonic roots Le Marié et al. (2014) Shovelomics (PennState Colleage of Agriculture, USA) • Involves manual excavation of plants and separating roots from shoots • Washed roots are then placed on a phenotyping board for root trait quantification • New algorithms allow extraction of several root traits in a high throughput manner Trachsel et al. (2011), Bucksch et al. (2014) Soil coring (CSIRO and The University of Queensland, Australia) • Uses a tractor-mounted, hydraulic soil corer to drive steel alloy sampling tubes into the soil • When combined with novel planting configurations (e.g. hill plots), this method allows for phenotyping deep rooted crop varieties Wasson et al. (2014) Rhizo-lysimetry (E. H. Graham Centre for Agricultural Innovation, Australia) • Elaborate facility consisting of an underground corridor and concrete silos and pipes to house soil-containing soil cores for direct root observation Eberbach et al. (2013) Minirhizotrons • A transparent observation tube permanently inserted in the soil. Images of roots growing along the minirhizotron wall at particular locations in the soil profile can be captured over time Iversen et al. (2011), Maeght et al. (2013) RhizoTube (INRA, France) • Cylindrical rhizotrons that allow full 2D visualization of the root system of a single or up to six plants simultaneously • The RhizoCab is designed to take images of the entire root systems of plants growing in RhizoTubes, and also permits a focus on some parts of the root systems Jeudy et al. (2016) RADIX (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • Rhizoslide platform allowing high throughput digital image analysis of root system expansion Le Marié et al. (2016) Scanner bank (The James Hutton Institute, UK) • Low-cost, high-resolution root phenotyping platform, requiring no sophisticated equipment and adaptable to most laboratory and glasshouse environments, and applied to quantify environmental and temporal variation in root traits Adu et al. (2014) The Plant Accelerator® (The University of Adelaide, Australia) • The facility offers modern plant growth environments and state-of-the-art high-throughput automated imaging and computing technologies to monitor the performance of plants under different environmental conditions (e.g. which genotype performs best under drought stress) • Research projects facilitated by this technology vary from large scale screening of early growth, salinity tolerance to water and nutrient use efficiency http://www.plantphenomics.org.au/services/accelerator/ DroughtSpotter (The University of Adelaide, Australia) • A gravimetric platform with precision irrigation to assess transpiration dynamics of plants with a precision of up to 1 g • Integrated irrigation units allow precise and reproducible water application for drought stress or related experiments that require an accurate control of water volume to 1 ml http://www.plantphenomics.org.au/services/droughtspotter/ Crop Plant Root Module (The University of Adelaide, Australia) • Comprises destructive and non-destructive measurement of root growth, architecture, morphology, and water uptake in soil in controlled environments and in the field http://www.plantphenomics.org.au/services/croproot/ Jülich Plant Phenotyping Centre (Jülich, Germany) • Reproducibly quantifies growth and architecture of roots • Elucidates dynamic establishment of roots in space and time • Determines interaction of root responses with above ground plant part http://www.fz-juelich.de/ibg/ibg-2/EN/organisation/JPPC/ JPPC_node.html PHENOPSIS (INRA, France) • Automated platform for reproducible phenotyping of plant responses to soil water deficit Granier et al. (2006), Bresson et al. (2015) WIWAM (SMO and VIB, Belgium) • Automated phenotyping platform for automated weighing, watering, and imaging of plants and, therefore, strictly controlling the applied watering regime Skirycz et al. (2011), Clauw et al. (2015) PHENOSCOPE (INRA, France) • Automated large-scale phenotyping platform that automatically adjusts watering and is equipped with a zenithal imaging system to monitor rosette size and expansion rate during the vegetative stage, with automatic image analysis allowing manual correction Tisné et al. (2013) LemnaTec (LemnaTec, Germany) • Versatile phenotyping system and analytics platform for diverse temporal responses to water availability Chen et al. (2014), Neumann et al. (2015) GlyPh (self-construction) • Simple and low-cost automatic platform for high throughput measurement of plant water use and growth to assess drought tolerance Pereyra-Irujo et al. (2012) View Large Robotic-assisted imaging platforms and computer vision-assisted analysis tools have also been developed for precise phenotyping of physiological growth, development, and other phenotypic properties (Fahlgren et al., 2015a,b). The data processing software that handles high throughput phenotyping for 2D, 3D, and anatomy cross sections of root have been extensively reviewed by Kuijken et al. (2015). The availability of automated high throughput phenotyping platforms enables testing of progenies from large germplasm populations across a range of target environments in terms of their impact on yield. The efficiency of these platforms, in this context, depends on the heritability and correlation of the traits with agronomically relevant breeding targets. Proper identification of the key constraints for a particular production ecology and an understanding of the biological processes can now be achieved with increasing scale and speed using existing platforms and designing new ones. Multiple challenges lie ahead for high throughput phenotyping such as (i) targeting the development of phenotyping protocols and technology specifically for the target traits of their breeding product profiles; (ii) development of precise protocols amenable for high throughput phenotyping for agronomically important traits; (iii) developing appropriate technology for collecting, collating, and analysing data; (iv) storage and processing of huge datasets collected during high throughput phenotyping; and (v) integration and deployment of high throughput phenotyping in breeding pipelines. On top of that, the cost per sample, especially when dealing with large-scale populations, still remains an important challenge to the phenotyping community. One important aspect of better integration of genomics compared with crop physiology in breeding is the lower cost of genotyping per sample compared with field-based phenotyping. Another area that needs immediate attention is the automation of data collection, analysis and delivery of phenotypic data to support the breeding community in a cost-effective and timely manner. Addressing these challenges will accelerate the integration of modern phenotyping tools into breeding programs and accelerate the development of crop varieties with wider adaptation, resilience, and increased productivity and profitability. Crop growth simulation and modelling approaches In recent years, crop simulation and modelling approaches have challenged the idea of broad adaptation cultivars and have brought a new paradigm in breeding for target environments. Crop models, with general genetic inputs, suggest how a given combination of alleles confers a positive or negative effect on plant performance in different locations and seasons (Hammer et al., 2002; Tardieu and Tuberosa, 2010; Messina et al., 2011). Modelling enables us to predict the effect of a change in the biological architecture of a plant type (for instance having faster growing roots, Vadez et al., 2012; for water use efficiency and stay green, Kholová et al., 2014), or of a change in the agronomic management (for instance increasing planting density, Vadez et al., 2017). As such, modelling should become an important resource-saving tool to guide the choice of breeding and agronomic management investment. Crop models have simulated the effect of crop management and climate on yield (Duncan et al., 1967; de Wit et al., 1970; de Wit, 1978; Kumar et al., 2009; Kim et al., 2010). Later, crop models were redesigned into farming systems models that could consider more broadly the effects of the management system including the carryover effects of rotations, planting density and intercropping. More broadly, systems models such as the Agricultural Production Systems sIMulator (APSIM) and the Decision Support System for Agrotechnology Transfer (DSSAT https://dssat.net/) simulate agricultural production systems (Keating et al., 2003; Jones et al., 2003; Holzworth et al., 2014; Hoogenboom et al., 2017), modelling interactions of plants, animals, climate, soil, and land management. Teixeira et al. (2015), using APSIM, demonstrated the importance of rotations for simulating climate impact assessments and Sennhenn et al. (2017) found new niches for short season legumes in Kenya. APSIM has been used in simulation studies for yield gap assessment in legumes such as soybean, groundnut, pigeonpea, and chickpea in India (Bhatia et al., 2007; Chauhan et al., 2008), in simulation of soil temperature in the podding zone of groundnut (Chauhan et al., 2007), and in assessment of the impacts of fertilizers and legumes on N2O and CO2 emissions from soils in subtropical agricultural systems (Huth et al., 2010). With the availability of high throughput phenotyping platforms, the use of crop models in a genetic context has become possible. A conceptual framework was developed by Tardieu and Tuberosa (2010) with four modules that take into consideration phenotyping, genetics, climate, and crop models for estimating the effects of genetic diversity on crop performance. More recently Parent and Tardieu (2014) compared the algorithms involved in 19 crop models developed for predicting the effects of climate or cultivation techniques on a reference genotype. This study reported significant differences in combined effects of temperature and water deficit on plant development, but these differences have a low impact on the yield prediction of a reference genotype because errors in the effects of different traits compensate each other. The impact of climate on crop yield is assessed using two common approaches: (i) process-based simulation models, which attempt to represent key dynamic processes affecting crop yields, and (ii) statistical models, which estimate functional relationships between historical observations of weather and yields (Lobell and Asseng, 2017). The first approach is closer to physiological processes and can explicitly take into account the genetic variability of traits as measured in phenotyping platforms, for both main effects and genotype×environment interaction. However, it is often over-parametrized and can lead to inaccurate predictions in some cases because the model follows its own logic even if not applicable to the considered set of conditions (Tardieu et al., 2017). The statistical approach is more conservative and safer, because predicted yields seldom depart from those observed experimentally. Both approaches are therefore complementary. The impact of drought on legume production is still not well understood in how it varies with legume species, region, agroecosystem, soil texture, and drought timing. For legumes specifically, a model has been developed (Simple Simulation Modelling; SSM) that uses the same model architecture across a range of grain legumes and is easy to use (Soltani and Sinclair, 2012). SSM has been used successfully to predict growth and yield in chickpea (Soltani and Sinclair, 2011; Vadez et al., 2012, 2013b), lentil (Ghanem et al., 2015a,b), common bean (Marrou et al., 2014), soybean (Sinclair et al., 2014), and groundnut (Vadez et al., 2017). In the case of soybean, a recent modelling study indicated that a faster rate of root growth has a negative impact on yield as available soil moisture was depleted faster by more vigorous root growth (Sinclair et al., 2010). While significant progress is being made in crop growth simulation and modelling, their integration with genetics and breeding is still in its infancy. Genomics-assisted breeding Most legume breeding programs focus on yield under drought stress. However, the pace of progress has been slow and the rate of yield gains has been minimal due to the complex nature of drought and non-availability of precise drought screening methods. As a result, the genetic gain achieved is about 1% per year in several species (Duvick, 2005; Cooper et al., 2009; Brisson et al., 2010; Lopes et al., 2012; Aisawi et al., 2015). The most promising option to design more drought resilient and sustainable production is to target the major traits of adaptation that include early flowering and seed set before the onset of terminal drought (Sennhenn et al., 2017). During the past two decades, the availability of molecular markers such as simple sequence repeats (Gupta and Varshney, 2000) and SNPs (Varshney et al., 2010) facilitated dissecting complex traits that hamper crop production, using QTL mapping and genome wide association mapping approaches (Varshney et al., 2015). Genome wide SNPs based on resequencing of several germplasm lines were also used to identify marker-trait associations in the cases of legumes such as pigeonpea (Varshney et al., 2017) and soybean (Zhou et al., 2015a). Genomics-assisted breeding (GAB; Varshney et al., 2005) was proposed to integrate genomics in breeding and it has been quite successful for several traits in cereals (Septiningsih et al., 2013; Varshney et al., 2006) and legumes (Varshney, 2016; Pratap et al., 2017). Recently, Reynolds and Langridge (2016) have suggested crossing parents with different complex but complementary traits to achieve cumulative gene action for yield, while selecting progeny using remote sensing in combination with genomic selection. Genetic and genomics approaches and innovative tools have been made available during the past decade to breed climate-resilient legumes; for instance, root xylem plasticity and its role in improving water use efficiency in soybean (Prince et al., 2017). The QTLs/genomic regions and markers associated with traits are used to breed for stress resilience using marker-assisted selection (MAS), marker-assisted back crossing (MABC), and MARS approaches (Varshney et al., 2012b). The legume community has been successful in developing several molecular breeding products despite the late arrival of genomic resources and trait-associated markers (Varshney et al., 2013a,b; Pandey et al., 2016; Varshney, 2016). Some key examples include resistance to Fusarium wilt and ascochyta blight (Varshney et al., 2013b) and improved drought tolerance (Varshney et al., 2013a) in chickpea; resistance to nematode and high oleic acid (Chu et al., 2011), resistance to leaf rust (Varshney et al., 2014), and resistance to high oleic acid (Janila et al., 2016) in groundnut; resistance to rust disease (Khanh et al., 2013), soybean mosaic virus (Saghai-Maroof et al., 2008; Shi et al., 2009; Parhe et al., 2017), and low phytate (Landau-Ellis and Pantalone, 2009) in soybean; Striga resistance and seed size in cowpea (Lucas et al., 2015; see Boukar et al., 2016); pyramid genes for resistance to ascochyta blight and anthracnose in lentil (Taran et al., 2003); powdery mildew resistance (Ghafoor and McPhee 2012), lodging resistance (Zhang et al., 2006), frost tolerance (see Tayeh et al., 2015b), and Aphanomyces root rot resistance (Lavaud et al., 2015) in pea; and resistance to common bacterial blight disease (Miklas et al., 2000, 2006; Mutlu et al., 2005; O’Boyle and Kelly, 2007), rust and viruses (Stavely, 2000), rust, anthracnose, and angular leaf spot (Oliveira et al., 2008), rust (Feleiro et al., 2001), and anthracnose (Alzate-Marin et al., 1999) in common bean. Several of these improved lines have either been released or are in the release pipeline in different countries. In addition to the above, the gene/QTL pyramiding efforts are at an advanced stage for several legume traits such as in groundnut (leaf rust resistance+late leaf spot+high oleic acid), chickpea (fusarium wilt resistance+ascochyta blight resistance+drought tolerance; see Pandey et al., 2016), and cowpea (resistance to Striga+aphid+macrophomina root rot; see Boukar et al., 2016). The majority of the legume crops now enjoy the availability of genomic resources and such examples are seen more frequently in grain legumes. The MARS approach was deployed in cowpea successfully for improving grain yield, drought tolerance, Striga resistance and Macrophomina resistance (see Boukar et al., 2016), but this approach did not work in the case of chickpea (see Varshney et al., 2013c). Next generation GAB approaches While some examples, as mentioned above, are available for the use of molecular breeding, they include introgression of one to a few traits with higher heritability through MABC. Due to advances in low-cost genotyping/sequencing and availability of diagnostic markers, we expect use of early generation selection, genomic selection, and genome editing technologies in future GAB approaches. Early generation selection Early generation selection (EGS) is performed in highly segregating populations (mostly F2) while advanced generation selection (AGS) begins with F4 onwards followed by assessment of yield performance and other important agronomically superior traits in subsequent generations. Both EGS and AGS can be used to breed improved lines followed by evaluating these lines in target populations of environment (i.e. the set of conditions under which the set of considered varieties has been bred) to select promising lines with wider adaptation and higher yield under water limited environments. The success of EGS depends on (i) efficiency and accuracy of phenotyping a selected trait in a segregating population of single plants, for instance phenotyping for leaf wilting and selection for root depth in tubes, and (ii) genetic makeup of the population; for instance, drought tolerance has been introgressed from the wild species resulting in the appearance of undesirable and agronomically segregating populations. EGS will be highly desirable only when F2s are from agronomically superior elite lines. In the case of soybean, EGS for pods per plant proved to be the most effective for enhancing the yield (Singh and Sharma, 2016). AGS holds considerable promise for traits that are best expressed in homozygous progeny. While EGS and AGS have been earlier practised based on phenotypic selection, diagnostic markers are now being integrated for traits, for both EGS and AGS. A high-throughput genotyping project (HTPG; http://cegsb.icrisat.org/high-throughput-genotyping-project-htpg/), an initiative led by ICRISAT in collaboration with the International Maize and Wheat Improvement Center (CIMMYT) and the International Rice Research Institute (IRRI) and funded by the Bill & Melinda Gates Foundation, is expected to accelerate the deployment of EGS/AGS. It is important to note that the HTPG project has made it possible to screen populations at a 2017 cost of US$ 1.5 per sample for 10 SNP markers including DNA extraction. While several institutes such as CIMMYT, ICRISAT and IRRI have started to deploy EGS at a large scale in maize, rice, wheat, and groundnut, the genomics community needs to work on developing additional diagnostic markers for more traits especially nutrition traits across all legume crops. Genomic selection Genomic selection (GS), proposed by Meuwissen et al. (2001), is a promising breeding approach for simultaneous improvement of complex traits, especially those with low heritability. GS uses genomic signatures or dense marker profiles of individuals as well as parents for predicting the breeding values. The use of GS for accelerating genetic gains, based on genomic-estimated breeding values predicted using genome-wide dense markers, has been deployed extensively in dairy cattle (Van Raden et al., 2009). García-Ruiz et al. (2016) reported a positive impact of GS on cattle breeding through drastic reduction of generation interval and increased selection intensity for low heritable traits. Initially, computer simulations and parametric and non-parametric statistical models were used on maize and wheat datasets to study the prediction accuracies in real plant breeding scenarios (Bernardo and Yu, 2007; de los Campos et al., 2009; Crossa et al., 2010, 2011). Promising results in major cereals provided a kick start for deploying GS in a few legume crops, such as pea, soybean, chickpea, groundnut, and pigeonpea (Varshney, 2016). GS in crop plants including legumes has been extensively reviewed recently elsewhere (Crossa et al., 2017). In GS, in brief, a training population (genotyped with genome wide markers and extensively phenotyped over years and in several locations) is trained with appropriate models to estimate the genomic-estimated breeding values in a breeding population (genotyped with genome wide markers) and validated in a validation population. Hickey et al. (2017) compared both plant and animal breeding approaches to make a case for bringing the two together through the application of GS. In recent years, the GS approach has been deployed in different types of populations, for instance bi-parental, multi-parental (Beyene et al., 2015), and synthetic populations (Schopp et al., 2017) and diverse panel of inbred lines, which basically differ in the pedigree, number of founders used to develop the population, linkage disequilibrium decay, and population structure. The genetic advance achieved through selection depends on the total variation (σ2), the repeatability of the trait (h2) and the selection pressure (S) imposed. The selection pressure implies the proportion of the population selected. Even if the repeatability is high the genetic advance (genetic gain, R) would be small without a large genetic variation. With selection, the genetic variation and consequently the repeatability and thus the advance from one to the next generation decline while the mean value of the trait increases (or decreases, depending on the trait). Genotyping platforms, training populations and statistical models Although the availability of genome wide markers and cost of genotyping were two major factors that restricted implementation of GS in crop breeding, they are no longer constraints as most of the legume communities are now equipped with genome sequence information as well as high density SNP arrays. For instance, high density SNP arrays have become available in case of chickpea (Roorkiwal et al., 2017), groundnut (Pandey et al., 2017a), and pigeonpea (Saxena et al., 2017 and unpublished). In addition, the efficiency and accuracy with which superior lines can be predicted also depend on the size of the reference population (Jannink et al., 2010; Lorenz et al., 2011). The genetic relatedness or population structure (Saatchi et al., 2011; Riedelsheimer et al., 2013; Wray et al., 2013) may result in overestimating the heritability of the traits (Price et al., 2010; Visscher et al., 2012; Wray et al., 2013). The population structure of the training population can be determined with greater accuracy using genome wide SNPs compared with the simple sequence repeats and SNP arrays (Isidro et al., 2015). Accuracy of estimating breeding values is important in GS. For instance, Jarquín et al. (2014) obtained a high prediction accuracy of 0.64 by deploying GS for improving yield and agronomic traits using genotyping-by-sequencing in a breeding program. In addition, GS for yield ridge regression best linear unbiased prediction (BLUP) coupled with fivefold cross-validations and marker preselection based on haplotype blocks is an interesting option for a cost-efficient implementation of genomic selection for grain yield in soybean breeding (Ma et al., 2016). In the case of pea, using GS, Tayeh et al. (2015c) reported mean cross-environment prediction accuracies of 0.83 for thousand-seed weight, 0.68 for number of seeds per plant, and 0.65 for date of flowering. Population structure did not impact the prediction accuracy, and modelling genotype by environment by management (G×E×M) indicated improved prediction efficiency in chickpea (Roorkiwal et al., 2016). Further, Roorkiwal et al. (2016) used statistical models such as RR-BLUP, Kinship GAUSS, Bayes Cπ, Bayes B, Baysian LASSO, and random forest regression and reported high prediction accuracies for days to maturity, days to flowering, and seed dry weight. The development and deployment of improved and more precise statistical models will eventually enhance further the prediction accuracy leading to enhanced GS efficiency in legume crops. Development of models that use deep machine-learning methods and multi-trait and multi-environment information is essential to realize the potential of GS in legume breeding and to enhance the prediction accuracies (Crossa et al., 2017). Selection intensity and generation interval The selection differential (S) is defined as the difference of base population mean and the mean of the selected parents R=h2S where h2 is the narrow-sense heritability (Arruda et al., 2016). The selection intensity depends on how different the selected parents are from the overall population average. The closer the selected parents to the population average, the smaller the selection intensity and vice versa. The impact of GS and generation intervals in cattle breeding was first systematically studied by de Roos et al. (2010). Recently, García-Ruiz et al. (2016) reported that decreased generation interval and increased selection intensity for low heritable traits is as a result of the positive impact of GS on the US cattle industry. Genomic selection in combination with a reduced generation interval may double the rate of genetic gain while keeping the rate of inbreeding per generation constant. Rapid cycling The usefulness of rapid cycling genomic selection (RCGS) was first reported in a bi-parental mapping population derived from B73 and Mo17. Massman et al. (2013) reported that RCGS had a superior response for stover yield, as well as stover and grain yield indices that were 14–50% higher than those of MARS. Recently, in tropical maize, by employing RCGS on eight bi-parental populations evaluated under drought stress environments, Beyene et al. (2015) reported an average gain of 0.086 ton ha−1 per cycle, and hybrids derived from cycle 3 produced 7.3% (0.176 ton ha−1) higher grain yield than those from cycle 0 and conventional pedigree breeding methods. Further, Zhang et al. (2017) using RCGS in a tropical maize multi-parental population reported a genetic gain for grain yield from cycle 1 to cycle 4 reached 0.225 ton ha−1 per cycle, which is equivalent to 0.100 ton ha−1 year−1 over a 4.5 year breeding period from the initial cross to the last cycle. While similar rates of gain have not been reported in legumes, recent reports on over three generations per year in traditional breeding for pigeonpea (Saxena et al., 2017) offer new opportunities to apply RCGS to accelerate rates of genetic gain in grain legumes. Genome editing The sequencing-based trait dissection and gene discovery are enabling the functional characterization of genes. An alternative approach for modifying targeted genomic regions, genome editing, may be better suited for improving the deficiency of one or two traits within popular varieties that have suitable production and market traits. Enhancing the performance of market-preferred varieties offers the advantage of higher productivity and higher replacement rates by farmers (Dar et al., 2013). Genome editing harnesses the strength of programmable nucleases in cutting and pasting the specific genetic information in living cells (Kim, 2016). The availability of more and more information on functionality of genes or haplotypes for economically important traits will facilitate modification of multiple SNPs through genome editing without leading to a deterioration in the original behavior of a popular cultivar. In several of the legume crop species, with genetic barriers such as ploidy differences, the exchange of genetic material through natural process is restricted and therefore the huge genetic variation lying within crop’s wild relatives remains untapped and unutilized. In this context, if the function of the specific genome variations in wild relatives can be assigned, genome editing can enable targeted editing of a popular cultivar without fear of linkage drag. The recently developed targeted genome-editing technologies such as zinc finger nucleases, RNA-guided endonucleases, transcription-activator-like effector nucleases, and CRISPR (cluster of regularly interspaced palindromic repeats) have undergone tremendous improvement over the past couple of years (Wang et al., 2017). Of the several approaches, CRISPR received wider acceptance among researchers due to its huge potential role in crop improvement in coming years (Kim, 2016). This technology has already been optimized in three model legume crops, soybean (Sun et al., 2015; Li et al., 2015), Lotus japonicas (Wang et al., 2016) and Medicago truncatula (Meng et al., 2017). The efficiency of genome editing using various endogenous and exogenous RNA polymerase III promoters has been compared (Sun et al., 2015; Du et al., 2016) and in soybean the Cas9–single guide RNA together with donor DNA fragments was successfully transformed into soybean using the particle bombardment method (Li et al., 2015). Medicago and Lotus, having smaller diploid genomes, are the model legume crops for conducting research in several areas including symbiotic nitrogen fixation. The efficacy of the CRISPR system has been assessed in these two crops by modifying the multiple genes related to symbiotic nitrogen fixation (Michno et al., 2015; Wang et al., 2016; Meng et al., 2017). Recently, Benson Hill Biosystems, an agricultural technology company based in the UK, launched a commercial service for genome editing for crop improvement (https://www.prnewswire.com/news/benson-hill-biosystems). These developments will further accelerate the deployment of this promising technology in improving desired traits in legume crops. Genome editing is considered to be a safe technology for developing improved crop plants by US regulators, who indicated that genome editing products developed using oligonucleotide-directed mutagenesis or site-directed nucleases should not be subjected to regulation as these technologies are similar to mutagenesis (see Sprink et al., 2016). There are multiple examples of development of improved plant varieties using genome editing technology and 37 such examples in Arabidopsis, tobacco, maize, rice, wheat, canola, soybean, camelina, tomato, and cucumber have been well documented recently (see Kamburova et al., 2017). Of these successful efforts, several of these products have been approved in different countries for cultivation, such as herbicide-tolerant SU Canola (Cibus 5715) in Canada in 2014 (https://cen.acs.org/articles/95/i24/CRISPR-new-toolbox-better-crops.html), browning-resistant Arctic apples in the USA and Canada in 2015 (Smyth, 2017), and less-acrylamide-producing potatoes for chip processing in Canada in 2016 (Smyth, 2017), while tomato with a greater number of fruits and soybean with high oleic acid are near to reaching approval. Expected products in Canada and the USA in the coming years include glyphosate-tolerant flax in 2019. Products expected for 2020 include late-blight-resistant potato, herbicide-tolerant rice, and starchy or ‘waxy’ corn. Improving the operational efficiency in breeding programs Improvement in operational efficiency in the breeding programs will help in conducting precise and well-documented research experiments. Currently, the majority of breeding programs still record data on paper. The digitalization of breeding programs and improved operational efficiency will reduce the chance of error, which in turn will allow breeders to assess the potential of improved breeding procedures (Varshney et al., 2016). In this context, digitalization, automation, and mechanization in breeding procedures need to be adopted during the breeding process. More emphasis is required on uniform ontology, digitalized pedigree information, appropriate experimental design, barcoding of breeding material, digital data recording in the field, data management systems, and quality control. Integrated Breeding Platform (http://www.integratedbreeding.net) is one such initiative that is helping breeding programs to modernize and improving their breeding efficiency. For instance, the Breeding Management System of Integrated Breeding Platform is being used extensively at ICRISAT and its partner breeding programs. Similarly, B4R (http://bbi.irri.org/home) is a program for managing breeding data that has been developed and is being used by IRRI. The volume of data generated throughout the breeding process has increased significantly due to the involvement of high throughput genotyping and phenotyping technologies. Data volume now poses a significant challenge in compilation, processing, and interpretation. Available tools engaged today may no longer be suitable for handling high volume data in the future, thereby necessitating development of improved tools/software that can drive science-based breeding at a faster pace with user-friendly tools and databases. The Genomic Open-source Breeding Informatics Initiative is one such endeavor (http://gobiiproject.org/) targeting deployment of high density genotyping information in public sector breeding programs for more precise selection to achieve higher genetic gains. Further, the Excellence in Breeding (EiB; http://excellenceinbreeding.org/) platform came into existence very recently to bring all the components of CGIAR crop improvement under a single platform to focus on better integration and modernizing of breeding programs in developing countries. The above efforts will help in improving the efficiency and resilience of agricultural systems to achieve more sustainable food production for increased profitability of farming, greater resilience and nutrition of global agrifood systems. Seed systems Once superior varieties are developed, they should reach farmers’ fields as early as possible such that the average age of the varieties in the farmers’ fields are not more than 10 years. Cultivar replacement is indeed one of the basic undertakings in the EiB platform. In the past, however, most released varieties of legumes have not enjoyed high rates of adoption (Atlin et al., 2017). One reason could be low performance of new varieties under farmers’ field management and conditions. The second reason is limited multi-location testing of public sector varieties due to limited funds. As a result, seed companies do not know the level of superiority of a new variety for a certain production ecology and are reluctant to invest in producing and promoting a new seed product. Our experience from the Tropical Legumes project (http://tropicallegumes.icrisat.org/) shows that adoption of new varieties requires promotion just like with new products in the manufacturing sector. As compared with hybrid maize, where the private sector invests heavily in thousands of demonstration plots to create awareness and popularize new varieties, only very limited demonstration plots of new legume varieties and accompanying agronomic practices are conducted. Farmers’ awareness of improved varieties has been reported to be strongly correlated with higher adoption rates for pigeonpea in Tanzania (Amare et al., 2012) and for chickpea in Ethiopia (Abate et al., 2012). However, unlike the private manufacturing sector where a new product is promoted and new demand can be met quickly, seed is the vehicle through which genetic gains are realized in farmers’ fields, especially with small-scale farmers in the case of legumes. Production of seed takes time through cycles of breeding, foundation and certified seed, each stage taking at least one cropping season. Compared with cereals and canola, legumes generally have a lower seed multiplication rate, thereby requiring extra space, labor, time, effort, and more generations to produce sufficient quantities of certified seed. This increases the cost of seed production to seed producers, which must be paid for by farmers, many of whom find the costs prohibitive compared with non-authentic sources. In addition, virtually all legumes are self-pollinated (pigeonpea is an exception) and farmers usually recycle their own seeds. Therefore, the incentive to develop a business around seed production in most grain legumes is low. Some legumes have relatively large seed rate requirements, sometimes more than 100 kg ha−1, meaning farmers have to buy large quantities to plant a relatively smaller area compared with cereals. This has additional implications in terms of requirement of seed storage space and packaging material, further increasing the cost of seed. In addition, transportation of the seed to remote areas is both cumbersome and expensive due to the bulky nature of grain legumes. Legume seeds also suffer from loss of germination faster than cereals, especially in hot, humid environments. Seed producers and traders must meet these challenges by investing in proper storage, thereby increasing the willingness of farmers to buy seed regularly off-farm with an assurance of better quality than farm-save seed (Sperling and McGuire, 2010). Besides these challenges, legumes are highly self-pollinated and it is assumed that farmers save their own seed for several seasons and do not need to buy new seed each season, which makes it difficult for seed companies to predict demand for seed. It was, however, reported in 2016 that a significant fraction of legume farmers (64.4%) actually do buy ‘seed’ or rather grain from local markets (McGuire and Sperling, 2016). However, due to high cost, limited availability, and access, farmers do not buy legume seed from the formal seed sector (seed companies and agrodealers). Innovative approaches are therefore necessary to tackle the challenge of legume seed supply to ensure availability, accessibility, affordability, and sustainability for the varieties to replace older varieties. This requires an integrated approach to seed production and delivery systems that includes formal public–private sector partnerships and a variety of farmer-based seed production and supply initiatives (Siddique et al., 2012; Johansen and Siddique 2017). One of the approaches used to increase total seed production and availability, particularly of early generation seed, is through licensing of public varieties for decentralized production. An example of this is in seed production and distribution of improved varieties of common bean in Ethiopia. Well-established innovation platforms along the bean value chain contributed to sustaining a decentralized seed production system (Rubyogo et al., 2010). Seed certification, normally by public agencies, is an important quality control step but it also increases the cost of seed significantly. Kenya and Zambia are examples of countries where certification is licensed to private institutions and individuals. Quality declared seed (QDS) is an approach to producing seed by competent seed producers and suppliers that accommodates the diversity of farming systems, particularly in the more difficult areas where highly organized seed systems do not function well (Plant Production and Protection Division, 2006). QDS standards and monitoring procedures are slightly less demanding, allowing for less expensive seed and more decentralized producers. The quality and performance of QDS seed on farmers’ fields is comparable—but the seed is less expensive (Granqvist, 2006). Innovative approaches are also needed for increasing availability, supply and access to high quality seed of improved varieties (e.g. through small packs, financial support, better coordination with extension or development programs). A small seed pack approach was pioneered by Pan-African Bean Research Alliance two decades ago to encourage informal varietal diffusion and adoption of new varieties (Grisley, 1993) and gained popularity as an efficient and cost effective means of reaching more farmers with bean seed of improved varieties (Maereka and Rubyogo, 2015). It has been adopted and popularized in the Tropical Legumes II/III project (Monyo and Varshney 2016), involving national programs and the private sector (Rubyogo et al., 2016). Demand creation for quality seed includes working with off-takers like grain exporters, demonstration plots of better varieties, post-harvest handling, and market linkages. Good examples of the role of off-takers can be cited with the Raphael Group Ltd in southern Tanzania that has promoted the adoption of the ‘Uyole 03’ sugar bean variety and ACOS Ltd in Ethiopia that has promoted market-orientated production of various varieties of white pea bean and kabuli chickpea. Reaching farmers ‘at the last mile’ by bundling seed with other products, piggybacking on existing product supply channels such as fertilizers and pesticides, and strengthening community seed production approaches are all critically important areas for innovation. Agronomy packages Improved seeds alone do not guarantee improved yields. Soil fertility is often a big constraint for smallholder farmers in the semi-arid tropics (Kamanga et al., 2013). Deployment of improved varieties should be accompanied by integrated crop management practices, underscoring the prevailing cropping practices. In fact, consideration of agronomic practices should be considered early in the development of product profiles so that breeding populations are developed accordingly and selection pressure integrates prevailing agronomic practices (e.g. soil fertility, crop density). Combining organic and mineral fertilizer is a sound management principle for smallholder farmers when combined with techniques such as crop rotations to improve soil fertility and nutrient availability. Two examples clearly illustrate the integration of integrated soil fertility management principles: (i) dual purpose grain legume–maize rotations with P fertilizer targeted at the legume phase and N fertilizer targeted at the cereal phase in the moist savanna agro-ecozone (Sanginga et al., 2003), and (ii) micro-dose fertilizer applications in legume–sorghum or legume–millet rotations with retention of crop residues and water harvesting techniques in semi-arid agro-ecozones (Bationo et al., 1998). In trials in West Africa, control of early season insect pests and diseases increased crop vigor, and greater yields have been reported in legumes such as cowpea (Tekwa et al., 2010) and groundnut (Yakubu et al., 2011) with ApronStar, a seed treatment fungicide–insecticide mixture for controlling downy mildew, damping-off diseases, as well as for protection of seeds and seedlings against early season insect pests and soil-borne diseases. Therefore, there is a need in the future to breed cultivars in a system perspective, i.e. to breed crops considering elements of the agro-ecosystem in which these will be deployed. Understanding abiotic and biotic stresses is important in designing breeding programs. For example, planting groundnuts early in the season when aphid populations are still low coupled with maintaining a good plant density can reduce aphids that are the vector for rosette disease, which in turn reduces disease incidence (Waliyar et al., 2007). This in turn would ensure optimum yield and that varieties developed for other traits, such as oil quality, early maturity, drought tolerance, and foliar disease resistance, are not lost due to rosette disease. Sowing time is also particularly important in pigeonpea, where there are different maturity groups. Early maturing varieties can compete with the cereals in intercropping systems but would be the best suited as a sole crop in places prone to short rainfall periods. Medium maturing varieties are planted after the main cereals and remain in the field when the main cereal is harvested. Similarly, proper timing is critical in chickpea to ensure the crop does not suffer from terminal drought (if planted too late) or seedling fungal diseases if planted too early during the rainy period. Integrated and coordinated approach Legumes are nutrient-rich crops with high levels of protein and other important nutritional components that offer great support in fighting malnutrition across the world. Genetic improvement for legumes needs to accommodate traditional yield-improvement priorities alongside nutritional characteristics. A single technology or process cannot be singled out to achieve higher yields in farmers’ fields with high consistency (Siddique et al 2012; Johansen and Siddique, 2017). Participatory feedback from different stakeholders along commodity value chains is required to develop appropriate product profiles so farmers can be more productive and profitable by accessing markets with surplus production. In this context, a holistic approach is required to realize higher genetic gains in less time and with more precision in the legume crops. We have proposed this strategy in Fig. 1. Fig. 1. View largeDownload slide Strategy for strengthening the breeding process and adopting improved technologies for achieving higher genetic gains in farmers’ field. Low-cost sequencing and genotyping technology will eventually help in molecular characterization of entire diverse germplasms. High density genotyping data together with phenotyping data on entire germplasm will help filter the most useful germplasm lines for use in breeding programs. Low-cost sequencing and genotyping data will also facilitate dissection of multiple traits even in complex genetic populations and faster and precise gene discovery. Improved and modern approaches need to be integrated in trait mapping and trait improvement pipeline in addition to adoption of improved breeding operational efficiency. Higher genetic gains achieved with improvement of desired traits, key to major stakeholders, will eventually help not only in attaining higher productivity but also in increasing the profitability of farming. Fig. 1. View largeDownload slide Strategy for strengthening the breeding process and adopting improved technologies for achieving higher genetic gains in farmers’ field. Low-cost sequencing and genotyping technology will eventually help in molecular characterization of entire diverse germplasms. High density genotyping data together with phenotyping data on entire germplasm will help filter the most useful germplasm lines for use in breeding programs. Low-cost sequencing and genotyping data will also facilitate dissection of multiple traits even in complex genetic populations and faster and precise gene discovery. Improved and modern approaches need to be integrated in trait mapping and trait improvement pipeline in addition to adoption of improved breeding operational efficiency. Higher genetic gains achieved with improvement of desired traits, key to major stakeholders, will eventually help not only in attaining higher productivity but also in increasing the profitability of farming. Our crop improvement programs, which include not just breeding but all other disciplines, need to start with demand and feedback from the stakeholders including farmers, consumers, and markets. Based on these demands, crop improvement programs prioritize traits (including nutrition) and define a product profile of the varieties to be developed. Subsequently, genebank, pre-breeding and trait discovery groups together with breeders and biometricians formulate a list of suitable genotypes with desired traits, and high breeding values are identified after a comprehensive genome characterization using sequencing and genotyping platforms and phenotypic characterization in target environments. In parallel, we need to have diagnostic markers with higher prediction accuracy and candidate genes with causal effect for a given trait. For trait discovery, instead of using a traditional mapping approach, it is possible now to use sequencing-based trait mapping approaches such as QTL-Seq, Bulk-Seq and Indel-Seq (Singh et al., 2016, 2017; Pandey et al., 2017b). Similarly, recent advances in functional genomics such as gene expression atlases (Pazhamala et al., 2017) and use of near-isogenic lines for transcriptomics approaches can identify the candidate genes for the target traits. In order to increase selection intensity, it is essential to increase the number of crosses and the population of each cross in public sector breeding programs. In our opinion, each breeding program should make at least 100 crosses every year and each cross should have at least 500–1000 F2s using the most appropriate parental combinations. Of course, the size of the population would also need to be adjusted to the genetics of the target trait. Early generation screening with the diagnostic markers or candidate genes for must-have traits can dramatically reduce early generation populations as a ‘forward breeding’ process. After selecting a small numbers of lines, genomic prediction can be undertaken and selected lines with higher breeding values can be targeted for testing in target population of environments. Subsequently, the selected lines can go to a national program for further multi-location and multi-year testing for possible release of improved varieties or can be included in the pool of parental selections in the new crossing schemes. We believe that such an approach will accelerate the development of superior grain legume varieties. In parallel, it is also possible to continue to undertake MAS or MABC in the ruling varieties for introgressing major genes/QTLs that have become deficient during their long-term cultivation in farmers’ fields. While undertaking GAB approaches, it is also essential that we improve our operational efficiency with appropriate analytical and decision support tools. For instance, appropriate experimental design for testing a segregating population/germplasm collection is very important. Similarly, having phenotypic data for different breeding populations/germplasm collections in databases and using these databases for planning a new set of experiments is critical. Deployment of breeding databases such as BMS and B4R in breeding programs needs to be accelerated. Use of tablets and mobile-based ‘Field Book’ programs will reduce error and turnaround time for data-driven breeding decisions. While using the GAB strategies, EGS, GS, MABC or MAS, it is important that breeding programs use the appropriate decision support tools to select the superior lines instead of using spreadsheets or stand-alone programs. All these efforts and good breeding practices together offer operational efficiency in the breeding programs. While use of GAB approaches together with operational efficiency methods can help in developing better varieties faster, the full genetic potential of these varieties can only be realized once they have been adopted and grown in farmers’ fields. It is essential to work with government agencies together with informal seed system actors in different developing countries to accelerate varietal replacement. We need to start using ‘digital seed road maps’ (http://seedsystems.icrisat.org/) that can help different stakeholders of the seed value chain to assess the need and make different kinds of seeds (breeders’ seed, foundation seeds, certified/truthfully labeled seeds) available to farmers. Finally, digital agriculture tools, for example, sowing apps (http://www.icrisat.org/tag/ sowing-app/), plantix (http://www.icrisat.org/tag/plantix/), should be deployed to provide increased access to equitable markets by farmers to increase profitability. We believe that use of such a holistic approach will not just deliver higher genetic gains in research plots of scientists but also in farmers’ field. By using the entire approach from science of discovery to science of delivery, the scientific community will be able to contribute to both enhancing crop productivity and increasing farm profitability. Furthermore, it will also help to make cheap raw material available to industry, and finally, availability of affordable nutritious food to the poorest of the poor in society. Abbreviations: Abbreviations: AGS advanced generation selection BLUP best linear unbiased prediction EGS early generation selection GAB genomics-assisted breeding GS genomic selection MABC marker-assisted back crossing MARS marker-assisted recurrent selection MAS marker-assisted selection QDS quality declared seed RCGS rapid cycling genomic selection SNP single nucleotide polymorphism. Acknowledgements The authors from ICRISAT are thankful to the Bill & Melinda Gates Foundation (Tropical Legumes I, II & III), United States Agency for International Development (USAID), MARS Chocolate Inc., Indian Council of Agricultural Research (ICAR), and Department of Biotechnology (DBT) of Government of India. The work reported in this article was undertaken as a part of the CGIAR Research Program on Grain Legumes and Dryland Cereals (GLDC). ICRISAT is a member of the CGIAR. References Abate T , Alene AD , Bergvinson D , Shiferaw B , Silim S , Orr A , Asfaw S . 2012 . Tropical grain legumes in Africa and south Asia: knowledge and opportunities . Nairobi, Kenya : International Crops Research Institute for the Semi-Arid Tropics . https://core.ac.uk/download/pdf/12107473.pdf. Adu MO , Chatot A , Wiesel L , Bennett MJ , Broadley MR , White PJ , Dupuy LX . 2014 . A scanner system for high-resolution quantification of variation in root growth dynamics of Brassica rapa genotypes . Journal of Experimental Botany 65 , 2039 – 2048 . Google Scholar CrossRef Search ADS Aisawi KAB , Reynolds MP , Singh RP , Foulkes MJ . 2015 . The physiological basis of the genetic progress in yield potential of CIMMYT spring wheat cultivars from 1966 to 2009 . Crop Science 55 , 1749 . Google Scholar CrossRef Search ADS Akibode CS . 2011 . Trends in the production, trade, and consumption of food-legume crops in sub-Saharan Africa . MSc Thesis, Michigan State University . Akond M , Liu S , Schoener L et al. 2013 . SNP-based genetic linkage map of soybean using the SoySNP6K illumina infinium BeadChip genotyping array . Journal of Plant Genome Science 1 , 80 – 89 . Alzate-Marin AL , Menarim H , de Carvalho GA , de Paula TJ , de Barros EG , Moreira MA . 1999 . Improved selection with newly identified RAPD markers linked to resistance gene to four pathotypes of Colletotrichum lindemuthianum in common bean . Phytopathology 89 , 281 – 285 . Google Scholar CrossRef Search ADS Amare M , Asfaw S , Shiferaw B . 2012 . Welfare impacts of maize-pigeon pea intensification in Tanzania . Agricultural Economics 43 , 27 – 43 . Google Scholar CrossRef Search ADS Arruda MP , Lipka AE , Brown PJ et al. 2016 . Comparing genomic selection and marker-assisted selection for Fusarium head blight resistance in wheat (Triticum aestivum L.) . Molecular Breeding 36 , 84 . Google Scholar CrossRef Search ADS Atlin GN , Cairns JE , Das B . 2017 . Rapid breeding and varietal replacement are critical to adaptation of cropping systems in the developing world to climate change . Global Food Security 12 , 31 – 37 . Google Scholar CrossRef Search ADS Bankole F , Menkir A , Olaoye G , Crossa J , Hearne S , Unachukwu N , Gedil M . 2017 . Genetic gains in yield and yield related traits under drought stress and favorable environments in a maize population improved using marker assisted recurrent selection . Frontiers in Plant Science 8 , 808 . Google Scholar CrossRef Search ADS Barker T , Campos H , Cooper M , Dolan D , Edmeades G , Habben J , Schussler , Wright D , Zinselmeier C . 2010 . Improving drought tolerance in maize . Plant Breeding Reviews 25 , 173 – 253 . Bationo A , Lompo F , Koala S . 1998 . Research on nutrient flows and balances in West Africa: state-of the art . Agriculture, Ecosystems and Environment 71 , 19 – 35 . Bernardo R , Yu J . 2007 . Prospects for genome-wide selection for quantitative traits in maize . Crop Science 47 , 1082 . Google Scholar CrossRef Search ADS Bertioli DJ , Cannon SB , Froenicke L et al. 2016 . The genome sequences of Arachis duranensis and Arachis ipaensis, the diploid ancestors of cultivated peanut . Nature Genetics 48 , 438 – 446 . Google Scholar CrossRef Search ADS Beyene Y , Semagn K , Mugo S et al. 2015 . Genetic gains in grain yield through genomic selection in eight bi-parental maize populations under drought stress . Crop Science 55 , 154 . Google Scholar CrossRef Search ADS Bhatia VS , Singh P , Wani SP , Rao AVRK , Srinivas K . 2007 . Yield gap analysis of soybean, groundnut, pigeonpea and chickpea in India using simulation modeling . Journal of SAT Agricultural Research 5 , 1 – 160 . Bohra A , Pandey MK , Jha UC , Singh B , Singh IP , Datta D , Chaturvedi SK , Nadarajan N , Varshney RK . 2014 . Genomics-assisted breeding in four major pulse crops of developing countries: present status and prospects . Theoretical and Applied Genetics 127 , 1263 – 1291 . Google Scholar CrossRef Search ADS Boukar O , Fatokun CA , Huynh BL , Roberts PA , Close TJ . 2016 . Genomic tools in cowpea breeding programs: status and perspectives . Frontiers in Plant Science 7 , 757 . Google Scholar CrossRef Search ADS Bresson J , Vasseur F , Dauzat M , Koch G , Granier C , Vile D . 2015 . Quantifying spatial heterogeneity of chlorophyll fluorescence during plant growth and in response to water stress . Plant Methods 11 , 23 . Google Scholar CrossRef Search ADS Brisson N , Gate P , Gouache D , Charmet G , Oury F-X , Huard F . 2010 . Why are wheat yields stagnating in Europe? A comprehensive data analysis for France . Field Crops Research 119 , 201 – 212 . Google Scholar CrossRef Search ADS Bucksch A , Burridge J , York LM , Das A , Nord E , Weitz JS , Lynch JP . 2014 . Image-based high-throughput field phenotyping of crop roots . Plant Physiology 166 , 470 – 486 . Google Scholar CrossRef Search ADS Cabrera-Bosquet L , Fournier C , Brichet N , Welcker C , Suard B , Tardieu F . 2016 . High-throughput estimation of incident light, light interception and radiation-use efficiency of thousands of plants in a phenotyping platform . New Phytologist 212 , 269 – 281 . Google Scholar CrossRef Search ADS Chauhan Y , Wright G , Rachaputi NR , Krosch S , Robertson M , Hargreaves J , Broome A . 2007 . Using APSIM-soiltemp to simulate soil temperature in the podding zone of peanut . Australian Journal of Experimental Agriculture 47 , 992 – 999 . Google Scholar CrossRef Search ADS Chauhan Y , Wright G , Rachaputi N , McCosker K . 2008 . Identifying chickpea homoclimes using the APSIM chickpea model . Australian Journal of Agricultural Research 59 , 260 – 269 . Google Scholar CrossRef Search ADS Chen D , Neumann K , Friedel S , Kilian B , Chen M , Altmann T , Klukas C . 2014 . Dissecting the phenotypic components of crop plant growth and drought responses based on high-throughput image analysis . The Plant Cell 26 , 4636 – 4655 . Google Scholar CrossRef Search ADS Chen X , Li H , Pandey MK et al. 2016 . Draft genome of the peanut A-genome progenitor (Arachis duranensis) provides insights into geocarpy, oil biosynthesis and allergens . Proceedings of the National Academy of Sciences, USA 113 , 6785 – 6790 . Google Scholar CrossRef Search ADS Chen YL , Dunbabin VM , Diggle AJ , Siddique KHM , Rengel Z . 2011 . Development of a novel semi-hydroponic phenotyping system for studying root architecture . Functional Plant Biology 38 , 355 – 363 . Google Scholar CrossRef Search ADS Chu Y , Wu CL , Holbrook CC , Tillman BL , Person G , Ozias-Akins P . 2011 . Marker-assisted selection to pyramid nematode resistance and high oleic trait in peanut . The Plant Genome 4 , 110 – 117 . Google Scholar CrossRef Search ADS Clauw P , Coppens F , De Beuf K , Dhondt S , Van Daele T , Maleux K , Storme V , Clement L , Gonzalez N , Inzé D . 2015 . Leaf responses to mild drought stress in natural variants of Arabidopsis . Plant Physiology 167 , 800 – 816 . Google Scholar CrossRef Search ADS Clevenger J , Chu Y , Chavarro C et al. 2017 . Genome-wide SNP genotyping resolves signatures of selection and tetrasomic recombination in peanut . Molecular Plant 10 , 309 – 322 . Google Scholar CrossRef Search ADS Contreras-Soto RI , de Oliveira MB , Costenaro-da-Silva , Scapim CA , Schuster I . 2017 . Population structure, genetic relatedness and linkage disequilibrium blocks in cultivars of tropical soybean (Glycine max) . Euphytica 213 , 173 . Google Scholar CrossRef Search ADS Cooper M , van Eeuwijk FA , Hammer GL , Podlich DW , Messina C . 2009 . Modeling QTL for complex traits: detection and context for plant breeding . Current Opinion in Plant Biology 12 , 231 – 240 . Google Scholar CrossRef Search ADS Crossa J , de Campos GL , Pérez P et al. 2010 . Prediction of genetic values of quantitative traits in plant breeding using pedigree and molecular markers . Genetics 186 , 713 – 724 . Google Scholar CrossRef Search ADS Crossa J , Pérez P , de los Campos G , Mahuku G , Dreisigacker S , Magorokosho C . 2011 . Genomic selection and prediction in plant breeding . Journal of Crop Improvement 25 , 239 – 261 . Google Scholar CrossRef Search ADS Crossa J , Pérez-Rodríguez P , Cuevas J et al. 2017 . Genomic selection in plant breeding: methods, models, and perspectives . Trends in Plant Science 22 , 961 – 975 . Google Scholar CrossRef Search ADS Dar MH , de Janvry A , Emerick K , Raitzer D , Sadoulet E . 2013 . Flood-tolerant rice reduces yield variability and raises expected yield, differentially benefitting socially disadvantaged groups . Scientific Reports 3 , 3315 . Google Scholar CrossRef Search ADS de los Campos G , Naya H , Gianola D , Crossa J , Legarra A , Manfredi E , Weigel K , Cotes JM . 2009 . Predicting quantitative traits with regression models for dense molecular markers and pedigree . Genetics 182 , 375 – 385 . Google Scholar CrossRef Search ADS de Roos AP , Schrooten C , Veerkamp RF , van Arendonk JA . 2010 . The impact of genomic selection and short generation interval on dairy cattle breeding programs . In: Proceedings of 9th World Congress on Genetics Applied to Livestock Production . http://www2.naut.is/Files/Skra_0043312.pdf. Deulvot C , Charrel H , Marty A , Jacquin F , Donnadieu C , Lejeune-Hénaut I , Burstin J , Aubert G . 2010 . Highly-multiplexed SNP genotyping for genetic mapping and germplasm diversity studies in pea . BMC Genomics 11 , 468 . Google Scholar CrossRef Search ADS de Wit CT , Brouwer R , Penning de Vries FWT . 1970 . The simulation of photosynthetic systems . In: Setlik I , ed. Prediction and measurements of photosynthetic productivity . Proceedings of the IBP/PP Technical Meeting , Trebon . Wageningen : Pudoc , 47 – 50 . de Wit CT . 1978 . Simulation of assimilation, respiration and transpiration of crops. Simulation Monographs . Wageningen : Centre for Agricultural Publishing and Documentation . Doyle JJ , Luckow MA . 2003 . The rest of the iceberg. Legume diversity and evolution in a phylogenetic context . Plant Physiology 131 , 900 – 910 . Google Scholar CrossRef Search ADS Du H , Zeng X , Zhao M , Cui X , Wang Q , Yang H , Cheng H , Yu D . 2016 . Efficient targeted mutagenesis in soybean by TALENs and CRISPR/Cas9 . Journal of Biotechnology 217 , 90 – 97 . Google Scholar CrossRef Search ADS Duncan WG , Loomis RS , Williams WA , Hanau R . 1967 . A model for simulating photosynthesis in plant communities . Hilgardia 38 , 181 – 205 . Google Scholar CrossRef Search ADS Duvick DN . 2005 . Genetic progress in yield of United States maize . Maydica 50 , 193 – 202 . Eberbach PL , Hoffmann J , Moroni SJ , Wade LJ , Weston LA . 2013 . Rhizo-lysimetry: facilities for the simultaneous study of root behaviour and resource use by agricultural crop and pasture systems . Plant Methods 9 , 3 . Google Scholar CrossRef Search ADS Fahlgren N , Feldman M , Gehan MA et al. 2015a. A versatile phenotyping system and analytics platform reveals diverse temporal responses to water availability in Setaria . Molecular Plant 8 , 1520 – 1535 . Google Scholar CrossRef Search ADS Fahlgren N , Gehan MA , Baxter I . 2015b. Lights, camera, action: high-throughput plant phenotyping is ready for a close-up . Current Opinion in Plant Biology 24 , 93 – 99 . Google Scholar CrossRef Search ADS FAOSTAT . 2016 . Review of CGIAR priorities and strategies . Rome : Food and Agriculture Organization of the United Nations . www.fao.org/wairdocs/tac/x5756e/x5756e08.htm. Feleiro FG , Vinhadelli WS , Ragagnin VA , Vinhadelli WS , Moreira MA , Stavely JR , de Barros EG . 2001 . Resistance of bean lines to four races of Uromyces appenduculata isolated in the state of Minas Gerais . Fitopatologia Brasileiara 26 , 77 – 80 . Google Scholar CrossRef Search ADS Fiorani F , Schurr U . 2013 . Future scenarios for plant phenotyping . Annual Review of Plant Biology 64 , 267 – 291 . Google Scholar CrossRef Search ADS Foyer CH , Lam HM , Nguyen HT et al. 2016 . Neglecting legumes has compromised human health and sustainable food production . Nature Plants 2 , 16112 . Google Scholar CrossRef Search ADS García-Ruiz A , Cole JB , VanRaden PM et al. 2016 . Changes in genetic selection differentials and generation intervals in US Holstein dairy cattle as a result of genomic selection . Proceedings of the National Academy of Sciences, USA 113 , E3995 – E4004 . Google Scholar CrossRef Search ADS Ghafoor A , and McPhee K . 2012 . Marker assisted selection (MAS) for developing powdery mildew resistant pea cultivars . Euphytica 186 , 593 – 607 . Google Scholar CrossRef Search ADS Ghanem ME , Marrou H , Biradar C , Sinclair TR . 2015a. Production potential of lentil (Lens culinaris Medik.) in East Africa . Agricultural Systems 137 , 24 – 38 . Google Scholar CrossRef Search ADS Ghanem ME , Marrou H , Soltani A , Kumar S , Sinclair TR . 2015b. Lentil variation in phenology and yield evaluated with a model . Agronomy Journal 107 , 1967 – 77 . Google Scholar CrossRef Search ADS Gioia T , Galinski A , Lenz H et al. 2016 . GrowScreen-PaGe, a non-invasive, high-throughput phenotyping system based on germination paper to quantify crop phenotypic diversity and plasticity of root traits under varying nutrient supply . Functional Plant Biology 44 , 76 – 93 . Google Scholar CrossRef Search ADS Granier C , Aguirrezabal L , Chenu K et al. 2006 . PHENOPSIS, an automated platform for reproducible phenotyping of plant responses to soil water deficit in Arabidopsis thaliana permitted the identification of an accession with low sensitivity to soil water deficit . New Phytologist 169 , 623 – 635 . Google Scholar CrossRef Search ADS Granqvist B . 2006 . Is quality declared seed production an effective way to address seed and food security in Africa ?http://knowledge.cta.int/Dossiers/S-T-Policy/ACP-agricultural-S-T-dialogue/Demanding-Innovation/Feature-articles/Is-Quality-Declared-Seed-Production-an-effective-and-sustainable-way-to-address-Seed-and-Food-Security-in-Africa. Grisley W . 1993 . Seed for bean production in sub-Saharan Africa: issues, problems, and possible solutions . Agricultural Systems 43 , 19 – 33 . Google Scholar CrossRef Search ADS Gupta S , Nawaz K , Parween S , Roy R , Sahu K , Kumar Pole A , Khandal H , Srivastava R , Kumar Parida S , Chattopadhyay D . 2017 . Draft genome sequence of Cicer reticulatum L., the wild progenitor of chickpea provides a resource for agronomic trait improvement . DNA Research 24 , 1 – 10 . Gupta PK , Varshney RK . 2000 . The development and use of microsatellite markers for genetics and plant breeding with emphasis on bread wheat . Euphytica 113 , 163 – 185 . Google Scholar CrossRef Search ADS Hammer GL , Kropff MJ , Sinclair TR , Porter JR . 2002 . Future contributions of crop modelling—from heuristics and supporting decision making to understanding genetic regulation and aiding crop improvement . European Journal of Agronomy 18 , 15 – 31 . Google Scholar CrossRef Search ADS Hickey JM , Chiurugwi T , Mackay I , Powell W , Implementing Genomic Selection in CGIAR Breeding Programs Workshop Participants . 2017 . Genomic prediction unifies animal and plant breeding programs to form platforms for biological discovery . Nature Genetics 49 , 1297 – 1303 . Google Scholar CrossRef Search ADS Holzworth DP , Huth NI , Zurcher EJ et al. 2014 . APSIM – evolution towards a new generation of agricultural systems simulation . Environmental Modelling & Software 62 , 327 – 350 . Google Scholar CrossRef Search ADS Hoogenboom G , Porter CH , Shelia V et al. 2017 . Decision Support System for Agrotechnology Transfer (DSSAT) Version 4.7 , https://DSSAT.net. Gainesville, FL, USA: DSSAT Foundation . Huth NI , Thorburn PJ , Radford BJ , Thornton CM . 2010 . Impacts of fertilisers and legumes on N2O and CO2 emissions from soils in subtropical agricultural systems: a simulation study . Agriculture Ecosystems & Environment 136 , 351 – 357 . Google Scholar CrossRef Search ADS Hyten DL , Song Q , Choi IY , Yoon MS , Specht JE , Matukumalli LK , Nelson RL , Shoemaker RC , Young ND , Cregan PB . 2008 . High-throughput genotyping with the GoldenGate assay in the complex genome of soybean . Theoretical and Applied Genetics 116 , 945 – 952 . Google Scholar CrossRef Search ADS Isidro J , Jannink JL , Akdemir D , Poland J , Heslot N , Sorrells ME . 2015 . Training set optimization under population structure in genomic selection . Theoretical and Applied Genetics 128 , 145 – 158 . Google Scholar CrossRef Search ADS Iversen CM , Murphy MT , Allen MF , Childs J , Eissenstat DM , Lilleskov EA , Sarjala TM , Sloan VL , Sullivan PF . 2011 . Advancing the use of minirhizotrons in wetlands . Plant Soil 352 , 23 – 39 . Google Scholar CrossRef Search ADS Jaganathan D , Thudi M , Kale S , Azam S , Roorkiwal M , Gaur PM , Kishor PB , Nguyen H , Sutton T , Varshney RK . 2015 . Genotyping-by-sequencing based intra-specific genetic map refines a “QTL-hotspot” region for drought tolerance in chickpea . Molecular Genetics and Genomics 290 , 559 – 571 . Google Scholar CrossRef Search ADS Janila P , Pandey MK , Shasidhar Y et al. 2016 . Molecular breeding for introgression of fatty acid desaturase mutant alleles (ahFAD2A and ahFAD2B) enhances oil quality in high and low oil containing peanut genotypes . Plant Science 242 , 203 – 213 . Google Scholar CrossRef Search ADS Jannink JL , Lorenz AJ , Iwata H . 2010 . Genomic selection in plant breeding: from theory to practice . Briefings in Functional Genomics 9 , 166 – 177 . Google Scholar CrossRef Search ADS Jarquín D , Kocak K , Posadas L , Hyma K , Jedlicka J , Graef G , Lorenz A . 2014 . Genotyping by sequencing for genomic prediction in a soybean breeding population . BMC Genomics 15 , 740 . Google Scholar CrossRef Search ADS Jeudy C , Adrian M , Baussard C et al. 2016 . RhizoTubes as a new tool for high throughput imaging of plant root development and architecture: test, comparison with pot grown plants and validation . Plant Methods 12 , 31 . Google Scholar CrossRef Search ADS Johansen C , Siddique K . 2017 . Grain legumes in integrated crop . In: Shivasankar S , Bergvinson D , Gaur P , Kumar S , Beebe S , Tamo M , eds. Achieving sustainable cultivation of grain legumes. Chapter 10, Grain legumes in integrated crop management systems . Sawston, UK : Burleigh Dodds Science Publishing Limited . Jones JW , Hoogenboom G , Porter CH , Boote KJ , Batchelor WD , Hunt LA , Wilkens PW , Singh U , Gijsman AJ , Ritchie JT . 2003 . DSSAT cropping system model . European Journal of Agronomy 18 , 235 – 265 . Google Scholar CrossRef Search ADS Kale SM , Jaganathan D , Ruperao P et al. 2015 . Prioritization of candidate genes in “QTL-hotspot” region for drought tolerance in chickpea (Cicer arietinum L.) . Scientific Reports 5 , 15296 . Google Scholar CrossRef Search ADS Kamanga BCG , Waddington SR , Whitbread A , Almekinders CJM , Giller K . 2013 . Improving the efficiency of use of small amounts of nitrogen and phosphorus fertiliser on smallholder maize in central Malawi . Experimental Agriculture 50 , 229 – 249 . Google Scholar CrossRef Search ADS Kamburova VS , Nikitina EV , Shermatov SE , Buriev ZT , Kumpatla SP , Emani C , Abdurakhmonov IY . 2017 . Genome editing in plants: an overview of tools and applications . International Journal of Agronomy 2017 , 7315351 . Google Scholar CrossRef Search ADS Kang YJ , Kim SK , Kim MY et al. 2014 . Genome sequence of mungbean and insights into evolution within Vigna species . Nature Communications 5 , 5443 . Google Scholar CrossRef Search ADS Kang YJ , Satyawan D , Shim S et al. 2015 . Draft genome sequence of adzuki bean, Vigna angularis . Scientific Reports 5 , 8069 . Google Scholar CrossRef Search ADS Keating BA , Carberry PS , Hammer GL et al. 2003 . An overview of APSIM, a model designed for farming systems simulation . European Journal of Agronomy 18 , 267 – 288 . Google Scholar CrossRef Search ADS Khanh T , Anh T , Buu B et al. 2013 . Applying molecular breeding to improve soybean rust resistance in Vietnamese elite soybean . American Journal of Plant Sciences 4 , 1 – 6 . Google Scholar CrossRef Search ADS Kholová J , Murugesan T , Kaliamoorthy S et al. 2014 . Modelling the effect of plant water use traits on yield and stay-green expression in sorghum . Functional Plant Biology 41 , 1019 – 1034 . Google Scholar CrossRef Search ADS Kim HK , van Oosterom E , Dingkuhn M , Luquet D , Hammer G . 2010 . Regulation of tillering in sorghum: environmental effects . Annals of Botany 106 , 57 – 67 . Google Scholar CrossRef Search ADS Kim JS . 2016 . Genome editing comes of age . Nature Protocols 11 , 1573 – 1578 . Google Scholar CrossRef Search ADS Koester RP , Skoneczka JA , Cary TR , Diers BW , Ainsworth EA . 2014 . Historical gains in soybean (Glycine max Merr.) seed yield are driven by linear increases in light interception, energy conversion, and partitioning efficiencies . Journal of Experimental Botany 65 , 3311 – 3321 . Google Scholar CrossRef Search ADS Kuijken RC , van Eeuwijk FA , Marcelis LF , Bouwmeester HJ . 2015 . Root phenotyping: from component trait in the lab to breeding . Journal of Experimental Botany 66 , 5389 – 5401 . Google Scholar CrossRef Search ADS Kumar SR , Hammer GL , Broad I , Harland P , McLean G . 2009 . Modelling environmental effects on phenology and canopy development of diverse sorghum genotypes . Field Crops Research 111 , 157 – 165 . Google Scholar CrossRef Search ADS Kumar V , Khan AW , Saxena RK , Garg V , Varshney RK . 2016 . First-generation HapMap in Cajanus spp. reveals untapped variations in parental lines of mapping 1 populations . Plant Biotechnology Journal 14 , 1673 – 1681 . Google Scholar CrossRef Search ADS Lam HM , Xu X , Liu X et al. 2010 . Resequencing of 31 wild and cultivated soybean genomes identifies patterns of genetic diversity and selection . Nature Genetics 42 , 1053 – 1059 . Google Scholar CrossRef Search ADS Landau-Ellis D , Pantalone VR . 2009 . Marker-assisted backcrossing to incorporate two low phytate alleles into the Tennessee soybean cultivar 5601T . In: Shu GY , ed. Induced plant mutations in the genomics era . Rome : Food and Agriculture Organization of the United Nations (FAO) , 316 – 318 . Lavaud C , Lesné A , Piriou C , Le Roy G , Boutet G , Moussart A , Poncet C , Delourme R , Baranger A , Pilet-Nayel ML . 2015 . Validation of QTL for resistance to Aphanomyces euteiches in different pea genetic backgrounds using near-isogenic lines . Theoretical and Applied Genetics 128 , 2273 – 2288 . Google Scholar CrossRef Search ADS Le Marié C , Kirchgessner N , Flütsch P , Pfeifer J , Walter A , Hund A . 2016 . RADIX: rhizoslide platform allowing high throughput digital image analysis of root system expansion . Plant Methods 12 , 40 . Google Scholar CrossRef Search ADS Le Marié C , Kirchgessner N , Marschall D , Walter A , Hund A . 2014 . Rhizoslides: paper-based growth system for non-destructive, high throughput phenotyping of root development by means of image analysis . Plant Methods 10 , 13 . Google Scholar CrossRef Search ADS Leamy LJ , Zhang H , Li C , Chen CY , Song BH . 2017 . A genome-wide association study of seed composition traits in wild soybean (Glycine soja) . BMC Genomics 18 , 18 . Google Scholar CrossRef Search ADS Lee YG , Jeong N , Kim JH et al. 2015 . Development, validation and genetic analysis of a large soybean SNP genotyping array . The Plant Journal 81 , 625 – 636 . Google Scholar CrossRef Search ADS Li Z , Liu ZB , Xing A , Moon BP , Koellhoffer JP , Huang L , Ward RT , Clifton E , Falco SC , Cigan AM . 2015 . Cas9-Guide RNA directed genome editing in soybean . Plant Physiology 169 , 960 – 970 . Google Scholar CrossRef Search ADS Lobell DB , Asseng S . 2017 . Comparing estimates of climate change impacts from process based and statistical crop models . Environment Research Letters 12 , 015001 . Google Scholar CrossRef Search ADS Lopes MS , Reynolds MP , Manes Y , Singh RP , Crossa J , Braun HJ . 2012 . Genetic yield gains and changes in associated traits of CIMMYT spring bread wheat in an historic set representing 30 years of breeding . Crop Science 52 , 1123 . Google Scholar CrossRef Search ADS Lorenz AJ , Chao S , Asoro FG , Heffner EL . 2011 . Genomic selection in plant breeding: knowledge and prospects . Advances in Agronomy 110 , 77 – 123 . Google Scholar CrossRef Search ADS Lucas MR , Huynh BL , Roberts PA , Close TJ . 2015 . Introgression of a rare haplotype from Southeastern Africa to breed California blackeyes with larger seeds . Frontiers in Plant Science 6 , 126 . Ma Y , Reif JC , Jiang Y et al. 2016 . Potential of marker selection to increase prediction accuracy of genomic selection in soybean (Glycine max L.) . Molecular Breeding 36 , 113 . Google Scholar CrossRef Search ADS Maeght JL , Rewald B , Pierret A . 2013 . How to study deep roots—and why it matters . Frontiers in Plant Science 4 , 299 . Google Scholar CrossRef Search ADS Maereka EK , Rubyogo JC . 2015 . Integrated bean seed systems in Africa: implications for community seed production . In: Ojiewo CO , Kugbei S , Bishaw Z , Rubyogo JC , eds. Community seed production . Workshop proceedings , December 2013 . Addis Ababa : ICRISAT and Rome : FAO , 9 – 11 . Maldonado dos Santos JV , Valliyodan B , Joshi T et al. 2016 . Evaluation of genetic variation among Brazilian soybean cultivars through genome resequencing . BMC Genomics 17 , 110 . Google Scholar CrossRef Search ADS Mairhofer S , Zappala S , Tracy SR , Sturrock C , Bennett M , Mooney SJ , Pridmore T . 2012 . RooTrak: automated recovery of three-dimensional plant root architecture in soil from x-ray microcomputed tomography images using visual tracking . Plant Physiology 158 , 561 – 569 . Google Scholar CrossRef Search ADS Marrou H , Sinclair TR , Metral R . 2014 . Assessment of irrigation scenarios to improve performances of Lingot bean (Phaseolus vulgaris) in southwest France . European Journal of Agronomy 59 , 5922 – 5928 . Google Scholar CrossRef Search ADS Massman JM , Gordillo A , Lorenzana RE , Bernardo R . 2013 . Genomewide predictions from maize single-cross data . Theoretical and Applied Genetics 126 , 13 – 22 . Google Scholar CrossRef Search ADS Mathieu L , Lobet G , Tocquin P , Périlleux C . 2015 . “Rhizoponics”: a novel hydroponic rhizotron for root system analyses on mature Arabidopsis thaliana plants . Plant Methods 11 , 3 . Google Scholar CrossRef Search ADS McGuire S , Sperling L . 2016 . Seed systems smallholder farmers use . Food Security 8 , 179 – 195 . Google Scholar CrossRef Search ADS Meng Y , Hou Y , Wang H , Ji R , Liu B , Wen J , Niu L , Lin H . 2017 . Targeted mutagenesis by CRISPR/Cas9 system in the model legume Medicago truncatula . Plant Cell Reports 36 , 371 – 374 . Google Scholar CrossRef Search ADS Messina CD , Podlich D , Dong Z , Samples M , Cooper M . 2011 . Yield-trait performance landscapes: from theory to application in breeding maize for drought tolerance . Journal of Experimental Botany 62 , 855 – 868 . Google Scholar CrossRef Search ADS Meuwissen TH , Hayes BJ , Goddard ME . 2001 . Prediction of total genetic value using genome-wide dense marker maps . Genetics 157 , 1819 – 1829 . Michno JM , Wang X , Liu J , Curtin SJ , Kono TJ , Stupar RM . 2015 . CRISPR/Cas mutagenesis of soybean and Medicago truncatula using a new web-tool and a modified Cas9 enzyme . GM Crops & Food 6 , 243 – 252 . Google Scholar CrossRef Search ADS Miklas PN , Kelly JD , Beebe SE , Blair MW . 2006 . Common bean breeding for resistance against biotic and abiotic stresses: from classical to MAS breeding . Euphytica 147 , 105 – 131 . Google Scholar CrossRef Search ADS Miklas PN , Smith JR , Riley R , Grafton KF , Singh SP , Jung G , Coyne DP . 2000 . Marker-assisted breeding for pyramided resistance to common bacterial blight in common bean . Annual Reports Bean Improvement Cooperatives 43 , 39 – 40 . Monyo ES , Varshney RK . 2016 . Seven seasons of learning and engaging smallholder farmers in the drought-prone areas of sub-Saharan Africa and South Asia through Tropical Legumes, 2007–2014 . Patancheru, Telangana, India : International Crops Research Institute for the Semi-Arid Tropics . Mooney S , Pridmore T , Helliwell J , Bennett MJ . 2012 . Developing X-ray computed tomography to non-invasively image 3-D root systems architecture in soil . Plant Soil 352 , 1 – 22 . Google Scholar CrossRef Search ADS Muchero W , Diop NN , Bhat PR et al. 2009 . A consensus genetic map of cowpea [Vigna unguiculata (L) Walp.] and synteny based on EST-derived SNPs . Proceedings of the National Academy of Sciences, USA 106 , 18159 – 18164 . Google Scholar CrossRef Search ADS Muñoz-Amatriaín M , Mirebrahim H , Xu P et al. 2017 . Genome resources for climate-resilient cowpea, an essential crop for food security . The Plant Journal 89 , 1042 – 1054 . Google Scholar CrossRef Search ADS Mutlu N , Miklas PN , Reiser J et al. 2005 . Backcross breeding for improved resistance to common bacterial blight in pinto bean (Phaseolus vulgaris L.) . Plant Breeding 124 , 282 – 287 . Google Scholar CrossRef Search ADS Nagy ED , Guo Y , Tang S et al. 2012 . A high-density genetic map of Arachis duranensis, a diploid ancestor of cultivated peanut . BMC Genomics 13 , 469 . Google Scholar CrossRef Search ADS Neumann K , Klukas C , Friedel S , Rischbeck P , Chen D , Entzian A , Stein N , Graner A , Kilian B . 2015 . Dissecting spatiotemporal biomass accumulation in barley under different water regimes using high-throughput image analysis . Plant, Cell & Environment 38 , 1980 – 1996 . Google Scholar CrossRef Search ADS O’Boyle PD , Kelly JD . 2007 . Use of marker-assisted selection to breed for resistance to common bacterial blight in common bean . Journal of American Society of Horticultural Science 132 , 381 – 386 . Oliveira LK , Melo LC , Brondani C , Peloso MJ , Brondani RP . 2008 . Backcross assisted by microsatellite markers in common bean . Genetics and Molecular Research 7 , 1000 – 1010 . Google Scholar CrossRef Search ADS Pandey MK , Roorkiwal M , Singh VK , Ramalingam A , Kudapa H , Thudi M , Chitikineni A , Rathore A , Varshney RK . 2016 . Emerging genomic tools for legume breeding: current status and future prospects . Frontiers in Plant Science 7 , 455 . Pandey MK , Agarwal G , Kale SM et al. 2017a. Development and evaluation of a high density genotyping ‘Axiom_Arachis’ array with 58K SNPs for accelerating genetics and breeding in groundnut . Scientific Reports 7 , 40577 . Google Scholar CrossRef Search ADS Pandey MK , Khan AW , Singh VK et al. 2017b. QTL-seq approach identified genomic regions and diagnostic markers for rust and late leaf spot resistance in groundnut (Arachis hypogaea L.) . Plant Biotechnology Journal 15 , 927 – 941 . Google Scholar CrossRef Search ADS Parent B , Tardieu F . 2014 . Can current crop models be used in the phenotyping era for predicting the genetic variability of yield of plants subjected to drought or high temperature ? Journal of Experimental Botany 65 , 6179 – 6189 . Google Scholar CrossRef Search ADS Parhe SD , Chimote VP , Deshmukh MP , Chandra K , Akash M . 2017 . Marker-assisted pyramiding of four QTL/genes for Asian rust (Phakopsora pachyrhizi) resistance in soybean . Journal of Crop Improvement 31 , 689 – 711 . Google Scholar CrossRef Search ADS Parween S , Nawaz K , Roy R et al. 2015 . An advanced draft genome assembly of a desi type chickpea (Cicer arietinum L.) . Scientific Reports 5 , 12806 . Google Scholar CrossRef Search ADS Pazhamala LT , Purohit S , Saxena RK , Garg V , Krishnamurthy L , Verdier J , Varshney RK . 2017 . Gene expression atlas of pigeonpea and its application to gain insights into genes associated with pollen fertility implicated in seed formation . Journal of Experimental Botany 68 , 2037 – 2054 . Google Scholar CrossRef Search ADS Pereyra-Irujo GA , Gasco ED , Peirone LS , Aguirrezabal L . 2012 . GlyPh: a low-cost platform for phenotyping plant growth and water use . Functional Plant Biology 39 , 905 – 913 . Google Scholar CrossRef Search ADS Plant Production and Protection Division . 2006 . Quality declared seed system . FAO Plant Production and Protection Paper 185. Rome : Food and Agriculture Organization of the United Nations . PubMed PubMed Pratap A , Chaturvedi SK , Tomar R , Rajan N , Malviya N , Thudi M , Saabale PR , Prajapati U , Varshney RK , Singh NP . 2017 . Marker-assisted introgression of resistance to fusarium wilt race 2 in Pusa 256, an elite cultivar of desi chickpea . Molecular Genetics and Genomics 292 , 1237 – 1245 . Google Scholar CrossRef Search ADS Price AL , Zaitlen NA , Reich D , Patterson N . 2010 . New approaches to population stratification in genome-wide association studies . Nature Reviews Genetics 11 , 459 – 463 . Google Scholar CrossRef Search ADS Prince SJ , Murphy M , Mutava RN , Durnell LA , Valliyodan B , Shannon JG , Nguyen HT . 2017 . Root xylem plasticity to improve water use and yield in water-stressed soybean . Journal of Experimental Botany 68 , 2027 – 2036 . Rellán-Álvarez R , Lobet G , Lindner H et al. 2015 . GLO-Roots: an imaging platform enabling multidimensional characterization of soil-grown root systems . eLife 4 , 07597 . Google Scholar CrossRef Search ADS Reynolds M , Langridge P . 2016 . Physiological breeding . Current Opinion in Plant Biology 31 , 162 – 171 . Google Scholar CrossRef Search ADS Richard CA , Hickey LT , Fletcher S , Jennings R , Chenu K , Christopher JT . 2015 . High-throughput phenotyping of seminal root traits in wheat . Plant Methods 11 , 13 . Google Scholar CrossRef Search ADS Riedelsheimer C , Endelman JB , Stange M , Sorrells ME , Jannink JL , Melchinger AE . 2013 . Genomic predictability of interconnected biparental maize populations . Genetics 194 , 493 – 503 . Google Scholar CrossRef Search ADS Roorkiwal M , Jain A , Kale SM , Doddamani D , Chitikineni A , Thudi M , Varshney RK . 2017 . Development and evaluation of high density SNP array (Axiom® CicerSNP Array) for high resolution genetic mapping and breeding applications in chickpea . Plant Biotechnology Journal 16 , 890 – 901 . Google Scholar CrossRef Search ADS Roorkiwal M , Rathore A , Das RR et al. 2016 . Genome-enabled prediction models for yield related traits in Chickpea . Frontiers in Plant Science 7 , 1666 . Google Scholar CrossRef Search ADS Roorkiwal M , Sawargaonkar SL , Chitikineni A , Thudi M , Saxena RK , Upadhyaya HD , Vales MI , Riera-Lizarazu O , Varshney RK . 2013 . Single nucleotide polymorphism genotyping for breeding and genetics applications in chickpea and pigeonpea using the BeadXpress platform . The Plant Genome 6 , doi: 10.3835/plantgenome2013.05.0017 . Rubyogo JC , Myer GM , Ajeigbe H et al. 2016 . Integrated seed systems delivering on the promise: experiences from Tropical Legumes II . In: Monyo ES , Varshney RK . Seven seasons of learning and engaging smallholder farmers in the drought-prone areas of sub-Saharan Africa and South Asia through Tropical Legumes, 2007–2014 . Patancheru, Telangana, India : International Crops Research Institute for the Semi-Arid Tropics , 167 – 178 . Rubyogo JC , Sperling JC , Muthoni R , Buruchara R . 2010 . Bean seed delivery for small farmers in sub-saharan Africa: the power of partnerships . Society & Natural Resources 4 , 285 – 302 . Google Scholar CrossRef Search ADS Saatchi M , McClure MC , McKay SD et al. 2011 . Accuracies of genomic breeding values in American Angus beef cattle using K-means clustering for cross-validation . Genetics, Selection, Evolution 43 , 40 . Google Scholar CrossRef Search ADS Saghai-Maroof MA , Jeong SC , Gunduz I , Tucker DM , Buss GR , Tolin S . 2008 . Pyramiding of soybean mosaic virus resistance genes by marker assisted selection . Crop Science 48 , 517 – 526 . Google Scholar CrossRef Search ADS Sanginga N , Dashiell K , Diels J et al. 2003 . Sustainable resource management coupled to resilient germplasm to provide new intensive cereal–grain–legume–livestock systems in the dry Savanna . Agriculture, Ecosystems and Environment 100 , 305 – 314 . Google Scholar CrossRef Search ADS Saxena K , Saxena RK , Varshney RK . 2017 . Use of immature seed germination and single seed descent for rapid genetic gains in pigeonpea . Plant Breeding 136 , 954 – 957 . Google Scholar CrossRef Search ADS Schmutz J , Cannon SB , Schlueter J et al. 2010 . Genome sequence of the palaeopolyploid soybean . Nature 463 , 178 – 183 . Google Scholar CrossRef Search ADS Schmutz J , McClean PE , Mamidi S et al. 2014 . A reference genome for common bean and genome-wide analysis of dual domestications . Nature Genetics 46 , 707 – 713 . Google Scholar CrossRef Search ADS Schopp P , Müller D , Technow F , Melchinger AE . 2017 . Accuracy of genomic prediction in synthetic populations depending on the number of parents, relatedness, and ancestral linkage disequilibrium . Genetics 205 , 441 – 454 . Google Scholar CrossRef Search ADS Sennhenn A , Njarui DMG , Maass BL , Whitbread AM . 2017 . Exploring niches for short-season grain legumes in semi-arid Eastern Kenya—coping with the impacts of climate variability . Frontiers in Plant Science 8 , 699 . Google Scholar CrossRef Search ADS Septiningsih EM , Collard BC , Heuer S , Bailey-Serres J , Ismail AM , Mackill DJ . 2013 . Applying genomics tools for breeding submergence tolerance in rice . In: Varshney RK , Tuberosa R , eds. Translational genomics for crop breeding, Vol II, Abiotic stress, yield and quality . John Wiley & Sons, Inc ., 9 – 30 . Google Scholar CrossRef Search ADS Shi A , Chen P , Li D , Zheng C , Zhang B , Hou A . 2009 . Pyramiding multiple genes for resistance to soybean mosaic virus in soybean using molecular markers . Molecular Breeding 23 , 113 – 124 . Google Scholar CrossRef Search ADS Siddique KH , Johansen C , Turner NC , Jeuffroy M-H , Hashem A , Sakar D , Gan Y , Alghamdi S . 2012 . Innovations in agronomy for food legumes. A review . Agronomy for Sustainable Development 32 , 45 – 64 . Google Scholar CrossRef Search ADS Sinclair TR , Marrou H , Soltani A , Vadez V , Chandolu KC . 2014 . Soybean production potential in Africa . Global Food Security 3 , 31 – 40 . Google Scholar CrossRef Search ADS Sinclair TR , Messina CD , Beatty A , Samples M . 2010 . Assessment across the United States of the benefits of altered soybean drought traits . Agronomy Journal 102 , 475 – 482 . Google Scholar CrossRef Search ADS Sinclair TR , Vadez V . 2012 . The future of grain legumes in cropping systems . Crop and Pasture Science 63 , 501 – 512 . Google Scholar CrossRef Search ADS Singh VK , Khan AW , Jaganathan D et al. 2016 . QTL-seq for rapid identification of candidate genes for 100-seed weight and root/total plant dry weight ratio under rainfed conditions in chickpea . Plant Biotechnology Journal 14 , 2110 – 2119 . Google Scholar CrossRef Search ADS Singh VK , Khan AW , Saxena RK et al. 2017 . Indel-seq: a fast-forward genetics approach for identification of trait-associated putative candidate genomic regions and its application in pigeonpea (Cajanus cajan) . Plant Biotechnology Journal 15 , 906 – 914 . Google Scholar CrossRef Search ADS Singh T , Sharma A . 2016 . Early generation selection for yield and its related traits in soybean [Glycine max (L.) Merrill.] . Legume Research 39 , 343 – 348 . Sita K , Sehgal A , HanumanthaRao B et al. 2017 . Food legumes and rising temperatures: effects, adaptive functional mechanisms specific to reproductive growth stage and strategies to improve heat tolerance . Frontiers in Plant Science 8 , 1658 . Google Scholar CrossRef Search ADS Skirycz A , Vandenbroucke K , Clauw P et al. 2011 . Survival and growth of Arabidopsis plants given limited water are not equal . Nature Biotechnology 29 , 212 – 214 . Google Scholar CrossRef Search ADS Smyth SJ . 2017 . Canadian regulatory perspectives on genome engineered crops . GM Crops & Food 8 , 35 – 43 . Google Scholar CrossRef Search ADS Soltani A , Sinclair TR . 2011 . A simple model for chickpea development, growth and yield . Field Crops Research 124 , 252 – 260 . Google Scholar CrossRef Search ADS Soltani A , Sinclair TR . 2012 . Modeling physiology of crop development, growth and yield . Wallingford, UK : CABI . Google Scholar CrossRef Search ADS Song Q , Hyten DL , Jia G , Quigley CV , Fickus EW , Nelson RL , Cregan PB . 2013 . Development and evaluation of SoySNP50K, a high-density genotyping array for soybean . PLoS One 8 , e54985 . Google Scholar CrossRef Search ADS Song Q , Jia G , Hyten DL et al. 2015 . SNP assay development for linkage map construction, anchoring whole-genome sequence, and other genetic and genomic applications in common bean . G3: Genes, Genomes, Genetics 5 , 2285 – 2290 . Sperling L , McGuire S . 2010 . Understanding and strengthening informal seed markets . Experimental Agriculture 46 , 119 – 136 . Google Scholar CrossRef Search ADS Sprink T , Eriksson D , Schiemann J , Hartung F . 2016 . Regulatory hurdles for genome editing: process- vs. product-based approaches in different regulatory contexts . Plant Cell Reports 35 , 1493 – 1506 . Google Scholar CrossRef Search ADS Stagnari F , Maggio A , Galieni A , Pisante M . 2017 . Multiple benefits of legumes for agriculture sustainability: an overview . Chemical and Biological Technologies in Agriculture 4 , 2 . Google Scholar CrossRef Search ADS Stavely JR . 2000 . Pyramiding rust and viral resistance genes using traditional and marker techniques in common bean . Annual Reports Bean Improvement Cooperatives 43 , 1 – 4 . Sun X , Hu Z , Chen R , Jiang Q , Song G , Zhang H , Xi Y . 2015 . Targeted mutagenesis in soybean using the CRISPR-Cas9 system . Scientific Reports 5 , 10342 . Google Scholar CrossRef Search ADS Taran B , Buchwaldt L , Tullu A , Banniza S , Warkentin TD , Vandenberg A . 2003 . Using molecular markers to pyramid genes for resistance to ascochyta blight and anthracnose in lentil (Lens culinaris Medik.) . Euphytica 134 , 223 – 230 . Google Scholar CrossRef Search ADS Tardieu F , Cabrera-Bosquet L , Pridmore T , Bennett M . 2017 . Plant phenomics, from sensors to knowledge . Current Biology 27 , R770 – R783 . Google Scholar CrossRef Search ADS Tardieu F , Tuberosa R . 2010 . Dissection and modelling of abiotic stress tolerance in plants . Current Opinion in Plant Biology 13 , 206 – 212 . Google Scholar CrossRef Search ADS Tayeh N , Aluome C , Falque M et al. 2015a. Development of two major resources for pea genomics: the GenoPea 13.2K SNP array and a high-density, high-resolution consensus genetic map . The Plant Journal 84 , 1257 – 1273 . Google Scholar CrossRef Search ADS Tayeh N , Aubert G , Pilet-Nayel ML , Lejeune-Hénaut I , Warkentin TD , Burstin J . 2015b. Genomic tools in pea breeding programs: status and perspectives . Frontiers in Plant Science 6 , 1037 . Tayeh N , Klein A , Le Paslier MC et al. 2015c. Genomic prediction in pea: effect of marker density and training population size and composition on prediction accuracy . Frontiers in Plant Science 6 , 941 . Teixeira EI , Brown HE , Sharp J , Meenken ED , Ewert F . 2015 . Evaluating methods to simulate crop rotations for climate impact assessments—a case study on the Canterbury plains of New Zealand . Environmental Modelling & Software 72 , 304 – 313 . Google Scholar CrossRef Search ADS Tekwa IJ , Ijabula ST , Maijama’a NP . 2010 . Effect of herbicides, seed dressing chemicals and spray regimes on germination, insect infestation and yield of cowpea (Vigna unguiculata (L) Walp) . Australian Journal of Agricultural Engineering 1 , 14 – 17 . Thudi M , Chitikineni A , Liu X et al. 2016a. Recent breeding programs enhanced genetic diversity in both desi and kabuli varieties of chickpea (Cicer arietinum L.) . Scientific Reports 6 , 38636 . Google Scholar CrossRef Search ADS Thudi M , Khan AW , Kumar V , Gaur PM , Katta K , Garg V , Roorkiwal M , Samineni S , Varshney RK . 2016b. Whole genome re-sequencing reveals genome-wide variations among parental lines of 16 mapping populations in chickpea (Cicer arietinum L.) . BMC Plant Biology 16 Suppl 1 , 10 . Google Scholar CrossRef Search ADS Tisné S , Serrand Y , Bach L et al. 2013 . Phenoscope: an automated large-scale phenotyping platform offering high spatial homogeneity . The Plant Journal 74 , 534 – 544 . Google Scholar CrossRef Search ADS Trachsel S , Kaeppler SM , Brown KM , Lynch JP . 2011 . Shovelomics: High throughput phenotyping of maize (Zea mays L.) root architecture in the field . Plant Soil 341 , 75 – 87 . Google Scholar CrossRef Search ADS Vadez V . 2014 . Root hydraulics: the forgotten side of root in drought adaptation . Field Crops Research 165 , 15 – 24 . Google Scholar CrossRef Search ADS Vadez V , Halilou O , Hissene HM , Sibiry-Traore P , Sinclair TR , Soltani A . 2017 . Mapping water stress incidence and intensity, optimal plant populations, and cultivar duration for African groundnut productivity enhancement . Frontiers in Plant Science 8 , 432 . Google Scholar CrossRef Search ADS Vadez V , Kholová J , Hummel G , Zhokhavets U , Gupta SK , Hash CT . 2015 . LeasyScan: a novel concept combining 3D imaging and lysimetry for high-throughput phenotyping of traits controlling plant water budget . Journal of Experimental Botany 66 , 5581 – 5593 . Google Scholar CrossRef Search ADS Vadez V , Kholova J , Zaman-Allah M , Belko N . 2013a. Water: the most important ‘molecular’ component of water stress tolerance research . Functional Plant Biology 40 , 1310 – 1322 . Google Scholar CrossRef Search ADS Vadez V , Soltani A , Sinclair TR . 2012 . Modelling possible benefits of root related traits to enhance terminal drought adaptation of chickpea . Field Crops Research 137 , 108 – 115 . Google Scholar CrossRef Search ADS Vadez V , Soltani A , Sinclair TR . 2013b. Crop simulation analysis of phenological adaptation of chickpea to different latitudes of India . Field Crops Research 146 , 1 – 9 . Google Scholar CrossRef Search ADS Valdisser PAMR , Pereira WJ , Almeida Filho JE et al. 2017 . In-depth genome characterization of a Brazilian common bean core collection using DArTseq high-density SNP genotyping . BMC Genomics 18 , 423 . Google Scholar CrossRef Search ADS Valliyodan B , Dan Qiu , Patil G et al. 2016 . Landscape of genomic diversity and trait discovery in soybean . Scientific Reports 6 , 23598 . Google Scholar CrossRef Search ADS Van Raden PM , Van Tassell CP , Wiggans GR , Sonstegard TS , Schnabel RD , Taylor JF , Schenkel FS . 2009 . Reliability of genomic predictions for North American Holstein bulls . Journal of Dairy Science 92 , 16 – 24 . Google Scholar CrossRef Search ADS Varshney RK . 2016 . Exciting journey of 10 years from genomes to fields and markets: some success stories of genomics-assisted breeding in chickpea, pigeonpea and groundnut . Plant Science 242 , 98 – 107 . Google Scholar CrossRef Search ADS Varshney RK , Chen W , Li Y et al. 2012a. Draft genome sequence of pigeonpea (Cajanus cajan), an orphan legume crop of resource-poor farmers . Nature Biotechnology 30 , 83 – 89 . Google Scholar CrossRef Search ADS Varshney RK , Gaur PM , Chamarthi SK , Krishnamurthy L , Tripathi S , Kashiwagi J , Samineni S , Singh VK , Thudi M , Jaganathan D . 2013a. Fast-track introgression of “QTL-hotspot” for root traits and other drought tolerance traits in JG11, an elite and leading variety of chickpea . The Plant Genome 6 , 1 – 9 . Varshney RK , Glaszmann JC , Leung H , Ribaut JM . 2010 . More genomic resources for less-studied crops . Trends in Biotechnology 28 , 452 – 460 . Google Scholar CrossRef Search ADS Varshney RK , Graner A , Sorrells ME . 2005 . Genomics-assisted breeding for crop improvement . Trends in Plant Science 10 , 621 – 630 . Google Scholar CrossRef Search ADS Varshney RK , Hoisington DA , Tyagi AK . 2006 . Advances in cereal genomics and applications in crop breeding . Trends in Biotechnology 24 , 490 – 499 . Google Scholar CrossRef Search ADS Varshney RK , Kudapa H , Pazhamala L et al. 2015 . Translational genomics in agriculture: some examples in grain legumes . Critical Reviews in Plant Sciences 34 , 169 – 194 . Google Scholar CrossRef Search ADS Varshney RK , Mohan SM , Gaur PM et al. 2013b. Marker-assisted backcrossing to introgress resistance to Fusarium Wilt Race 1 and Ascochyta blight in C 214, an elite cultivar of chickpea . The Plant Genome 7 , 1 – 11 . Varshney RK , Pandey MK , Janila P , Nigam SN , Sudini H , Gowda MV , Sriswathi M , Radhakrishnan T , Manohar SS , Nagesh P . 2014 . Marker-assisted introgression of a QTL region to improve rust resistance in three elite and popular varieties of peanut (Arachis hypogaea L.) . Theoretical and Applied Genetics 127 , 1771 – 1781 . Google Scholar CrossRef Search ADS Varshney RK , Ribaut JM , Buckler ES , Tuberosa R , Rafalski JA , Langridge P . 2012b. Can genomics boost productivity of orphan crops ? Nature Biotechnology 30 , 1172 – 1176 . Google Scholar CrossRef Search ADS Varshney RK , Roorkiwal M , Nguyen HT . 2013c. Legume genomics: from genomic resources to molecular breeding . The Plant Genome 6 , 1 – 7 . Varshney RK , Saxena RK , Upadhyaya HD et al. 2017 . Whole-genome resequencing of 292 pigeonpea accessions identifies genomic regions associated with domestication and agronomic traits . Nature Genetics 49 , 1082 – 1088 . Google Scholar CrossRef Search ADS Varshney RK , Singh VK , Hickey J , Xun X , Marshall DF , Wang J , Edwards D , Ribaut J . 2016 . Analytical and decision support tools for genomics-assisted breeding . Trends in Plant Science 15 , S1360 – S1385 . Varshney RK , Song C , Saxena RK et al. 2013d. Draft genome sequence of chickpea (Cicer arietinum) provides a resource for trait improvement . Nature Biotechnology 31 , 240 – 246 . Google Scholar CrossRef Search ADS Visscher PM , Brown MA , McCarthy MI , Yang J . 2012 . Five years of GWAS discovery . American Journal of Human Genetics 90 , 7 – 24 . Google Scholar CrossRef Search ADS Waliyar F , Kumar PL , Ntare BR , Diallo AT . 2007 . A century of research on groundnut rosette disease and its management. Information Bulletin no. 75 . Patancheru, Andhra Pradesh, India : International Crops Research Institute for the Semi-Arid Tropics . Wang L , Wang L , Tan Q , Fan Q , Zhu H , Hong Z , Zhang Z , Duanmu D . 2016 . Efficient inactivation of symbiotic nitrogen fixation related genes in Lotus japonicus using CRISPR-Cas9 . Frontiers in Plant Science 7 , 1333 . Wang LL , Wang LX , Zhou Y , Duanmu D . 2017 . Use of CRISPR/Cas9 for symbiotic nitrogen fixation research in legumes . Progress in Molecular Biology and Translational Science 150 , 1 – 495 . Google Scholar CrossRef Search ADS Wasson AP , Rebetzke GJ , Kirkegaard JA , Christopher J , Richards RA , Watt M . 2014 . Soil coring at multiple field environments can directly quantify variation in deep root traits to select wheat genotypes for breeding . Journal of Experimental Botany 65 , 6231 – 6249 . Google Scholar CrossRef Search ADS Wojciechowski MF , Lavin M , Sanderson MJ . 2004 . A phylogeny of legumes (Leguminosae) based on analysis of the plastid matK gene resolves many well-supported subclades within the family . American Journal of Botany 91 , 1846 – 1862 . Google Scholar CrossRef Search ADS Wray NR , Yang J , Hayes BJ , Price AL , Goddard ME , Visscher PM . 2013 . Pitfalls of predicting complex traits from SNPs . Nature Reviews Genetics 14 , 507 – 515 . Google Scholar CrossRef Search ADS Yakubu H , Buji IB , Sandabe MK . 2011 . Effects of seed-dressing fungicides on germination, nodulation, N2-fixation and yields of two groundnut varieties in semi-arid region of Nigeria . Sandabe International Journal of Applied Agricultural Research 6 , 121 – 129 . Yang K , Tianb Z , Chenc C et al. 2015 . Genome sequencing of adzuki bean (Vigna angularis) provides insight into high starch and low fat accumulation and domestication . Proceedings of the National Academy of Sciences, USA 112 , 3213 – 18 . Zaman-Allah M , Jenkinson DM , Vadez V . 2011 . A conservative pattern of water use, rather than deep or profuse rooting, is critical for the terminal drought tolerance of chickpea . Journal of Experimental Botany 62 , 4239 – 4252 . Google Scholar CrossRef Search ADS Zhang C , Tar’an B , Warkentin T , Tullu A , Bett KE , Vandenberg B , Somers D . 2006 . Selection for lodging resistance in early generations of field pea by molecular markers . Crop Science 46 , 321 – 329 . Google Scholar CrossRef Search ADS Zhang X , Pérez-Rodríguez P , Burgueño J , Olsen M , Buckler E , Atlin G , Prasanna BM , Vargas M , San Vicente F , Crossa J . 2017 . Rapid cycling genomic selection in a multiparental tropical maize population . G3 7 , 2315 – 2326 . Google Scholar CrossRef Search ADS Zhou Z , Jiang Y , Wang Z et al. 2015a. Resequencing 302 wild and cultivated accessions identifies genes related to domestication and improvement in soybean . Nature Biotechnology 33 , 408 – 414 . Google Scholar CrossRef Search ADS Zhou L , Wang SB , Jian J , et al. 2015b. Identification of domestication-related loci associated with flowering time and seed size in soybean with the RAD-seq genotyping method . Scientific Reports 5 , 9350 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Experimental Botany Oxford University Press

Accelerating genetic gains in legumes for the development of prosperous smallholder agriculture: integrating genomics, phenotyping, systems modelling and agronomy

Loading next page...
 
/lp/ou_press/accelerating-genetic-gains-in-legumes-for-the-development-of-fsKPP4e08p
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
0022-0957
eISSN
1460-2431
D.O.I.
10.1093/jxb/ery088
Publisher site
See Article on Publisher Site

Abstract

Abstract Grain legumes form an important component of the human diet, provide feed for livestock, and replenish soil fertility through biological nitrogen fixation. Globally, the demand for food legumes is increasing as they complement cereals in protein requirements and possess a high percentage of digestible protein. Climate change has enhanced the frequency and intensity of drought stress, posing serious production constraints, especially in rainfed regions where most legumes are produced. Genetic improvement of legumes, like other crops, is mostly based on pedigree and performance-based selection over the past half century. To achieve faster genetic gains in legumes in rainfed conditions, this review proposes the integration of modern genomics approaches, high throughput phenomics, and simulation modelling in support of crop improvement that leads to improved varieties that perform with appropriate agronomy. Selection intensity, generation interval, and improved operational efficiencies in breeding are expected to further enhance the genetic gain in experimental plots. Improved seed access to farmers, combined with appropriate agronomic packages in farmers’ fields, will deliver higher genetic gains. Enhanced genetic gains, including not only productivity but also nutritional and market traits, will increase the profitability of farming and the availability of affordable nutritious food especially in developing countries. Drought stress, genetic gains, genomics-assisted breeding, legumes, rainfed agriculture Introduction Legumes are the third largest group among higher plants with more than 18 000 species in 650 genera. Grain legumes belong to subfamily Papilionoideae. Economically important food and feed legumes mostly fall in four clades of Papilionoideae: (i) genistoids, (ii) aeschynomenoids/dalbergioids, (iii) Hologalegina, and (iv) phaseoloids/millettioids. Recent molecular studies placed cool-season legumes in the inverted repeat loss clade of Hologalegina, while warm season legumes were included in the phaseoloid/millettioid clade (Wojciechowski et al., 2004). Cool season legumes include chickpea (Cicer arietinum), faba bean (Vicia faba), lentil (Lens culinaris), and pea (Pisum sativum). Warm season legumes include common bean (Phaseolus vulgaris), cowpea (Vigna unguiculata), pigeonpea (Cajanus cajan), soybean (Glycine max), and groundnut (Arachis hypogaea) (Doyle and Luckow, 2003). Signature features of legumes include biological nitrogen fixation in symbiotic association with bacteria and mycorrhiza, geotropic peg and pod development in groundnut, and vernalization in chickpea, each of which is of economic and scientific interest. Legumes form an important component of the human diet, provide animal feed, and replenish soil fertility through biological nitrogen fixation. In addition, legumes can recover unavailable forms of soil phosphorus, possibly lower the emission of greenhouse gases, and are prospective assets in future cropping systems (Stagnari et al., 2017). A recent review gives a good summary of the many virtues of food legumes and sets an agenda for their improvement in the decades to come (Sinclair and Vadez, 2012; Foyer et al., 2016). According to United States Census Bureau International Database, the global population is predicted to approach 9.8 billion by 2050. Further, the Global Alliance for Improved Nutrition, which works to strengthen the availability of nutritious and affordable foods for infants and women, estimated that 2 billion people across the world are undernourished and 2.6 million children die due to malnutrition (http://www.gainhealth.org/knowledge-centre/fast-facts-malnutrition/). The majority of crop improvement programs, however, have emphasized only crop productivity traits in the past including during the green revolution. As a result, despite the attaining of global food security, a continuous increase in micronutrient malnutrition has been recorded in most of the developing countries. Under such a scenario, the grain legumes are relevant in the fight against malnutrition and are considered to provide ‘nutritious seeds for a sustainable future’. Therefore, a comprehensive strategy and dedicated effort are required to produce more and dense nutritious legume crops to ensure food and nutritional security. Climate change has significant adverse impacts on all components of crop production—area, intensity, and yield— and hence producing more food to feed the growing population is a great challenge before agriculturists and other stakeholders. The World Resources Institute predicted that by 2040, the USA, China, and India are expected to face 40–70% more water stress. Globally the demand for food legumes is increasing as legumes may complement cereals in protein requirements and possess a high percentage of digestible protein. The majority of the area under legumes is in Asia (~76%), principally soybean, beans, groundnut, and chickpea. More recent studies indicate that between 1961 and 2014, 75% of the legume area in South America and 90% of that in North America are under soybean cultivation (FAOSTAT, 2016). In Sub-Saharan Africa, food legumes are cultivated on 20 million ha, and 44% of production comes from cowpea and 31% from dry beans. Sub-Saharan Africa alone contributes 54% of global cowpea production (Akibode, 2011). High tolerance to environmental stresses of many rainfed legumes, when compared with soybean, is required for sustainable production in often harsh soil and climatic conditions in dryland agriculture (Foyer et al., 2016; Sita et al., 2017). Erratic rainfall increases the occurrence of drought stress necessitating drought-tolerant legume varieties that can produce in water-limited environments. Drought tolerance is a very complex trait and hence an in-depth understanding of the genomic control and molecular mechanism of drought tolerance is essential for effective deployment in breeding. Genetic gain can be defined as the rate of increase in yield over a given time period—pure genetic gain is estimated against potential yield, but can also be assessed under defined stress conditions. Barker et al. (2010) reported a maize genetic gain of 211 kg ha−1 year−1 in favorable environments in Woodland, CA, USA. Further, the same study reported lower genetic gains under water limited environments, of 124 and 91 kg ha−1 year−1 expressed during flowering and mid-grain filling, respectively. Recently, by using marker-assisted recurrent selection (MARS) in maize, a 3% increase in yield per cycle under water stress conditions was reported (Bankole et al., 2017). Among legumes, Koester et al., (2014) reported a genetic gain of 26.5 kg ha–1 year–1 by studying the historical data of 80 years of soybean breeding at the Crop Research and Education Center in Urbana (IL, USA). This gain in grain yield is attributed to increases in light interception, energy conversion, and partitioning efficiencies. A contemporary question in legume improvement is how to enhance the synergy among the different disciplines of genomics, phenomics, crop physiology, growth modelling, and agronomy. In this article, we discuss the technological advances in different disciplines and advocate their integrated use both to understand the genetics of traits and to deploy them in crop improvement programs of soybean, chickpea, pigeonpea, and groundnut. A special emphasis has been placed on deployment of genomic selection in breeding programs to attaining faster genetic gain. Technological advances Agricultural research has benefitted from new technologies that have helped in achieving current food sufficiency in many parts of the world. However, to realize future demands, researchers need to apply new tools, technologies, and partnerships to achieve even higher productivity and profitability of farming and better nutrition for all consumers. Emerging crop improvement technologies are outlined in the following sections. Sequencing and genotyping During the past two decades, advances in molecular marker technologies and next-generation sequencing technologies have enhanced our understanding of several traits in crop plants including food legumes (Varshney et al., 2015). Genome sequences for most of the grain legumes are now available, for instance soybean (Schmutz et al., 2010), groundnut progenitors (Bertioli et al., 2016; Chen et al., 2016), chickpea (Varshney et al., 2013d; Parween et al., 2015), pigeonpea (Varshney et al., 2012a), common bean (Schmutz et al., 2014; Yang et al., 2015), and adzuki bean (Kang et al., 2015; Yang et al., 2015) (Table 1). Efforts are underway to sequence the remaining legume genomes. The past decade has witnessed an exponential increase in availability of genomic resources and their deployment in trait discovery and breeding (see Bohra et al., 2014; Varshney et al., 2015; Pandey et al., 2016). As a result, several of these legume crops have genomic resources and associated phenotypic data to support genomics-based discovery and breeding approaches to develop superior legume varieties. Table 1. Summary of genome sequence, resequencing efforts and available high throughput genotyping platforms in important legumes Crop Genome sequence Germplasm lines resequenced High throughput genotyping platforms Soybean • 85% of Glycine max var William 82 genome (1115 Mb); 46 430 protein coding genes (Schmutz et al., 2010) • 915.4 Mb of G. soja var. IT182932 (Kim et al., 2010) • 106 (7 wild, 43 landraces and 56 elite; Valliyodan et al., 2016) • 302 (62 wild, 130 landraces and 110 elite; Zhou et al., 2015a) • 28 (commercial cultivars; Maldonado dos Santos et al., 2016) • 89 lines (Lee et al., 2015) • 286 (14 wild, 153 landraces and 119 elite; Zhou et al., 2015b) • Illumina 384 SNP VeraCode assays (Lee et al., 2015) • NJAU 355K SoySNP array (Wang et al., 2016) • Illumina Infinium SoySNP6K BeadChip (Akond et al., 2013) • SoySNP50K array (Song et al., 2013) • 384 SNP GoldenGate assay (Hyten et al., 2008) • SoyaSNP180K Axiom (Lee et al., 2015) Pigeonpea • 72.7% of Cajanus cajan var Asha genome (833.07 Mb); 48 680 protein coding genes (Varshney et al., 2012a) • 292 lines (Varshney et al., 2017) • 20 (crossing parentals of recombinant inbred lines, introgression lines, MAGIC and NAM population; Kumar et al., 2016) • 60K Axiom®Cajanus SNP array (Saxena et al., 2017 and unpublished) Chickpea • 72.12% of Cicer arietinum var CDC Frontier genome (738 Mb); 28 269 protein coding genes (Varshney et al., 2013d) • 416 Mb of C. reticulatum var PI 489777 genome; 25 680 protein coding genes (Gupta et al., 2017) • 35 (parental genotypes of mapping populations; Thudi et al., 2016a) • 129 released varieties (Thudi et al., 2016b) • 300 lines (ICRISAT, unpublished) • 3000 lines (ICRISAT, unpublished) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) • 60K Axiom®Cicer SNP array (Roorkiwal et al., 2017) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) Groundnut • 1081 Mb of Arachis duranensis V14167 and 1371 Mb of A. ipaensis K30076 (Bertioli et al., 2016) • A. duranensis var PI475845; 50 324 protein coding genes (Chen et al., 2016) • 11 genotypes including synthetics and their diploid parents (Chen et al., 2016) • 41 diverse genotypes (30 tetraploids and 11 diploids) (Clevenger et al., 2017; Pandey et al., 2017a) • 58K Axiom®Arachis SNP array (Pandey et al., 2017a) •1536 SNP GoldenGate assay (Nagy et al., 2012) Common bean • 80.57% of Phaseolus vulgaris var G19833 genome (587 Mb); 26 279 protein coding genes (Schmutz et al., 2014) • 17 varieties (Song et al., 2015) • BARCBean6K_1, BARCBean6K_2 chips, BARCBean6K_3 SNP chips (Song et al., 2013, 2015) Mung bean • 84.48% of Vigna radiata var radiata VC1973A geome (548 Mb); 22 427 predicted genes (Kang et al., 2014) • 84.48% of Vigna radiata var sublobata TC1966A geome (501 Mb); 22 834 predicted genes (Kang et al., 2014) • 81.81% of Vigna radiata var glabra V1160 geome (968 Mb); 41 484 predicted genes (Kang et al., 2014) — — Adzuki bean • 96.56% of Vigna angularis var Gyeongwon genome (612 Mb); 21 532 predicted genes (Kang et al., 2015) • 83.02% of Vigna angularis var Jingnong geome (542 Mb); 34 183 predicted genes (Yang et al., 2015) — — Cowpea • 323 Mb of Phaseolus vulgaris var IT97K genome (724 Mb) (Muñoz-Amatriaín et al., 2017) • 36 diverse accessions (Muñoz-Amatriaín et al., 2017) • Cowpea iSelect Consortium Array (Muñoz-Amatriaín et al., 2017) • 1536 SNP GoldenGate assays (Muchero et al., 2009) Pea https://www.france-genomique.org/spip/spip. php?article141&lang=fr — • GenoPea 13.2K SNP Array (Tayeh et al., 2015a) • 384 SNP GoldenGate assays (Deulvot et al., 2010) Crop Genome sequence Germplasm lines resequenced High throughput genotyping platforms Soybean • 85% of Glycine max var William 82 genome (1115 Mb); 46 430 protein coding genes (Schmutz et al., 2010) • 915.4 Mb of G. soja var. IT182932 (Kim et al., 2010) • 106 (7 wild, 43 landraces and 56 elite; Valliyodan et al., 2016) • 302 (62 wild, 130 landraces and 110 elite; Zhou et al., 2015a) • 28 (commercial cultivars; Maldonado dos Santos et al., 2016) • 89 lines (Lee et al., 2015) • 286 (14 wild, 153 landraces and 119 elite; Zhou et al., 2015b) • Illumina 384 SNP VeraCode assays (Lee et al., 2015) • NJAU 355K SoySNP array (Wang et al., 2016) • Illumina Infinium SoySNP6K BeadChip (Akond et al., 2013) • SoySNP50K array (Song et al., 2013) • 384 SNP GoldenGate assay (Hyten et al., 2008) • SoyaSNP180K Axiom (Lee et al., 2015) Pigeonpea • 72.7% of Cajanus cajan var Asha genome (833.07 Mb); 48 680 protein coding genes (Varshney et al., 2012a) • 292 lines (Varshney et al., 2017) • 20 (crossing parentals of recombinant inbred lines, introgression lines, MAGIC and NAM population; Kumar et al., 2016) • 60K Axiom®Cajanus SNP array (Saxena et al., 2017 and unpublished) Chickpea • 72.12% of Cicer arietinum var CDC Frontier genome (738 Mb); 28 269 protein coding genes (Varshney et al., 2013d) • 416 Mb of C. reticulatum var PI 489777 genome; 25 680 protein coding genes (Gupta et al., 2017) • 35 (parental genotypes of mapping populations; Thudi et al., 2016a) • 129 released varieties (Thudi et al., 2016b) • 300 lines (ICRISAT, unpublished) • 3000 lines (ICRISAT, unpublished) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) • 60K Axiom®Cicer SNP array (Roorkiwal et al., 2017) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) Groundnut • 1081 Mb of Arachis duranensis V14167 and 1371 Mb of A. ipaensis K30076 (Bertioli et al., 2016) • A. duranensis var PI475845; 50 324 protein coding genes (Chen et al., 2016) • 11 genotypes including synthetics and their diploid parents (Chen et al., 2016) • 41 diverse genotypes (30 tetraploids and 11 diploids) (Clevenger et al., 2017; Pandey et al., 2017a) • 58K Axiom®Arachis SNP array (Pandey et al., 2017a) •1536 SNP GoldenGate assay (Nagy et al., 2012) Common bean • 80.57% of Phaseolus vulgaris var G19833 genome (587 Mb); 26 279 protein coding genes (Schmutz et al., 2014) • 17 varieties (Song et al., 2015) • BARCBean6K_1, BARCBean6K_2 chips, BARCBean6K_3 SNP chips (Song et al., 2013, 2015) Mung bean • 84.48% of Vigna radiata var radiata VC1973A geome (548 Mb); 22 427 predicted genes (Kang et al., 2014) • 84.48% of Vigna radiata var sublobata TC1966A geome (501 Mb); 22 834 predicted genes (Kang et al., 2014) • 81.81% of Vigna radiata var glabra V1160 geome (968 Mb); 41 484 predicted genes (Kang et al., 2014) — — Adzuki bean • 96.56% of Vigna angularis var Gyeongwon genome (612 Mb); 21 532 predicted genes (Kang et al., 2015) • 83.02% of Vigna angularis var Jingnong geome (542 Mb); 34 183 predicted genes (Yang et al., 2015) — — Cowpea • 323 Mb of Phaseolus vulgaris var IT97K genome (724 Mb) (Muñoz-Amatriaín et al., 2017) • 36 diverse accessions (Muñoz-Amatriaín et al., 2017) • Cowpea iSelect Consortium Array (Muñoz-Amatriaín et al., 2017) • 1536 SNP GoldenGate assays (Muchero et al., 2009) Pea https://www.france-genomique.org/spip/spip. php?article141&lang=fr — • GenoPea 13.2K SNP Array (Tayeh et al., 2015a) • 384 SNP GoldenGate assays (Deulvot et al., 2010) View Large Table 1. Summary of genome sequence, resequencing efforts and available high throughput genotyping platforms in important legumes Crop Genome sequence Germplasm lines resequenced High throughput genotyping platforms Soybean • 85% of Glycine max var William 82 genome (1115 Mb); 46 430 protein coding genes (Schmutz et al., 2010) • 915.4 Mb of G. soja var. IT182932 (Kim et al., 2010) • 106 (7 wild, 43 landraces and 56 elite; Valliyodan et al., 2016) • 302 (62 wild, 130 landraces and 110 elite; Zhou et al., 2015a) • 28 (commercial cultivars; Maldonado dos Santos et al., 2016) • 89 lines (Lee et al., 2015) • 286 (14 wild, 153 landraces and 119 elite; Zhou et al., 2015b) • Illumina 384 SNP VeraCode assays (Lee et al., 2015) • NJAU 355K SoySNP array (Wang et al., 2016) • Illumina Infinium SoySNP6K BeadChip (Akond et al., 2013) • SoySNP50K array (Song et al., 2013) • 384 SNP GoldenGate assay (Hyten et al., 2008) • SoyaSNP180K Axiom (Lee et al., 2015) Pigeonpea • 72.7% of Cajanus cajan var Asha genome (833.07 Mb); 48 680 protein coding genes (Varshney et al., 2012a) • 292 lines (Varshney et al., 2017) • 20 (crossing parentals of recombinant inbred lines, introgression lines, MAGIC and NAM population; Kumar et al., 2016) • 60K Axiom®Cajanus SNP array (Saxena et al., 2017 and unpublished) Chickpea • 72.12% of Cicer arietinum var CDC Frontier genome (738 Mb); 28 269 protein coding genes (Varshney et al., 2013d) • 416 Mb of C. reticulatum var PI 489777 genome; 25 680 protein coding genes (Gupta et al., 2017) • 35 (parental genotypes of mapping populations; Thudi et al., 2016a) • 129 released varieties (Thudi et al., 2016b) • 300 lines (ICRISAT, unpublished) • 3000 lines (ICRISAT, unpublished) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) • 60K Axiom®Cicer SNP array (Roorkiwal et al., 2017) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) Groundnut • 1081 Mb of Arachis duranensis V14167 and 1371 Mb of A. ipaensis K30076 (Bertioli et al., 2016) • A. duranensis var PI475845; 50 324 protein coding genes (Chen et al., 2016) • 11 genotypes including synthetics and their diploid parents (Chen et al., 2016) • 41 diverse genotypes (30 tetraploids and 11 diploids) (Clevenger et al., 2017; Pandey et al., 2017a) • 58K Axiom®Arachis SNP array (Pandey et al., 2017a) •1536 SNP GoldenGate assay (Nagy et al., 2012) Common bean • 80.57% of Phaseolus vulgaris var G19833 genome (587 Mb); 26 279 protein coding genes (Schmutz et al., 2014) • 17 varieties (Song et al., 2015) • BARCBean6K_1, BARCBean6K_2 chips, BARCBean6K_3 SNP chips (Song et al., 2013, 2015) Mung bean • 84.48% of Vigna radiata var radiata VC1973A geome (548 Mb); 22 427 predicted genes (Kang et al., 2014) • 84.48% of Vigna radiata var sublobata TC1966A geome (501 Mb); 22 834 predicted genes (Kang et al., 2014) • 81.81% of Vigna radiata var glabra V1160 geome (968 Mb); 41 484 predicted genes (Kang et al., 2014) — — Adzuki bean • 96.56% of Vigna angularis var Gyeongwon genome (612 Mb); 21 532 predicted genes (Kang et al., 2015) • 83.02% of Vigna angularis var Jingnong geome (542 Mb); 34 183 predicted genes (Yang et al., 2015) — — Cowpea • 323 Mb of Phaseolus vulgaris var IT97K genome (724 Mb) (Muñoz-Amatriaín et al., 2017) • 36 diverse accessions (Muñoz-Amatriaín et al., 2017) • Cowpea iSelect Consortium Array (Muñoz-Amatriaín et al., 2017) • 1536 SNP GoldenGate assays (Muchero et al., 2009) Pea https://www.france-genomique.org/spip/spip. php?article141&lang=fr — • GenoPea 13.2K SNP Array (Tayeh et al., 2015a) • 384 SNP GoldenGate assays (Deulvot et al., 2010) Crop Genome sequence Germplasm lines resequenced High throughput genotyping platforms Soybean • 85% of Glycine max var William 82 genome (1115 Mb); 46 430 protein coding genes (Schmutz et al., 2010) • 915.4 Mb of G. soja var. IT182932 (Kim et al., 2010) • 106 (7 wild, 43 landraces and 56 elite; Valliyodan et al., 2016) • 302 (62 wild, 130 landraces and 110 elite; Zhou et al., 2015a) • 28 (commercial cultivars; Maldonado dos Santos et al., 2016) • 89 lines (Lee et al., 2015) • 286 (14 wild, 153 landraces and 119 elite; Zhou et al., 2015b) • Illumina 384 SNP VeraCode assays (Lee et al., 2015) • NJAU 355K SoySNP array (Wang et al., 2016) • Illumina Infinium SoySNP6K BeadChip (Akond et al., 2013) • SoySNP50K array (Song et al., 2013) • 384 SNP GoldenGate assay (Hyten et al., 2008) • SoyaSNP180K Axiom (Lee et al., 2015) Pigeonpea • 72.7% of Cajanus cajan var Asha genome (833.07 Mb); 48 680 protein coding genes (Varshney et al., 2012a) • 292 lines (Varshney et al., 2017) • 20 (crossing parentals of recombinant inbred lines, introgression lines, MAGIC and NAM population; Kumar et al., 2016) • 60K Axiom®Cajanus SNP array (Saxena et al., 2017 and unpublished) Chickpea • 72.12% of Cicer arietinum var CDC Frontier genome (738 Mb); 28 269 protein coding genes (Varshney et al., 2013d) • 416 Mb of C. reticulatum var PI 489777 genome; 25 680 protein coding genes (Gupta et al., 2017) • 35 (parental genotypes of mapping populations; Thudi et al., 2016a) • 129 released varieties (Thudi et al., 2016b) • 300 lines (ICRISAT, unpublished) • 3000 lines (ICRISAT, unpublished) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) • 60K Axiom®Cicer SNP array (Roorkiwal et al., 2017) • GoldenGate assays based on VeraCode technology (Roorkiwal et al., 2013) Groundnut • 1081 Mb of Arachis duranensis V14167 and 1371 Mb of A. ipaensis K30076 (Bertioli et al., 2016) • A. duranensis var PI475845; 50 324 protein coding genes (Chen et al., 2016) • 11 genotypes including synthetics and their diploid parents (Chen et al., 2016) • 41 diverse genotypes (30 tetraploids and 11 diploids) (Clevenger et al., 2017; Pandey et al., 2017a) • 58K Axiom®Arachis SNP array (Pandey et al., 2017a) •1536 SNP GoldenGate assay (Nagy et al., 2012) Common bean • 80.57% of Phaseolus vulgaris var G19833 genome (587 Mb); 26 279 protein coding genes (Schmutz et al., 2014) • 17 varieties (Song et al., 2015) • BARCBean6K_1, BARCBean6K_2 chips, BARCBean6K_3 SNP chips (Song et al., 2013, 2015) Mung bean • 84.48% of Vigna radiata var radiata VC1973A geome (548 Mb); 22 427 predicted genes (Kang et al., 2014) • 84.48% of Vigna radiata var sublobata TC1966A geome (501 Mb); 22 834 predicted genes (Kang et al., 2014) • 81.81% of Vigna radiata var glabra V1160 geome (968 Mb); 41 484 predicted genes (Kang et al., 2014) — — Adzuki bean • 96.56% of Vigna angularis var Gyeongwon genome (612 Mb); 21 532 predicted genes (Kang et al., 2015) • 83.02% of Vigna angularis var Jingnong geome (542 Mb); 34 183 predicted genes (Yang et al., 2015) — — Cowpea • 323 Mb of Phaseolus vulgaris var IT97K genome (724 Mb) (Muñoz-Amatriaín et al., 2017) • 36 diverse accessions (Muñoz-Amatriaín et al., 2017) • Cowpea iSelect Consortium Array (Muñoz-Amatriaín et al., 2017) • 1536 SNP GoldenGate assays (Muchero et al., 2009) Pea https://www.france-genomique.org/spip/spip. php?article141&lang=fr — • GenoPea 13.2K SNP Array (Tayeh et al., 2015a) • 384 SNP GoldenGate assays (Deulvot et al., 2010) View Large Besides disclosing the blueprint of genomes of several legumes, different research groups also re-sequenced legume germplasm lines, which further increased our understanding of genome architecture, structural variations, genome evolution, and genome dynamics during domestication. Several million structural variations that aid in trait mapping and trait improvement were reported (Kumar et al., 2016; Thudi et al., 2016b). In addition, genome-wide single nucleotide polymorphisms (SNPs) were also used to identify significant marker trait associations for economically important traits (Zhou et al., 2015a; Varshney et al., 2017). Resequencing germplasm lines also enabled us to understand the spatial and temporal trends in diversity in released varieties of chickpea (Thudi et al., 2016a), cultivated and wild accessions of soybean (Lam et al., 2010; Zhou et al., 2015a), and a reference set of pigeonpea (Varshney et al., 2017; Table 1). Resequencing of 28 Brazilian soybean cultivars suggested that, despite the diversification of modern Brazilian cultivars, the soybean germplasm remains very narrow because of the large number of genome regions that exhibit low diversity (Maldonado dos Santos et al., 2016). In recent years, genotyping by sequencing, skim sequencing, diversity array technology (DArT)-seq and restriction site associated DNA sequencing approaches were also employed for developing high density genetic maps, refining the QTL mapping and identifying trait linked markers in legumes (Jaganathan et al., 2015; Kale et al., 2015; Contreras-Soto et al., 2017; Leamy et al., 2017; Valdisser et al., 2017). Nevertheless, the percentage of missing data points and the SNP calling rates greatly reduce the number of final SNPs for different studies and this has been a concern. To overcome these constraints several high-throughput SNP genotyping platforms, such as Veracode assays, Illumina GoldenGate assays, Infinium chips and Axiom arrays, are now available, which not only increase the precision of SNP calling but also enable genotyping of larger populations at reduced cost (Table 1). Axiom arrays with >50K SNPs with uniform genome coverage were developed and are being used for germplasm characterization, trait mapping and molecular breeding in chickpea (Roorkiwal et al., 2017), pigeonpea (ICRISAT unpublished data), groundnut (Pandey et al., 2017a) and soybean (Lee et al., 2015; Table 1). High-density and precise phenotyping Understanding the complex relationships between genotype and phenotype is a major challenge in plant sciences in view of the multi-scale nature of phenotypes and of the large phenotypic plasticity displayed by plants subjected to varying environmental conditions that results in different values of phenotypic variables and different rankings of genotypes in different experiments. It requires (i) capturing information on physiological traits and performance of large numbers of plants, together with their environment, (ii) analysing and organizing the resulting datasets, and (iii) developing models able to disentangle and simulate plant behavior in a range of scenarios (Tardieu et al., 2017). Good and relevant phenotyping starts with a detailed understanding of plant biological processes underlying tolerance/resistance to well-defined constraints. For instance, tolerance of chickpea to terminal drought in South Asia depends on plant traits that ensure the availability of water during the grain filling phase (Zaman-Allah et al., 2011; Vadez et al., 2013a; Vadez 2014). Therefore, phenotyping first needs to be well targeted and defined. Precise phenotyping is the next step and involves the design of protocols and approaches that ensure the assessment of key traits at the highest rate and the cheapest cost. The cost of phenotyping increases with increase in population size, especially in the case of deployment of genomic selection (see later), which requires a larger population for high prediction accuracy. The availability of non-destructive root phenotyping methods has, in the past, limited our understanding of traits involved in drought tolerance of legumes such as stomatal control or root system architecture. Hence, genome-wide association studies of traits associated with drought tolerance, based on large-scale germplasm re-sequencing (Varshney et al., 2017), require a suite of high-throughput and precise phenotyping technologies for capturing, analysing, and interpreting images in order to infer heritable traits that can be analysed genetically (Tardieu et al., 2017). Recognizing the importance of phenotyping, in recent years several large public, private and academic efforts have led to the establishment of state-of-the-art phenotyping facilities word-wide in an exponential phase of automated high-throughput phenotyping (Table 2). These include the German and French Plant Phenotyping Networks (http://www.dppn.de; https://www.phenome-fppn.fr/), European Plant Phenotyping Network (2020) (EPPN2020; https://eppn2020.plant-phenotyping.eu), European Plant Phenotyping Infrastructure (EMPHASIS), and International Plant Phenotyping Network (IPPN; https://www.plant-phenotyping.org/). These infrastructure networks enable access to the necessary tools for phenotyping, in particular robot-assisted image capture (Cooper et al., 2009; Fiorani and Schurr, 2013), statistical designs and models for extracting relevant physiological variables from raw data (Cabrera-Bosquet et al., 2016), and specialized information systems managing large datasets originating from phenotyping experiments (Tardieu et al., 2017). GrowScreen-PaGe is a non-invasive, high-throughput phenotyping system developed at the Institute of Biosciences and Geosciences suitable for investigating root systems and root plasticity of large sets of both dicots and monocots (Gioia et al., 2016). The LeasyScan platform developed at the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) offers a novel 3D scanning technique that capture leaf area development continuously and phenotype traits controlling plant water budget (Vadez et al., 2015). The University of Western Australia, Australia has developed a semi-hydroponic phenotyping system that examines architecture of deep root system and is amenable for phenotyping large sets of plants in a relatively small space (Chen et al., 2011). The salient features of several high throughput phenotyping platforms, such as PHENOPSIS (Bresson et al., 2015; Granier et al., 2006), PhenoArch (Cabrera-Bosquet et al., 2016), DEEPER PHENOSCOPE (Bresson et al., 2015), and RADIX (Le Marié et al., 2016), are summarized in Table 2. Table 2. A list of key high-throughput phenotyping platforms Phenotyping platform (and institute or company) Salient features Reference LeasyScan phenotyping platform (ICRISAT, India) • A novel 3D scanning technique to capture leaf area development continuously • Scanner-to-plant concept to increase imaging throughput and analytical scales to combine gravimetric transpiration measurements. • Combines 3D imaging and lysimetry for high- throughput phenotyping of traits controlling plant water budget Vadez et al. (2015) Semi-hydroponic phenotyping system (The University of Western Australia, Australia) • Permits mapping and digital measurement of dynamic growth of taproot and lateral roots • Desirable tool for examining root architecture of deep root systems and large sets of plants in a relatively small space Chen et al. (2011) DEEPER: an integrated phenotyping platform for deeper roots (PennState Colleage of Agriculture, USA) • A platform for identifying the traits of deeper rooted crops in non-destructive field phenotyping of rooting depth, root modeling, high-throughput 3D imaging of root architecture, and anatomy http://plantscience.psu.edu/research/labs/roots/projects/ deeper-an-integrated-phenotyping-platform-for-deeper-roots GLO-Roots: luminescence-based imaging system (Dinnenylab, USA) • Combines custom-made growth vessels and new image analysis algorithms to non-destructively monitor RSA development over space (2D) and time • Allows information on soil properties (e.g. moisture) to be integrated with root growth data • Makes use of luminescence imaging of roots expressing plant codon-optimize luciferase Rellán-Álvarez et al. (2015) X-Ray computed tomography (University of Nottingham, UK) • Non-destructively visualizes opaque root structures by measuring the attenuation of ionizing radiation as it passes through the root • A series of projections are acquired and combined to reconstruct a 3D image of the root system Mairhofer et al. (2012), Mooney et al. (2012) Rhizophonics (University of Liège, Belgium) • Combines hydroponics and rhizotrons • System is made of a nylon fabric supported by an aluminum frame • The set-up is immersed in a tank filled with liquid medium • Allows non-destructive, 2D imaging of root architecture while simultaneously sampling shoots Mathieu et al. (2015) Clear pot method (The University of Queensland, Australia) • Uses transparent pots filled with soil or other p otting media • Seeds are planted close to the pot wall to enable high- throughput imaging of roots along the clear pot wall • To prevent light exposure, the clear pot is placed in black pots while roots are developing Richard et al. (2015) Rhizoslides (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • The set-up consists of a Plexiglas sheet covered with moistened germination paper. Seeds are planted on the slit of the Plexiglas • Allows separation of crown roots from embryonic roots Le Marié et al. (2014) Shovelomics (PennState Colleage of Agriculture, USA) • Involves manual excavation of plants and separating roots from shoots • Washed roots are then placed on a phenotyping board for root trait quantification • New algorithms allow extraction of several root traits in a high throughput manner Trachsel et al. (2011), Bucksch et al. (2014) Soil coring (CSIRO and The University of Queensland, Australia) • Uses a tractor-mounted, hydraulic soil corer to drive steel alloy sampling tubes into the soil • When combined with novel planting configurations (e.g. hill plots), this method allows for phenotyping deep rooted crop varieties Wasson et al. (2014) Rhizo-lysimetry (E. H. Graham Centre for Agricultural Innovation, Australia) • Elaborate facility consisting of an underground corridor and concrete silos and pipes to house soil-containing soil cores for direct root observation Eberbach et al. (2013) Minirhizotrons • A transparent observation tube permanently inserted in the soil. Images of roots growing along the minirhizotron wall at particular locations in the soil profile can be captured over time Iversen et al. (2011), Maeght et al. (2013) RhizoTube (INRA, France) • Cylindrical rhizotrons that allow full 2D visualization of the root system of a single or up to six plants simultaneously • The RhizoCab is designed to take images of the entire root systems of plants growing in RhizoTubes, and also permits a focus on some parts of the root systems Jeudy et al. (2016) RADIX (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • Rhizoslide platform allowing high throughput digital image analysis of root system expansion Le Marié et al. (2016) Scanner bank (The James Hutton Institute, UK) • Low-cost, high-resolution root phenotyping platform, requiring no sophisticated equipment and adaptable to most laboratory and glasshouse environments, and applied to quantify environmental and temporal variation in root traits Adu et al. (2014) The Plant Accelerator® (The University of Adelaide, Australia) • The facility offers modern plant growth environments and state-of-the-art high-throughput automated imaging and computing technologies to monitor the performance of plants under different environmental conditions (e.g. which genotype performs best under drought stress) • Research projects facilitated by this technology vary from large scale screening of early growth, salinity tolerance to water and nutrient use efficiency http://www.plantphenomics.org.au/services/accelerator/ DroughtSpotter (The University of Adelaide, Australia) • A gravimetric platform with precision irrigation to assess transpiration dynamics of plants with a precision of up to 1 g • Integrated irrigation units allow precise and reproducible water application for drought stress or related experiments that require an accurate control of water volume to 1 ml http://www.plantphenomics.org.au/services/droughtspotter/ Crop Plant Root Module (The University of Adelaide, Australia) • Comprises destructive and non-destructive measurement of root growth, architecture, morphology, and water uptake in soil in controlled environments and in the field http://www.plantphenomics.org.au/services/croproot/ Jülich Plant Phenotyping Centre (Jülich, Germany) • Reproducibly quantifies growth and architecture of roots • Elucidates dynamic establishment of roots in space and time • Determines interaction of root responses with above ground plant part http://www.fz-juelich.de/ibg/ibg-2/EN/organisation/JPPC/ JPPC_node.html PHENOPSIS (INRA, France) • Automated platform for reproducible phenotyping of plant responses to soil water deficit Granier et al. (2006), Bresson et al. (2015) WIWAM (SMO and VIB, Belgium) • Automated phenotyping platform for automated weighing, watering, and imaging of plants and, therefore, strictly controlling the applied watering regime Skirycz et al. (2011), Clauw et al. (2015) PHENOSCOPE (INRA, France) • Automated large-scale phenotyping platform that automatically adjusts watering and is equipped with a zenithal imaging system to monitor rosette size and expansion rate during the vegetative stage, with automatic image analysis allowing manual correction Tisné et al. (2013) LemnaTec (LemnaTec, Germany) • Versatile phenotyping system and analytics platform for diverse temporal responses to water availability Chen et al. (2014), Neumann et al. (2015) GlyPh (self-construction) • Simple and low-cost automatic platform for high throughput measurement of plant water use and growth to assess drought tolerance Pereyra-Irujo et al. (2012) Phenotyping platform (and institute or company) Salient features Reference LeasyScan phenotyping platform (ICRISAT, India) • A novel 3D scanning technique to capture leaf area development continuously • Scanner-to-plant concept to increase imaging throughput and analytical scales to combine gravimetric transpiration measurements. • Combines 3D imaging and lysimetry for high- throughput phenotyping of traits controlling plant water budget Vadez et al. (2015) Semi-hydroponic phenotyping system (The University of Western Australia, Australia) • Permits mapping and digital measurement of dynamic growth of taproot and lateral roots • Desirable tool for examining root architecture of deep root systems and large sets of plants in a relatively small space Chen et al. (2011) DEEPER: an integrated phenotyping platform for deeper roots (PennState Colleage of Agriculture, USA) • A platform for identifying the traits of deeper rooted crops in non-destructive field phenotyping of rooting depth, root modeling, high-throughput 3D imaging of root architecture, and anatomy http://plantscience.psu.edu/research/labs/roots/projects/ deeper-an-integrated-phenotyping-platform-for-deeper-roots GLO-Roots: luminescence-based imaging system (Dinnenylab, USA) • Combines custom-made growth vessels and new image analysis algorithms to non-destructively monitor RSA development over space (2D) and time • Allows information on soil properties (e.g. moisture) to be integrated with root growth data • Makes use of luminescence imaging of roots expressing plant codon-optimize luciferase Rellán-Álvarez et al. (2015) X-Ray computed tomography (University of Nottingham, UK) • Non-destructively visualizes opaque root structures by measuring the attenuation of ionizing radiation as it passes through the root • A series of projections are acquired and combined to reconstruct a 3D image of the root system Mairhofer et al. (2012), Mooney et al. (2012) Rhizophonics (University of Liège, Belgium) • Combines hydroponics and rhizotrons • System is made of a nylon fabric supported by an aluminum frame • The set-up is immersed in a tank filled with liquid medium • Allows non-destructive, 2D imaging of root architecture while simultaneously sampling shoots Mathieu et al. (2015) Clear pot method (The University of Queensland, Australia) • Uses transparent pots filled with soil or other p otting media • Seeds are planted close to the pot wall to enable high- throughput imaging of roots along the clear pot wall • To prevent light exposure, the clear pot is placed in black pots while roots are developing Richard et al. (2015) Rhizoslides (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • The set-up consists of a Plexiglas sheet covered with moistened germination paper. Seeds are planted on the slit of the Plexiglas • Allows separation of crown roots from embryonic roots Le Marié et al. (2014) Shovelomics (PennState Colleage of Agriculture, USA) • Involves manual excavation of plants and separating roots from shoots • Washed roots are then placed on a phenotyping board for root trait quantification • New algorithms allow extraction of several root traits in a high throughput manner Trachsel et al. (2011), Bucksch et al. (2014) Soil coring (CSIRO and The University of Queensland, Australia) • Uses a tractor-mounted, hydraulic soil corer to drive steel alloy sampling tubes into the soil • When combined with novel planting configurations (e.g. hill plots), this method allows for phenotyping deep rooted crop varieties Wasson et al. (2014) Rhizo-lysimetry (E. H. Graham Centre for Agricultural Innovation, Australia) • Elaborate facility consisting of an underground corridor and concrete silos and pipes to house soil-containing soil cores for direct root observation Eberbach et al. (2013) Minirhizotrons • A transparent observation tube permanently inserted in the soil. Images of roots growing along the minirhizotron wall at particular locations in the soil profile can be captured over time Iversen et al. (2011), Maeght et al. (2013) RhizoTube (INRA, France) • Cylindrical rhizotrons that allow full 2D visualization of the root system of a single or up to six plants simultaneously • The RhizoCab is designed to take images of the entire root systems of plants growing in RhizoTubes, and also permits a focus on some parts of the root systems Jeudy et al. (2016) RADIX (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • Rhizoslide platform allowing high throughput digital image analysis of root system expansion Le Marié et al. (2016) Scanner bank (The James Hutton Institute, UK) • Low-cost, high-resolution root phenotyping platform, requiring no sophisticated equipment and adaptable to most laboratory and glasshouse environments, and applied to quantify environmental and temporal variation in root traits Adu et al. (2014) The Plant Accelerator® (The University of Adelaide, Australia) • The facility offers modern plant growth environments and state-of-the-art high-throughput automated imaging and computing technologies to monitor the performance of plants under different environmental conditions (e.g. which genotype performs best under drought stress) • Research projects facilitated by this technology vary from large scale screening of early growth, salinity tolerance to water and nutrient use efficiency http://www.plantphenomics.org.au/services/accelerator/ DroughtSpotter (The University of Adelaide, Australia) • A gravimetric platform with precision irrigation to assess transpiration dynamics of plants with a precision of up to 1 g • Integrated irrigation units allow precise and reproducible water application for drought stress or related experiments that require an accurate control of water volume to 1 ml http://www.plantphenomics.org.au/services/droughtspotter/ Crop Plant Root Module (The University of Adelaide, Australia) • Comprises destructive and non-destructive measurement of root growth, architecture, morphology, and water uptake in soil in controlled environments and in the field http://www.plantphenomics.org.au/services/croproot/ Jülich Plant Phenotyping Centre (Jülich, Germany) • Reproducibly quantifies growth and architecture of roots • Elucidates dynamic establishment of roots in space and time • Determines interaction of root responses with above ground plant part http://www.fz-juelich.de/ibg/ibg-2/EN/organisation/JPPC/ JPPC_node.html PHENOPSIS (INRA, France) • Automated platform for reproducible phenotyping of plant responses to soil water deficit Granier et al. (2006), Bresson et al. (2015) WIWAM (SMO and VIB, Belgium) • Automated phenotyping platform for automated weighing, watering, and imaging of plants and, therefore, strictly controlling the applied watering regime Skirycz et al. (2011), Clauw et al. (2015) PHENOSCOPE (INRA, France) • Automated large-scale phenotyping platform that automatically adjusts watering and is equipped with a zenithal imaging system to monitor rosette size and expansion rate during the vegetative stage, with automatic image analysis allowing manual correction Tisné et al. (2013) LemnaTec (LemnaTec, Germany) • Versatile phenotyping system and analytics platform for diverse temporal responses to water availability Chen et al. (2014), Neumann et al. (2015) GlyPh (self-construction) • Simple and low-cost automatic platform for high throughput measurement of plant water use and growth to assess drought tolerance Pereyra-Irujo et al. (2012) View Large Table 2. A list of key high-throughput phenotyping platforms Phenotyping platform (and institute or company) Salient features Reference LeasyScan phenotyping platform (ICRISAT, India) • A novel 3D scanning technique to capture leaf area development continuously • Scanner-to-plant concept to increase imaging throughput and analytical scales to combine gravimetric transpiration measurements. • Combines 3D imaging and lysimetry for high- throughput phenotyping of traits controlling plant water budget Vadez et al. (2015) Semi-hydroponic phenotyping system (The University of Western Australia, Australia) • Permits mapping and digital measurement of dynamic growth of taproot and lateral roots • Desirable tool for examining root architecture of deep root systems and large sets of plants in a relatively small space Chen et al. (2011) DEEPER: an integrated phenotyping platform for deeper roots (PennState Colleage of Agriculture, USA) • A platform for identifying the traits of deeper rooted crops in non-destructive field phenotyping of rooting depth, root modeling, high-throughput 3D imaging of root architecture, and anatomy http://plantscience.psu.edu/research/labs/roots/projects/ deeper-an-integrated-phenotyping-platform-for-deeper-roots GLO-Roots: luminescence-based imaging system (Dinnenylab, USA) • Combines custom-made growth vessels and new image analysis algorithms to non-destructively monitor RSA development over space (2D) and time • Allows information on soil properties (e.g. moisture) to be integrated with root growth data • Makes use of luminescence imaging of roots expressing plant codon-optimize luciferase Rellán-Álvarez et al. (2015) X-Ray computed tomography (University of Nottingham, UK) • Non-destructively visualizes opaque root structures by measuring the attenuation of ionizing radiation as it passes through the root • A series of projections are acquired and combined to reconstruct a 3D image of the root system Mairhofer et al. (2012), Mooney et al. (2012) Rhizophonics (University of Liège, Belgium) • Combines hydroponics and rhizotrons • System is made of a nylon fabric supported by an aluminum frame • The set-up is immersed in a tank filled with liquid medium • Allows non-destructive, 2D imaging of root architecture while simultaneously sampling shoots Mathieu et al. (2015) Clear pot method (The University of Queensland, Australia) • Uses transparent pots filled with soil or other p otting media • Seeds are planted close to the pot wall to enable high- throughput imaging of roots along the clear pot wall • To prevent light exposure, the clear pot is placed in black pots while roots are developing Richard et al. (2015) Rhizoslides (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • The set-up consists of a Plexiglas sheet covered with moistened germination paper. Seeds are planted on the slit of the Plexiglas • Allows separation of crown roots from embryonic roots Le Marié et al. (2014) Shovelomics (PennState Colleage of Agriculture, USA) • Involves manual excavation of plants and separating roots from shoots • Washed roots are then placed on a phenotyping board for root trait quantification • New algorithms allow extraction of several root traits in a high throughput manner Trachsel et al. (2011), Bucksch et al. (2014) Soil coring (CSIRO and The University of Queensland, Australia) • Uses a tractor-mounted, hydraulic soil corer to drive steel alloy sampling tubes into the soil • When combined with novel planting configurations (e.g. hill plots), this method allows for phenotyping deep rooted crop varieties Wasson et al. (2014) Rhizo-lysimetry (E. H. Graham Centre for Agricultural Innovation, Australia) • Elaborate facility consisting of an underground corridor and concrete silos and pipes to house soil-containing soil cores for direct root observation Eberbach et al. (2013) Minirhizotrons • A transparent observation tube permanently inserted in the soil. Images of roots growing along the minirhizotron wall at particular locations in the soil profile can be captured over time Iversen et al. (2011), Maeght et al. (2013) RhizoTube (INRA, France) • Cylindrical rhizotrons that allow full 2D visualization of the root system of a single or up to six plants simultaneously • The RhizoCab is designed to take images of the entire root systems of plants growing in RhizoTubes, and also permits a focus on some parts of the root systems Jeudy et al. (2016) RADIX (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • Rhizoslide platform allowing high throughput digital image analysis of root system expansion Le Marié et al. (2016) Scanner bank (The James Hutton Institute, UK) • Low-cost, high-resolution root phenotyping platform, requiring no sophisticated equipment and adaptable to most laboratory and glasshouse environments, and applied to quantify environmental and temporal variation in root traits Adu et al. (2014) The Plant Accelerator® (The University of Adelaide, Australia) • The facility offers modern plant growth environments and state-of-the-art high-throughput automated imaging and computing technologies to monitor the performance of plants under different environmental conditions (e.g. which genotype performs best under drought stress) • Research projects facilitated by this technology vary from large scale screening of early growth, salinity tolerance to water and nutrient use efficiency http://www.plantphenomics.org.au/services/accelerator/ DroughtSpotter (The University of Adelaide, Australia) • A gravimetric platform with precision irrigation to assess transpiration dynamics of plants with a precision of up to 1 g • Integrated irrigation units allow precise and reproducible water application for drought stress or related experiments that require an accurate control of water volume to 1 ml http://www.plantphenomics.org.au/services/droughtspotter/ Crop Plant Root Module (The University of Adelaide, Australia) • Comprises destructive and non-destructive measurement of root growth, architecture, morphology, and water uptake in soil in controlled environments and in the field http://www.plantphenomics.org.au/services/croproot/ Jülich Plant Phenotyping Centre (Jülich, Germany) • Reproducibly quantifies growth and architecture of roots • Elucidates dynamic establishment of roots in space and time • Determines interaction of root responses with above ground plant part http://www.fz-juelich.de/ibg/ibg-2/EN/organisation/JPPC/ JPPC_node.html PHENOPSIS (INRA, France) • Automated platform for reproducible phenotyping of plant responses to soil water deficit Granier et al. (2006), Bresson et al. (2015) WIWAM (SMO and VIB, Belgium) • Automated phenotyping platform for automated weighing, watering, and imaging of plants and, therefore, strictly controlling the applied watering regime Skirycz et al. (2011), Clauw et al. (2015) PHENOSCOPE (INRA, France) • Automated large-scale phenotyping platform that automatically adjusts watering and is equipped with a zenithal imaging system to monitor rosette size and expansion rate during the vegetative stage, with automatic image analysis allowing manual correction Tisné et al. (2013) LemnaTec (LemnaTec, Germany) • Versatile phenotyping system and analytics platform for diverse temporal responses to water availability Chen et al. (2014), Neumann et al. (2015) GlyPh (self-construction) • Simple and low-cost automatic platform for high throughput measurement of plant water use and growth to assess drought tolerance Pereyra-Irujo et al. (2012) Phenotyping platform (and institute or company) Salient features Reference LeasyScan phenotyping platform (ICRISAT, India) • A novel 3D scanning technique to capture leaf area development continuously • Scanner-to-plant concept to increase imaging throughput and analytical scales to combine gravimetric transpiration measurements. • Combines 3D imaging and lysimetry for high- throughput phenotyping of traits controlling plant water budget Vadez et al. (2015) Semi-hydroponic phenotyping system (The University of Western Australia, Australia) • Permits mapping and digital measurement of dynamic growth of taproot and lateral roots • Desirable tool for examining root architecture of deep root systems and large sets of plants in a relatively small space Chen et al. (2011) DEEPER: an integrated phenotyping platform for deeper roots (PennState Colleage of Agriculture, USA) • A platform for identifying the traits of deeper rooted crops in non-destructive field phenotyping of rooting depth, root modeling, high-throughput 3D imaging of root architecture, and anatomy http://plantscience.psu.edu/research/labs/roots/projects/ deeper-an-integrated-phenotyping-platform-for-deeper-roots GLO-Roots: luminescence-based imaging system (Dinnenylab, USA) • Combines custom-made growth vessels and new image analysis algorithms to non-destructively monitor RSA development over space (2D) and time • Allows information on soil properties (e.g. moisture) to be integrated with root growth data • Makes use of luminescence imaging of roots expressing plant codon-optimize luciferase Rellán-Álvarez et al. (2015) X-Ray computed tomography (University of Nottingham, UK) • Non-destructively visualizes opaque root structures by measuring the attenuation of ionizing radiation as it passes through the root • A series of projections are acquired and combined to reconstruct a 3D image of the root system Mairhofer et al. (2012), Mooney et al. (2012) Rhizophonics (University of Liège, Belgium) • Combines hydroponics and rhizotrons • System is made of a nylon fabric supported by an aluminum frame • The set-up is immersed in a tank filled with liquid medium • Allows non-destructive, 2D imaging of root architecture while simultaneously sampling shoots Mathieu et al. (2015) Clear pot method (The University of Queensland, Australia) • Uses transparent pots filled with soil or other p otting media • Seeds are planted close to the pot wall to enable high- throughput imaging of roots along the clear pot wall • To prevent light exposure, the clear pot is placed in black pots while roots are developing Richard et al. (2015) Rhizoslides (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • The set-up consists of a Plexiglas sheet covered with moistened germination paper. Seeds are planted on the slit of the Plexiglas • Allows separation of crown roots from embryonic roots Le Marié et al. (2014) Shovelomics (PennState Colleage of Agriculture, USA) • Involves manual excavation of plants and separating roots from shoots • Washed roots are then placed on a phenotyping board for root trait quantification • New algorithms allow extraction of several root traits in a high throughput manner Trachsel et al. (2011), Bucksch et al. (2014) Soil coring (CSIRO and The University of Queensland, Australia) • Uses a tractor-mounted, hydraulic soil corer to drive steel alloy sampling tubes into the soil • When combined with novel planting configurations (e.g. hill plots), this method allows for phenotyping deep rooted crop varieties Wasson et al. (2014) Rhizo-lysimetry (E. H. Graham Centre for Agricultural Innovation, Australia) • Elaborate facility consisting of an underground corridor and concrete silos and pipes to house soil-containing soil cores for direct root observation Eberbach et al. (2013) Minirhizotrons • A transparent observation tube permanently inserted in the soil. Images of roots growing along the minirhizotron wall at particular locations in the soil profile can be captured over time Iversen et al. (2011), Maeght et al. (2013) RhizoTube (INRA, France) • Cylindrical rhizotrons that allow full 2D visualization of the root system of a single or up to six plants simultaneously • The RhizoCab is designed to take images of the entire root systems of plants growing in RhizoTubes, and also permits a focus on some parts of the root systems Jeudy et al. (2016) RADIX (Institute of Agricultural Sciences, ETH Zurich, Switzerland) • Rhizoslide platform allowing high throughput digital image analysis of root system expansion Le Marié et al. (2016) Scanner bank (The James Hutton Institute, UK) • Low-cost, high-resolution root phenotyping platform, requiring no sophisticated equipment and adaptable to most laboratory and glasshouse environments, and applied to quantify environmental and temporal variation in root traits Adu et al. (2014) The Plant Accelerator® (The University of Adelaide, Australia) • The facility offers modern plant growth environments and state-of-the-art high-throughput automated imaging and computing technologies to monitor the performance of plants under different environmental conditions (e.g. which genotype performs best under drought stress) • Research projects facilitated by this technology vary from large scale screening of early growth, salinity tolerance to water and nutrient use efficiency http://www.plantphenomics.org.au/services/accelerator/ DroughtSpotter (The University of Adelaide, Australia) • A gravimetric platform with precision irrigation to assess transpiration dynamics of plants with a precision of up to 1 g • Integrated irrigation units allow precise and reproducible water application for drought stress or related experiments that require an accurate control of water volume to 1 ml http://www.plantphenomics.org.au/services/droughtspotter/ Crop Plant Root Module (The University of Adelaide, Australia) • Comprises destructive and non-destructive measurement of root growth, architecture, morphology, and water uptake in soil in controlled environments and in the field http://www.plantphenomics.org.au/services/croproot/ Jülich Plant Phenotyping Centre (Jülich, Germany) • Reproducibly quantifies growth and architecture of roots • Elucidates dynamic establishment of roots in space and time • Determines interaction of root responses with above ground plant part http://www.fz-juelich.de/ibg/ibg-2/EN/organisation/JPPC/ JPPC_node.html PHENOPSIS (INRA, France) • Automated platform for reproducible phenotyping of plant responses to soil water deficit Granier et al. (2006), Bresson et al. (2015) WIWAM (SMO and VIB, Belgium) • Automated phenotyping platform for automated weighing, watering, and imaging of plants and, therefore, strictly controlling the applied watering regime Skirycz et al. (2011), Clauw et al. (2015) PHENOSCOPE (INRA, France) • Automated large-scale phenotyping platform that automatically adjusts watering and is equipped with a zenithal imaging system to monitor rosette size and expansion rate during the vegetative stage, with automatic image analysis allowing manual correction Tisné et al. (2013) LemnaTec (LemnaTec, Germany) • Versatile phenotyping system and analytics platform for diverse temporal responses to water availability Chen et al. (2014), Neumann et al. (2015) GlyPh (self-construction) • Simple and low-cost automatic platform for high throughput measurement of plant water use and growth to assess drought tolerance Pereyra-Irujo et al. (2012) View Large Robotic-assisted imaging platforms and computer vision-assisted analysis tools have also been developed for precise phenotyping of physiological growth, development, and other phenotypic properties (Fahlgren et al., 2015a,b). The data processing software that handles high throughput phenotyping for 2D, 3D, and anatomy cross sections of root have been extensively reviewed by Kuijken et al. (2015). The availability of automated high throughput phenotyping platforms enables testing of progenies from large germplasm populations across a range of target environments in terms of their impact on yield. The efficiency of these platforms, in this context, depends on the heritability and correlation of the traits with agronomically relevant breeding targets. Proper identification of the key constraints for a particular production ecology and an understanding of the biological processes can now be achieved with increasing scale and speed using existing platforms and designing new ones. Multiple challenges lie ahead for high throughput phenotyping such as (i) targeting the development of phenotyping protocols and technology specifically for the target traits of their breeding product profiles; (ii) development of precise protocols amenable for high throughput phenotyping for agronomically important traits; (iii) developing appropriate technology for collecting, collating, and analysing data; (iv) storage and processing of huge datasets collected during high throughput phenotyping; and (v) integration and deployment of high throughput phenotyping in breeding pipelines. On top of that, the cost per sample, especially when dealing with large-scale populations, still remains an important challenge to the phenotyping community. One important aspect of better integration of genomics compared with crop physiology in breeding is the lower cost of genotyping per sample compared with field-based phenotyping. Another area that needs immediate attention is the automation of data collection, analysis and delivery of phenotypic data to support the breeding community in a cost-effective and timely manner. Addressing these challenges will accelerate the integration of modern phenotyping tools into breeding programs and accelerate the development of crop varieties with wider adaptation, resilience, and increased productivity and profitability. Crop growth simulation and modelling approaches In recent years, crop simulation and modelling approaches have challenged the idea of broad adaptation cultivars and have brought a new paradigm in breeding for target environments. Crop models, with general genetic inputs, suggest how a given combination of alleles confers a positive or negative effect on plant performance in different locations and seasons (Hammer et al., 2002; Tardieu and Tuberosa, 2010; Messina et al., 2011). Modelling enables us to predict the effect of a change in the biological architecture of a plant type (for instance having faster growing roots, Vadez et al., 2012; for water use efficiency and stay green, Kholová et al., 2014), or of a change in the agronomic management (for instance increasing planting density, Vadez et al., 2017). As such, modelling should become an important resource-saving tool to guide the choice of breeding and agronomic management investment. Crop models have simulated the effect of crop management and climate on yield (Duncan et al., 1967; de Wit et al., 1970; de Wit, 1978; Kumar et al., 2009; Kim et al., 2010). Later, crop models were redesigned into farming systems models that could consider more broadly the effects of the management system including the carryover effects of rotations, planting density and intercropping. More broadly, systems models such as the Agricultural Production Systems sIMulator (APSIM) and the Decision Support System for Agrotechnology Transfer (DSSAT https://dssat.net/) simulate agricultural production systems (Keating et al., 2003; Jones et al., 2003; Holzworth et al., 2014; Hoogenboom et al., 2017), modelling interactions of plants, animals, climate, soil, and land management. Teixeira et al. (2015), using APSIM, demonstrated the importance of rotations for simulating climate impact assessments and Sennhenn et al. (2017) found new niches for short season legumes in Kenya. APSIM has been used in simulation studies for yield gap assessment in legumes such as soybean, groundnut, pigeonpea, and chickpea in India (Bhatia et al., 2007; Chauhan et al., 2008), in simulation of soil temperature in the podding zone of groundnut (Chauhan et al., 2007), and in assessment of the impacts of fertilizers and legumes on N2O and CO2 emissions from soils in subtropical agricultural systems (Huth et al., 2010). With the availability of high throughput phenotyping platforms, the use of crop models in a genetic context has become possible. A conceptual framework was developed by Tardieu and Tuberosa (2010) with four modules that take into consideration phenotyping, genetics, climate, and crop models for estimating the effects of genetic diversity on crop performance. More recently Parent and Tardieu (2014) compared the algorithms involved in 19 crop models developed for predicting the effects of climate or cultivation techniques on a reference genotype. This study reported significant differences in combined effects of temperature and water deficit on plant development, but these differences have a low impact on the yield prediction of a reference genotype because errors in the effects of different traits compensate each other. The impact of climate on crop yield is assessed using two common approaches: (i) process-based simulation models, which attempt to represent key dynamic processes affecting crop yields, and (ii) statistical models, which estimate functional relationships between historical observations of weather and yields (Lobell and Asseng, 2017). The first approach is closer to physiological processes and can explicitly take into account the genetic variability of traits as measured in phenotyping platforms, for both main effects and genotype×environment interaction. However, it is often over-parametrized and can lead to inaccurate predictions in some cases because the model follows its own logic even if not applicable to the considered set of conditions (Tardieu et al., 2017). The statistical approach is more conservative and safer, because predicted yields seldom depart from those observed experimentally. Both approaches are therefore complementary. The impact of drought on legume production is still not well understood in how it varies with legume species, region, agroecosystem, soil texture, and drought timing. For legumes specifically, a model has been developed (Simple Simulation Modelling; SSM) that uses the same model architecture across a range of grain legumes and is easy to use (Soltani and Sinclair, 2012). SSM has been used successfully to predict growth and yield in chickpea (Soltani and Sinclair, 2011; Vadez et al., 2012, 2013b), lentil (Ghanem et al., 2015a,b), common bean (Marrou et al., 2014), soybean (Sinclair et al., 2014), and groundnut (Vadez et al., 2017). In the case of soybean, a recent modelling study indicated that a faster rate of root growth has a negative impact on yield as available soil moisture was depleted faster by more vigorous root growth (Sinclair et al., 2010). While significant progress is being made in crop growth simulation and modelling, their integration with genetics and breeding is still in its infancy. Genomics-assisted breeding Most legume breeding programs focus on yield under drought stress. However, the pace of progress has been slow and the rate of yield gains has been minimal due to the complex nature of drought and non-availability of precise drought screening methods. As a result, the genetic gain achieved is about 1% per year in several species (Duvick, 2005; Cooper et al., 2009; Brisson et al., 2010; Lopes et al., 2012; Aisawi et al., 2015). The most promising option to design more drought resilient and sustainable production is to target the major traits of adaptation that include early flowering and seed set before the onset of terminal drought (Sennhenn et al., 2017). During the past two decades, the availability of molecular markers such as simple sequence repeats (Gupta and Varshney, 2000) and SNPs (Varshney et al., 2010) facilitated dissecting complex traits that hamper crop production, using QTL mapping and genome wide association mapping approaches (Varshney et al., 2015). Genome wide SNPs based on resequencing of several germplasm lines were also used to identify marker-trait associations in the cases of legumes such as pigeonpea (Varshney et al., 2017) and soybean (Zhou et al., 2015a). Genomics-assisted breeding (GAB; Varshney et al., 2005) was proposed to integrate genomics in breeding and it has been quite successful for several traits in cereals (Septiningsih et al., 2013; Varshney et al., 2006) and legumes (Varshney, 2016; Pratap et al., 2017). Recently, Reynolds and Langridge (2016) have suggested crossing parents with different complex but complementary traits to achieve cumulative gene action for yield, while selecting progeny using remote sensing in combination with genomic selection. Genetic and genomics approaches and innovative tools have been made available during the past decade to breed climate-resilient legumes; for instance, root xylem plasticity and its role in improving water use efficiency in soybean (Prince et al., 2017). The QTLs/genomic regions and markers associated with traits are used to breed for stress resilience using marker-assisted selection (MAS), marker-assisted back crossing (MABC), and MARS approaches (Varshney et al., 2012b). The legume community has been successful in developing several molecular breeding products despite the late arrival of genomic resources and trait-associated markers (Varshney et al., 2013a,b; Pandey et al., 2016; Varshney, 2016). Some key examples include resistance to Fusarium wilt and ascochyta blight (Varshney et al., 2013b) and improved drought tolerance (Varshney et al., 2013a) in chickpea; resistance to nematode and high oleic acid (Chu et al., 2011), resistance to leaf rust (Varshney et al., 2014), and resistance to high oleic acid (Janila et al., 2016) in groundnut; resistance to rust disease (Khanh et al., 2013), soybean mosaic virus (Saghai-Maroof et al., 2008; Shi et al., 2009; Parhe et al., 2017), and low phytate (Landau-Ellis and Pantalone, 2009) in soybean; Striga resistance and seed size in cowpea (Lucas et al., 2015; see Boukar et al., 2016); pyramid genes for resistance to ascochyta blight and anthracnose in lentil (Taran et al., 2003); powdery mildew resistance (Ghafoor and McPhee 2012), lodging resistance (Zhang et al., 2006), frost tolerance (see Tayeh et al., 2015b), and Aphanomyces root rot resistance (Lavaud et al., 2015) in pea; and resistance to common bacterial blight disease (Miklas et al., 2000, 2006; Mutlu et al., 2005; O’Boyle and Kelly, 2007), rust and viruses (Stavely, 2000), rust, anthracnose, and angular leaf spot (Oliveira et al., 2008), rust (Feleiro et al., 2001), and anthracnose (Alzate-Marin et al., 1999) in common bean. Several of these improved lines have either been released or are in the release pipeline in different countries. In addition to the above, the gene/QTL pyramiding efforts are at an advanced stage for several legume traits such as in groundnut (leaf rust resistance+late leaf spot+high oleic acid), chickpea (fusarium wilt resistance+ascochyta blight resistance+drought tolerance; see Pandey et al., 2016), and cowpea (resistance to Striga+aphid+macrophomina root rot; see Boukar et al., 2016). The majority of the legume crops now enjoy the availability of genomic resources and such examples are seen more frequently in grain legumes. The MARS approach was deployed in cowpea successfully for improving grain yield, drought tolerance, Striga resistance and Macrophomina resistance (see Boukar et al., 2016), but this approach did not work in the case of chickpea (see Varshney et al., 2013c). Next generation GAB approaches While some examples, as mentioned above, are available for the use of molecular breeding, they include introgression of one to a few traits with higher heritability through MABC. Due to advances in low-cost genotyping/sequencing and availability of diagnostic markers, we expect use of early generation selection, genomic selection, and genome editing technologies in future GAB approaches. Early generation selection Early generation selection (EGS) is performed in highly segregating populations (mostly F2) while advanced generation selection (AGS) begins with F4 onwards followed by assessment of yield performance and other important agronomically superior traits in subsequent generations. Both EGS and AGS can be used to breed improved lines followed by evaluating these lines in target populations of environment (i.e. the set of conditions under which the set of considered varieties has been bred) to select promising lines with wider adaptation and higher yield under water limited environments. The success of EGS depends on (i) efficiency and accuracy of phenotyping a selected trait in a segregating population of single plants, for instance phenotyping for leaf wilting and selection for root depth in tubes, and (ii) genetic makeup of the population; for instance, drought tolerance has been introgressed from the wild species resulting in the appearance of undesirable and agronomically segregating populations. EGS will be highly desirable only when F2s are from agronomically superior elite lines. In the case of soybean, EGS for pods per plant proved to be the most effective for enhancing the yield (Singh and Sharma, 2016). AGS holds considerable promise for traits that are best expressed in homozygous progeny. While EGS and AGS have been earlier practised based on phenotypic selection, diagnostic markers are now being integrated for traits, for both EGS and AGS. A high-throughput genotyping project (HTPG; http://cegsb.icrisat.org/high-throughput-genotyping-project-htpg/), an initiative led by ICRISAT in collaboration with the International Maize and Wheat Improvement Center (CIMMYT) and the International Rice Research Institute (IRRI) and funded by the Bill & Melinda Gates Foundation, is expected to accelerate the deployment of EGS/AGS. It is important to note that the HTPG project has made it possible to screen populations at a 2017 cost of US$ 1.5 per sample for 10 SNP markers including DNA extraction. While several institutes such as CIMMYT, ICRISAT and IRRI have started to deploy EGS at a large scale in maize, rice, wheat, and groundnut, the genomics community needs to work on developing additional diagnostic markers for more traits especially nutrition traits across all legume crops. Genomic selection Genomic selection (GS), proposed by Meuwissen et al. (2001), is a promising breeding approach for simultaneous improvement of complex traits, especially those with low heritability. GS uses genomic signatures or dense marker profiles of individuals as well as parents for predicting the breeding values. The use of GS for accelerating genetic gains, based on genomic-estimated breeding values predicted using genome-wide dense markers, has been deployed extensively in dairy cattle (Van Raden et al., 2009). García-Ruiz et al. (2016) reported a positive impact of GS on cattle breeding through drastic reduction of generation interval and increased selection intensity for low heritable traits. Initially, computer simulations and parametric and non-parametric statistical models were used on maize and wheat datasets to study the prediction accuracies in real plant breeding scenarios (Bernardo and Yu, 2007; de los Campos et al., 2009; Crossa et al., 2010, 2011). Promising results in major cereals provided a kick start for deploying GS in a few legume crops, such as pea, soybean, chickpea, groundnut, and pigeonpea (Varshney, 2016). GS in crop plants including legumes has been extensively reviewed recently elsewhere (Crossa et al., 2017). In GS, in brief, a training population (genotyped with genome wide markers and extensively phenotyped over years and in several locations) is trained with appropriate models to estimate the genomic-estimated breeding values in a breeding population (genotyped with genome wide markers) and validated in a validation population. Hickey et al. (2017) compared both plant and animal breeding approaches to make a case for bringing the two together through the application of GS. In recent years, the GS approach has been deployed in different types of populations, for instance bi-parental, multi-parental (Beyene et al., 2015), and synthetic populations (Schopp et al., 2017) and diverse panel of inbred lines, which basically differ in the pedigree, number of founders used to develop the population, linkage disequilibrium decay, and population structure. The genetic advance achieved through selection depends on the total variation (σ2), the repeatability of the trait (h2) and the selection pressure (S) imposed. The selection pressure implies the proportion of the population selected. Even if the repeatability is high the genetic advance (genetic gain, R) would be small without a large genetic variation. With selection, the genetic variation and consequently the repeatability and thus the advance from one to the next generation decline while the mean value of the trait increases (or decreases, depending on the trait). Genotyping platforms, training populations and statistical models Although the availability of genome wide markers and cost of genotyping were two major factors that restricted implementation of GS in crop breeding, they are no longer constraints as most of the legume communities are now equipped with genome sequence information as well as high density SNP arrays. For instance, high density SNP arrays have become available in case of chickpea (Roorkiwal et al., 2017), groundnut (Pandey et al., 2017a), and pigeonpea (Saxena et al., 2017 and unpublished). In addition, the efficiency and accuracy with which superior lines can be predicted also depend on the size of the reference population (Jannink et al., 2010; Lorenz et al., 2011). The genetic relatedness or population structure (Saatchi et al., 2011; Riedelsheimer et al., 2013; Wray et al., 2013) may result in overestimating the heritability of the traits (Price et al., 2010; Visscher et al., 2012; Wray et al., 2013). The population structure of the training population can be determined with greater accuracy using genome wide SNPs compared with the simple sequence repeats and SNP arrays (Isidro et al., 2015). Accuracy of estimating breeding values is important in GS. For instance, Jarquín et al. (2014) obtained a high prediction accuracy of 0.64 by deploying GS for improving yield and agronomic traits using genotyping-by-sequencing in a breeding program. In addition, GS for yield ridge regression best linear unbiased prediction (BLUP) coupled with fivefold cross-validations and marker preselection based on haplotype blocks is an interesting option for a cost-efficient implementation of genomic selection for grain yield in soybean breeding (Ma et al., 2016). In the case of pea, using GS, Tayeh et al. (2015c) reported mean cross-environment prediction accuracies of 0.83 for thousand-seed weight, 0.68 for number of seeds per plant, and 0.65 for date of flowering. Population structure did not impact the prediction accuracy, and modelling genotype by environment by management (G×E×M) indicated improved prediction efficiency in chickpea (Roorkiwal et al., 2016). Further, Roorkiwal et al. (2016) used statistical models such as RR-BLUP, Kinship GAUSS, Bayes Cπ, Bayes B, Baysian LASSO, and random forest regression and reported high prediction accuracies for days to maturity, days to flowering, and seed dry weight. The development and deployment of improved and more precise statistical models will eventually enhance further the prediction accuracy leading to enhanced GS efficiency in legume crops. Development of models that use deep machine-learning methods and multi-trait and multi-environment information is essential to realize the potential of GS in legume breeding and to enhance the prediction accuracies (Crossa et al., 2017). Selection intensity and generation interval The selection differential (S) is defined as the difference of base population mean and the mean of the selected parents R=h2S where h2 is the narrow-sense heritability (Arruda et al., 2016). The selection intensity depends on how different the selected parents are from the overall population average. The closer the selected parents to the population average, the smaller the selection intensity and vice versa. The impact of GS and generation intervals in cattle breeding was first systematically studied by de Roos et al. (2010). Recently, García-Ruiz et al. (2016) reported that decreased generation interval and increased selection intensity for low heritable traits is as a result of the positive impact of GS on the US cattle industry. Genomic selection in combination with a reduced generation interval may double the rate of genetic gain while keeping the rate of inbreeding per generation constant. Rapid cycling The usefulness of rapid cycling genomic selection (RCGS) was first reported in a bi-parental mapping population derived from B73 and Mo17. Massman et al. (2013) reported that RCGS had a superior response for stover yield, as well as stover and grain yield indices that were 14–50% higher than those of MARS. Recently, in tropical maize, by employing RCGS on eight bi-parental populations evaluated under drought stress environments, Beyene et al. (2015) reported an average gain of 0.086 ton ha−1 per cycle, and hybrids derived from cycle 3 produced 7.3% (0.176 ton ha−1) higher grain yield than those from cycle 0 and conventional pedigree breeding methods. Further, Zhang et al. (2017) using RCGS in a tropical maize multi-parental population reported a genetic gain for grain yield from cycle 1 to cycle 4 reached 0.225 ton ha−1 per cycle, which is equivalent to 0.100 ton ha−1 year−1 over a 4.5 year breeding period from the initial cross to the last cycle. While similar rates of gain have not been reported in legumes, recent reports on over three generations per year in traditional breeding for pigeonpea (Saxena et al., 2017) offer new opportunities to apply RCGS to accelerate rates of genetic gain in grain legumes. Genome editing The sequencing-based trait dissection and gene discovery are enabling the functional characterization of genes. An alternative approach for modifying targeted genomic regions, genome editing, may be better suited for improving the deficiency of one or two traits within popular varieties that have suitable production and market traits. Enhancing the performance of market-preferred varieties offers the advantage of higher productivity and higher replacement rates by farmers (Dar et al., 2013). Genome editing harnesses the strength of programmable nucleases in cutting and pasting the specific genetic information in living cells (Kim, 2016). The availability of more and more information on functionality of genes or haplotypes for economically important traits will facilitate modification of multiple SNPs through genome editing without leading to a deterioration in the original behavior of a popular cultivar. In several of the legume crop species, with genetic barriers such as ploidy differences, the exchange of genetic material through natural process is restricted and therefore the huge genetic variation lying within crop’s wild relatives remains untapped and unutilized. In this context, if the function of the specific genome variations in wild relatives can be assigned, genome editing can enable targeted editing of a popular cultivar without fear of linkage drag. The recently developed targeted genome-editing technologies such as zinc finger nucleases, RNA-guided endonucleases, transcription-activator-like effector nucleases, and CRISPR (cluster of regularly interspaced palindromic repeats) have undergone tremendous improvement over the past couple of years (Wang et al., 2017). Of the several approaches, CRISPR received wider acceptance among researchers due to its huge potential role in crop improvement in coming years (Kim, 2016). This technology has already been optimized in three model legume crops, soybean (Sun et al., 2015; Li et al., 2015), Lotus japonicas (Wang et al., 2016) and Medicago truncatula (Meng et al., 2017). The efficiency of genome editing using various endogenous and exogenous RNA polymerase III promoters has been compared (Sun et al., 2015; Du et al., 2016) and in soybean the Cas9–single guide RNA together with donor DNA fragments was successfully transformed into soybean using the particle bombardment method (Li et al., 2015). Medicago and Lotus, having smaller diploid genomes, are the model legume crops for conducting research in several areas including symbiotic nitrogen fixation. The efficacy of the CRISPR system has been assessed in these two crops by modifying the multiple genes related to symbiotic nitrogen fixation (Michno et al., 2015; Wang et al., 2016; Meng et al., 2017). Recently, Benson Hill Biosystems, an agricultural technology company based in the UK, launched a commercial service for genome editing for crop improvement (https://www.prnewswire.com/news/benson-hill-biosystems). These developments will further accelerate the deployment of this promising technology in improving desired traits in legume crops. Genome editing is considered to be a safe technology for developing improved crop plants by US regulators, who indicated that genome editing products developed using oligonucleotide-directed mutagenesis or site-directed nucleases should not be subjected to regulation as these technologies are similar to mutagenesis (see Sprink et al., 2016). There are multiple examples of development of improved plant varieties using genome editing technology and 37 such examples in Arabidopsis, tobacco, maize, rice, wheat, canola, soybean, camelina, tomato, and cucumber have been well documented recently (see Kamburova et al., 2017). Of these successful efforts, several of these products have been approved in different countries for cultivation, such as herbicide-tolerant SU Canola (Cibus 5715) in Canada in 2014 (https://cen.acs.org/articles/95/i24/CRISPR-new-toolbox-better-crops.html), browning-resistant Arctic apples in the USA and Canada in 2015 (Smyth, 2017), and less-acrylamide-producing potatoes for chip processing in Canada in 2016 (Smyth, 2017), while tomato with a greater number of fruits and soybean with high oleic acid are near to reaching approval. Expected products in Canada and the USA in the coming years include glyphosate-tolerant flax in 2019. Products expected for 2020 include late-blight-resistant potato, herbicide-tolerant rice, and starchy or ‘waxy’ corn. Improving the operational efficiency in breeding programs Improvement in operational efficiency in the breeding programs will help in conducting precise and well-documented research experiments. Currently, the majority of breeding programs still record data on paper. The digitalization of breeding programs and improved operational efficiency will reduce the chance of error, which in turn will allow breeders to assess the potential of improved breeding procedures (Varshney et al., 2016). In this context, digitalization, automation, and mechanization in breeding procedures need to be adopted during the breeding process. More emphasis is required on uniform ontology, digitalized pedigree information, appropriate experimental design, barcoding of breeding material, digital data recording in the field, data management systems, and quality control. Integrated Breeding Platform (http://www.integratedbreeding.net) is one such initiative that is helping breeding programs to modernize and improving their breeding efficiency. For instance, the Breeding Management System of Integrated Breeding Platform is being used extensively at ICRISAT and its partner breeding programs. Similarly, B4R (http://bbi.irri.org/home) is a program for managing breeding data that has been developed and is being used by IRRI. The volume of data generated throughout the breeding process has increased significantly due to the involvement of high throughput genotyping and phenotyping technologies. Data volume now poses a significant challenge in compilation, processing, and interpretation. Available tools engaged today may no longer be suitable for handling high volume data in the future, thereby necessitating development of improved tools/software that can drive science-based breeding at a faster pace with user-friendly tools and databases. The Genomic Open-source Breeding Informatics Initiative is one such endeavor (http://gobiiproject.org/) targeting deployment of high density genotyping information in public sector breeding programs for more precise selection to achieve higher genetic gains. Further, the Excellence in Breeding (EiB; http://excellenceinbreeding.org/) platform came into existence very recently to bring all the components of CGIAR crop improvement under a single platform to focus on better integration and modernizing of breeding programs in developing countries. The above efforts will help in improving the efficiency and resilience of agricultural systems to achieve more sustainable food production for increased profitability of farming, greater resilience and nutrition of global agrifood systems. Seed systems Once superior varieties are developed, they should reach farmers’ fields as early as possible such that the average age of the varieties in the farmers’ fields are not more than 10 years. Cultivar replacement is indeed one of the basic undertakings in the EiB platform. In the past, however, most released varieties of legumes have not enjoyed high rates of adoption (Atlin et al., 2017). One reason could be low performance of new varieties under farmers’ field management and conditions. The second reason is limited multi-location testing of public sector varieties due to limited funds. As a result, seed companies do not know the level of superiority of a new variety for a certain production ecology and are reluctant to invest in producing and promoting a new seed product. Our experience from the Tropical Legumes project (http://tropicallegumes.icrisat.org/) shows that adoption of new varieties requires promotion just like with new products in the manufacturing sector. As compared with hybrid maize, where the private sector invests heavily in thousands of demonstration plots to create awareness and popularize new varieties, only very limited demonstration plots of new legume varieties and accompanying agronomic practices are conducted. Farmers’ awareness of improved varieties has been reported to be strongly correlated with higher adoption rates for pigeonpea in Tanzania (Amare et al., 2012) and for chickpea in Ethiopia (Abate et al., 2012). However, unlike the private manufacturing sector where a new product is promoted and new demand can be met quickly, seed is the vehicle through which genetic gains are realized in farmers’ fields, especially with small-scale farmers in the case of legumes. Production of seed takes time through cycles of breeding, foundation and certified seed, each stage taking at least one cropping season. Compared with cereals and canola, legumes generally have a lower seed multiplication rate, thereby requiring extra space, labor, time, effort, and more generations to produce sufficient quantities of certified seed. This increases the cost of seed production to seed producers, which must be paid for by farmers, many of whom find the costs prohibitive compared with non-authentic sources. In addition, virtually all legumes are self-pollinated (pigeonpea is an exception) and farmers usually recycle their own seeds. Therefore, the incentive to develop a business around seed production in most grain legumes is low. Some legumes have relatively large seed rate requirements, sometimes more than 100 kg ha−1, meaning farmers have to buy large quantities to plant a relatively smaller area compared with cereals. This has additional implications in terms of requirement of seed storage space and packaging material, further increasing the cost of seed. In addition, transportation of the seed to remote areas is both cumbersome and expensive due to the bulky nature of grain legumes. Legume seeds also suffer from loss of germination faster than cereals, especially in hot, humid environments. Seed producers and traders must meet these challenges by investing in proper storage, thereby increasing the willingness of farmers to buy seed regularly off-farm with an assurance of better quality than farm-save seed (Sperling and McGuire, 2010). Besides these challenges, legumes are highly self-pollinated and it is assumed that farmers save their own seed for several seasons and do not need to buy new seed each season, which makes it difficult for seed companies to predict demand for seed. It was, however, reported in 2016 that a significant fraction of legume farmers (64.4%) actually do buy ‘seed’ or rather grain from local markets (McGuire and Sperling, 2016). However, due to high cost, limited availability, and access, farmers do not buy legume seed from the formal seed sector (seed companies and agrodealers). Innovative approaches are therefore necessary to tackle the challenge of legume seed supply to ensure availability, accessibility, affordability, and sustainability for the varieties to replace older varieties. This requires an integrated approach to seed production and delivery systems that includes formal public–private sector partnerships and a variety of farmer-based seed production and supply initiatives (Siddique et al., 2012; Johansen and Siddique 2017). One of the approaches used to increase total seed production and availability, particularly of early generation seed, is through licensing of public varieties for decentralized production. An example of this is in seed production and distribution of improved varieties of common bean in Ethiopia. Well-established innovation platforms along the bean value chain contributed to sustaining a decentralized seed production system (Rubyogo et al., 2010). Seed certification, normally by public agencies, is an important quality control step but it also increases the cost of seed significantly. Kenya and Zambia are examples of countries where certification is licensed to private institutions and individuals. Quality declared seed (QDS) is an approach to producing seed by competent seed producers and suppliers that accommodates the diversity of farming systems, particularly in the more difficult areas where highly organized seed systems do not function well (Plant Production and Protection Division, 2006). QDS standards and monitoring procedures are slightly less demanding, allowing for less expensive seed and more decentralized producers. The quality and performance of QDS seed on farmers’ fields is comparable—but the seed is less expensive (Granqvist, 2006). Innovative approaches are also needed for increasing availability, supply and access to high quality seed of improved varieties (e.g. through small packs, financial support, better coordination with extension or development programs). A small seed pack approach was pioneered by Pan-African Bean Research Alliance two decades ago to encourage informal varietal diffusion and adoption of new varieties (Grisley, 1993) and gained popularity as an efficient and cost effective means of reaching more farmers with bean seed of improved varieties (Maereka and Rubyogo, 2015). It has been adopted and popularized in the Tropical Legumes II/III project (Monyo and Varshney 2016), involving national programs and the private sector (Rubyogo et al., 2016). Demand creation for quality seed includes working with off-takers like grain exporters, demonstration plots of better varieties, post-harvest handling, and market linkages. Good examples of the role of off-takers can be cited with the Raphael Group Ltd in southern Tanzania that has promoted the adoption of the ‘Uyole 03’ sugar bean variety and ACOS Ltd in Ethiopia that has promoted market-orientated production of various varieties of white pea bean and kabuli chickpea. Reaching farmers ‘at the last mile’ by bundling seed with other products, piggybacking on existing product supply channels such as fertilizers and pesticides, and strengthening community seed production approaches are all critically important areas for innovation. Agronomy packages Improved seeds alone do not guarantee improved yields. Soil fertility is often a big constraint for smallholder farmers in the semi-arid tropics (Kamanga et al., 2013). Deployment of improved varieties should be accompanied by integrated crop management practices, underscoring the prevailing cropping practices. In fact, consideration of agronomic practices should be considered early in the development of product profiles so that breeding populations are developed accordingly and selection pressure integrates prevailing agronomic practices (e.g. soil fertility, crop density). Combining organic and mineral fertilizer is a sound management principle for smallholder farmers when combined with techniques such as crop rotations to improve soil fertility and nutrient availability. Two examples clearly illustrate the integration of integrated soil fertility management principles: (i) dual purpose grain legume–maize rotations with P fertilizer targeted at the legume phase and N fertilizer targeted at the cereal phase in the moist savanna agro-ecozone (Sanginga et al., 2003), and (ii) micro-dose fertilizer applications in legume–sorghum or legume–millet rotations with retention of crop residues and water harvesting techniques in semi-arid agro-ecozones (Bationo et al., 1998). In trials in West Africa, control of early season insect pests and diseases increased crop vigor, and greater yields have been reported in legumes such as cowpea (Tekwa et al., 2010) and groundnut (Yakubu et al., 2011) with ApronStar, a seed treatment fungicide–insecticide mixture for controlling downy mildew, damping-off diseases, as well as for protection of seeds and seedlings against early season insect pests and soil-borne diseases. Therefore, there is a need in the future to breed cultivars in a system perspective, i.e. to breed crops considering elements of the agro-ecosystem in which these will be deployed. Understanding abiotic and biotic stresses is important in designing breeding programs. For example, planting groundnuts early in the season when aphid populations are still low coupled with maintaining a good plant density can reduce aphids that are the vector for rosette disease, which in turn reduces disease incidence (Waliyar et al., 2007). This in turn would ensure optimum yield and that varieties developed for other traits, such as oil quality, early maturity, drought tolerance, and foliar disease resistance, are not lost due to rosette disease. Sowing time is also particularly important in pigeonpea, where there are different maturity groups. Early maturing varieties can compete with the cereals in intercropping systems but would be the best suited as a sole crop in places prone to short rainfall periods. Medium maturing varieties are planted after the main cereals and remain in the field when the main cereal is harvested. Similarly, proper timing is critical in chickpea to ensure the crop does not suffer from terminal drought (if planted too late) or seedling fungal diseases if planted too early during the rainy period. Integrated and coordinated approach Legumes are nutrient-rich crops with high levels of protein and other important nutritional components that offer great support in fighting malnutrition across the world. Genetic improvement for legumes needs to accommodate traditional yield-improvement priorities alongside nutritional characteristics. A single technology or process cannot be singled out to achieve higher yields in farmers’ fields with high consistency (Siddique et al 2012; Johansen and Siddique, 2017). Participatory feedback from different stakeholders along commodity value chains is required to develop appropriate product profiles so farmers can be more productive and profitable by accessing markets with surplus production. In this context, a holistic approach is required to realize higher genetic gains in less time and with more precision in the legume crops. We have proposed this strategy in Fig. 1. Fig. 1. View largeDownload slide Strategy for strengthening the breeding process and adopting improved technologies for achieving higher genetic gains in farmers’ field. Low-cost sequencing and genotyping technology will eventually help in molecular characterization of entire diverse germplasms. High density genotyping data together with phenotyping data on entire germplasm will help filter the most useful germplasm lines for use in breeding programs. Low-cost sequencing and genotyping data will also facilitate dissection of multiple traits even in complex genetic populations and faster and precise gene discovery. Improved and modern approaches need to be integrated in trait mapping and trait improvement pipeline in addition to adoption of improved breeding operational efficiency. Higher genetic gains achieved with improvement of desired traits, key to major stakeholders, will eventually help not only in attaining higher productivity but also in increasing the profitability of farming. Fig. 1. View largeDownload slide Strategy for strengthening the breeding process and adopting improved technologies for achieving higher genetic gains in farmers’ field. Low-cost sequencing and genotyping technology will eventually help in molecular characterization of entire diverse germplasms. High density genotyping data together with phenotyping data on entire germplasm will help filter the most useful germplasm lines for use in breeding programs. Low-cost sequencing and genotyping data will also facilitate dissection of multiple traits even in complex genetic populations and faster and precise gene discovery. Improved and modern approaches need to be integrated in trait mapping and trait improvement pipeline in addition to adoption of improved breeding operational efficiency. Higher genetic gains achieved with improvement of desired traits, key to major stakeholders, will eventually help not only in attaining higher productivity but also in increasing the profitability of farming. Our crop improvement programs, which include not just breeding but all other disciplines, need to start with demand and feedback from the stakeholders including farmers, consumers, and markets. Based on these demands, crop improvement programs prioritize traits (including nutrition) and define a product profile of the varieties to be developed. Subsequently, genebank, pre-breeding and trait discovery groups together with breeders and biometricians formulate a list of suitable genotypes with desired traits, and high breeding values are identified after a comprehensive genome characterization using sequencing and genotyping platforms and phenotypic characterization in target environments. In parallel, we need to have diagnostic markers with higher prediction accuracy and candidate genes with causal effect for a given trait. For trait discovery, instead of using a traditional mapping approach, it is possible now to use sequencing-based trait mapping approaches such as QTL-Seq, Bulk-Seq and Indel-Seq (Singh et al., 2016, 2017; Pandey et al., 2017b). Similarly, recent advances in functional genomics such as gene expression atlases (Pazhamala et al., 2017) and use of near-isogenic lines for transcriptomics approaches can identify the candidate genes for the target traits. In order to increase selection intensity, it is essential to increase the number of crosses and the population of each cross in public sector breeding programs. In our opinion, each breeding program should make at least 100 crosses every year and each cross should have at least 500–1000 F2s using the most appropriate parental combinations. Of course, the size of the population would also need to be adjusted to the genetics of the target trait. Early generation screening with the diagnostic markers or candidate genes for must-have traits can dramatically reduce early generation populations as a ‘forward breeding’ process. After selecting a small numbers of lines, genomic prediction can be undertaken and selected lines with higher breeding values can be targeted for testing in target population of environments. Subsequently, the selected lines can go to a national program for further multi-location and multi-year testing for possible release of improved varieties or can be included in the pool of parental selections in the new crossing schemes. We believe that such an approach will accelerate the development of superior grain legume varieties. In parallel, it is also possible to continue to undertake MAS or MABC in the ruling varieties for introgressing major genes/QTLs that have become deficient during their long-term cultivation in farmers’ fields. While undertaking GAB approaches, it is also essential that we improve our operational efficiency with appropriate analytical and decision support tools. For instance, appropriate experimental design for testing a segregating population/germplasm collection is very important. Similarly, having phenotypic data for different breeding populations/germplasm collections in databases and using these databases for planning a new set of experiments is critical. Deployment of breeding databases such as BMS and B4R in breeding programs needs to be accelerated. Use of tablets and mobile-based ‘Field Book’ programs will reduce error and turnaround time for data-driven breeding decisions. While using the GAB strategies, EGS, GS, MABC or MAS, it is important that breeding programs use the appropriate decision support tools to select the superior lines instead of using spreadsheets or stand-alone programs. All these efforts and good breeding practices together offer operational efficiency in the breeding programs. While use of GAB approaches together with operational efficiency methods can help in developing better varieties faster, the full genetic potential of these varieties can only be realized once they have been adopted and grown in farmers’ fields. It is essential to work with government agencies together with informal seed system actors in different developing countries to accelerate varietal replacement. We need to start using ‘digital seed road maps’ (http://seedsystems.icrisat.org/) that can help different stakeholders of the seed value chain to assess the need and make different kinds of seeds (breeders’ seed, foundation seeds, certified/truthfully labeled seeds) available to farmers. Finally, digital agriculture tools, for example, sowing apps (http://www.icrisat.org/tag/ sowing-app/), plantix (http://www.icrisat.org/tag/plantix/), should be deployed to provide increased access to equitable markets by farmers to increase profitability. We believe that use of such a holistic approach will not just deliver higher genetic gains in research plots of scientists but also in farmers’ field. By using the entire approach from science of discovery to science of delivery, the scientific community will be able to contribute to both enhancing crop productivity and increasing farm profitability. Furthermore, it will also help to make cheap raw material available to industry, and finally, availability of affordable nutritious food to the poorest of the poor in society. Abbreviations: Abbreviations: AGS advanced generation selection BLUP best linear unbiased prediction EGS early generation selection GAB genomics-assisted breeding GS genomic selection MABC marker-assisted back crossing MARS marker-assisted recurrent selection MAS marker-assisted selection QDS quality declared seed RCGS rapid cycling genomic selection SNP single nucleotide polymorphism. Acknowledgements The authors from ICRISAT are thankful to the Bill & Melinda Gates Foundation (Tropical Legumes I, II & III), United States Agency for International Development (USAID), MARS Chocolate Inc., Indian Council of Agricultural Research (ICAR), and Department of Biotechnology (DBT) of Government of India. The work reported in this article was undertaken as a part of the CGIAR Research Program on Grain Legumes and Dryland Cereals (GLDC). ICRISAT is a member of the CGIAR. References Abate T , Alene AD , Bergvinson D , Shiferaw B , Silim S , Orr A , Asfaw S . 2012 . Tropical grain legumes in Africa and south Asia: knowledge and opportunities . Nairobi, Kenya : International Crops Research Institute for the Semi-Arid Tropics . https://core.ac.uk/download/pdf/12107473.pdf. Adu MO , Chatot A , Wiesel L , Bennett MJ , Broadley MR , White PJ , Dupuy LX . 2014 . A scanner system for high-resolution quantification of variation in root growth dynamics of Brassica rapa genotypes . Journal of Experimental Botany 65 , 2039 – 2048 . Google Scholar CrossRef Search ADS Aisawi KAB , Reynolds MP , Singh RP , Foulkes MJ . 2015 . The physiological basis of the genetic progress in yield potential of CIMMYT spring wheat cultivars from 1966 to 2009 . Crop Science 55 , 1749 . Google Scholar CrossRef Search ADS Akibode CS . 2011 . Trends in the production, trade, and consumption of food-legume crops in sub-Saharan Africa . MSc Thesis, Michigan State University . Akond M , Liu S , Schoener L et al. 2013 . SNP-based genetic linkage map of soybean using the SoySNP6K illumina infinium BeadChip genotyping array . Journal of Plant Genome Science 1 , 80 – 89 . Alzate-Marin AL , Menarim H , de Carvalho GA , de Paula TJ , de Barros EG , Moreira MA . 1999 . Improved selection with newly identified RAPD markers linked to resistance gene to four pathotypes of Colletotrichum lindemuthianum in common bean . Phytopathology 89 , 281 – 285 . Google Scholar CrossRef Search ADS Amare M , Asfaw S , Shiferaw B . 2012 . Welfare impacts of maize-pigeon pea intensification in Tanzania . Agricultural Economics 43 , 27 – 43 . Google Scholar CrossRef Search ADS Arruda MP , Lipka AE , Brown PJ et al. 2016 . Comparing genomic selection and marker-assisted selection for Fusarium head blight resistance in wheat (Triticum aestivum L.) . Molecular Breeding 36 , 84 . Google Scholar CrossRef Search ADS Atlin GN , Cairns JE , Das B . 2017 . Rapid breeding and varietal replacement are critical to adaptation of cropping systems in the developing world to climate change . Global Food Security 12 , 31 – 37 . Google Scholar CrossRef Search ADS Bankole F , Menkir A , Olaoye G , Crossa J , Hearne S , Unachukwu N , Gedil M . 2017 . Genetic gains in yield and yield related traits under drought stress and favorable environments in a maize population improved using marker assisted recurrent selection . Frontiers in Plant Science 8 , 808 . Google Scholar CrossRef Search ADS Barker T , Campos H , Cooper M , Dolan D , Edmeades G , Habben J , Schussler , Wright D , Zinselmeier C . 2010 . Improving drought tolerance in maize . Plant Breeding Reviews 25 , 173 – 253 . Bationo A , Lompo F , Koala S . 1998 . Research on nutrient flows and balances in West Africa: state-of the art . Agriculture, Ecosystems and Environment 71 , 19 – 35 . Bernardo R , Yu J . 2007 . Prospects for genome-wide selection for quantitative traits in maize . Crop Science 47 , 1082 . Google Scholar CrossRef Search ADS Bertioli DJ , Cannon SB , Froenicke L et al. 2016 . The genome sequences of Arachis duranensis and Arachis ipaensis, the diploid ancestors of cultivated peanut . Nature Genetics 48 , 438 – 446 . Google Scholar CrossRef Search ADS Beyene Y , Semagn K , Mugo S et al. 2015 . Genetic gains in grain yield through genomic selection in eight bi-parental maize populations under drought stress . Crop Science 55 , 154 . Google Scholar CrossRef Search ADS Bhatia VS , Singh P , Wani SP , Rao AVRK , Srinivas K . 2007 . Yield gap analysis of soybean, groundnut, pigeonpea and chickpea in India using simulation modeling . Journal of SAT Agricultural Research 5 , 1 – 160 . Bohra A , Pandey MK , Jha UC , Singh B , Singh IP , Datta D , Chaturvedi SK , Nadarajan N , Varshney RK . 2014 . Genomics-assisted breeding in four major pulse crops of developing countries: present status and prospects . Theoretical and Applied Genetics 127 , 1263 – 1291 . Google Scholar CrossRef Search ADS Boukar O , Fatokun CA , Huynh BL , Roberts PA , Close TJ . 2016 . Genomic tools in cowpea breeding programs: status and perspectives . Frontiers in Plant Science 7 , 757 . Google Scholar CrossRef Search ADS Bresson J , Vasseur F , Dauzat M , Koch G , Granier C , Vile D . 2015 . Quantifying spatial heterogeneity of chlorophyll fluorescence during plant growth and in response to water stress . Plant Methods 11 , 23 . Google Scholar CrossRef Search ADS Brisson N , Gate P , Gouache D , Charmet G , Oury F-X , Huard F . 2010 . Why are wheat yields stagnating in Europe? A comprehensive data analysis for France . Field Crops Research 119 , 201 – 212 . Google Scholar CrossRef Search ADS Bucksch A , Burridge J , York LM , Das A , Nord E , Weitz JS , Lynch JP . 2014 . Image-based high-throughput field phenotyping of crop roots . Plant Physiology 166 , 470 – 486 . Google Scholar CrossRef Search ADS Cabrera-Bosquet L , Fournier C , Brichet N , Welcker C , Suard B , Tardieu F . 2016 . High-throughput estimation of incident light, light interception and radiation-use efficiency of thousands of plants in a phenotyping platform . New Phytologist 212 , 269 – 281 . Google Scholar CrossRef Search ADS Chauhan Y , Wright G , Rachaputi NR , Krosch S , Robertson M , Hargreaves J , Broome A . 2007 . Using APSIM-soiltemp to simulate soil temperature in the podding zone of peanut . Australian Journal of Experimental Agriculture 47 , 992 – 999 . Google Scholar CrossRef Search ADS Chauhan Y , Wright G , Rachaputi N , McCosker K . 2008 . Identifying chickpea homoclimes using the APSIM chickpea model . Australian Journal of Agricultural Research 59 , 260 – 269 . Google Scholar CrossRef Search ADS Chen D , Neumann K , Friedel S , Kilian B , Chen M , Altmann T , Klukas C . 2014 . Dissecting the phenotypic components of crop plant growth and drought responses based on high-throughput image analysis . The Plant Cell 26 , 4636 – 4655 . Google Scholar CrossRef Search ADS Chen X , Li H , Pandey MK et al. 2016 . Draft genome of the peanut A-genome progenitor (Arachis duranensis) provides insights into geocarpy, oil biosynthesis and allergens . Proceedings of the National Academy of Sciences, USA 113 , 6785 – 6790 . Google Scholar CrossRef Search ADS Chen YL , Dunbabin VM , Diggle AJ , Siddique KHM , Rengel Z . 2011 . Development of a novel semi-hydroponic phenotyping system for studying root architecture . Functional Plant Biology 38 , 355 – 363 . Google Scholar CrossRef Search ADS Chu Y , Wu CL , Holbrook CC , Tillman BL , Person G , Ozias-Akins P . 2011 . Marker-assisted selection to pyramid nematode resistance and high oleic trait in peanut . The Plant Genome 4 , 110 – 117 . Google Scholar CrossRef Search ADS Clauw P , Coppens F , De Beuf K , Dhondt S , Van Daele T , Maleux K , Storme V , Clement L , Gonzalez N , Inzé D . 2015 . Leaf responses to mild drought stress in natural variants of Arabidopsis . Plant Physiology 167 , 800 – 816 . Google Scholar CrossRef Search ADS Clevenger J , Chu Y , Chavarro C et al. 2017 . Genome-wide SNP genotyping resolves signatures of selection and tetrasomic recombination in peanut . Molecular Plant 10 , 309 – 322 . Google Scholar CrossRef Search ADS Contreras-Soto RI , de Oliveira MB , Costenaro-da-Silva , Scapim CA , Schuster I . 2017 . Population structure, genetic relatedness and linkage disequilibrium blocks in cultivars of tropical soybean (Glycine max) . Euphytica 213 , 173 . Google Scholar CrossRef Search ADS Cooper M , van Eeuwijk FA , Hammer GL , Podlich DW , Messina C . 2009 . Modeling QTL for complex traits: detection and context for plant breeding . Current Opinion in Plant Biology 12 , 231 – 240 . Google Scholar CrossRef Search ADS Crossa J , de Campos GL , Pérez P et al. 2010 . Prediction of genetic values of quantitative traits in plant breeding using pedigree and molecular markers . Genetics 186 , 713 – 724 . Google Scholar CrossRef Search ADS Crossa J , Pérez P , de los Campos G , Mahuku G , Dreisigacker S , Magorokosho C . 2011 . Genomic selection and prediction in plant breeding . Journal of Crop Improvement 25 , 239 – 261 . Google Scholar CrossRef Search ADS Crossa J , Pérez-Rodríguez P , Cuevas J et al. 2017 . Genomic selection in plant breeding: methods, models, and perspectives . Trends in Plant Science 22 , 961 – 975 . Google Scholar CrossRef Search ADS Dar MH , de Janvry A , Emerick K , Raitzer D , Sadoulet E . 2013 . Flood-tolerant rice reduces yield variability and raises expected yield, differentially benefitting socially disadvantaged groups . Scientific Reports 3 , 3315 . Google Scholar CrossRef Search ADS de los Campos G , Naya H , Gianola D , Crossa J , Legarra A , Manfredi E , Weigel K , Cotes JM . 2009 . Predicting quantitative traits with regression models for dense molecular markers and pedigree . Genetics 182 , 375 – 385 . Google Scholar CrossRef Search ADS de Roos AP , Schrooten C , Veerkamp RF , van Arendonk JA . 2010 . The impact of genomic selection and short generation interval on dairy cattle breeding programs . In: Proceedings of 9th World Congress on Genetics Applied to Livestock Production . http://www2.naut.is/Files/Skra_0043312.pdf. Deulvot C , Charrel H , Marty A , Jacquin F , Donnadieu C , Lejeune-Hénaut I , Burstin J , Aubert G . 2010 . Highly-multiplexed SNP genotyping for genetic mapping and germplasm diversity studies in pea . BMC Genomics 11 , 468 . Google Scholar CrossRef Search ADS de Wit CT , Brouwer R , Penning de Vries FWT . 1970 . The simulation of photosynthetic systems . In: Setlik I , ed. Prediction and measurements of photosynthetic productivity . Proceedings of the IBP/PP Technical Meeting , Trebon . Wageningen : Pudoc , 47 – 50 . de Wit CT . 1978 . Simulation of assimilation, respiration and transpiration of crops. Simulation Monographs . Wageningen : Centre for Agricultural Publishing and Documentation . Doyle JJ , Luckow MA . 2003 . The rest of the iceberg. Legume diversity and evolution in a phylogenetic context . Plant Physiology 131 , 900 – 910 . Google Scholar CrossRef Search ADS Du H , Zeng X , Zhao M , Cui X , Wang Q , Yang H , Cheng H , Yu D . 2016 . Efficient targeted mutagenesis in soybean by TALENs and CRISPR/Cas9 . Journal of Biotechnology 217 , 90 – 97 . Google Scholar CrossRef Search ADS Duncan WG , Loomis RS , Williams WA , Hanau R . 1967 . A model for simulating photosynthesis in plant communities . Hilgardia 38 , 181 – 205 . Google Scholar CrossRef Search ADS Duvick DN . 2005 . Genetic progress in yield of United States maize . Maydica 50 , 193 – 202 . Eberbach PL , Hoffmann J , Moroni SJ , Wade LJ , Weston LA . 2013 . Rhizo-lysimetry: facilities for the simultaneous study of root behaviour and resource use by agricultural crop and pasture systems . Plant Methods 9 , 3 . Google Scholar CrossRef Search ADS Fahlgren N , Feldman M , Gehan MA et al. 2015a. A versatile phenotyping system and analytics platform reveals diverse temporal responses to water availability in Setaria . Molecular Plant 8 , 1520 – 1535 . Google Scholar CrossRef Search ADS Fahlgren N , Gehan MA , Baxter I . 2015b. Lights, camera, action: high-throughput plant phenotyping is ready for a close-up . Current Opinion in Plant Biology 24 , 93 – 99 . Google Scholar CrossRef Search ADS FAOSTAT . 2016 . Review of CGIAR priorities and strategies . Rome : Food and Agriculture Organization of the United Nations . www.fao.org/wairdocs/tac/x5756e/x5756e08.htm. Feleiro FG , Vinhadelli WS , Ragagnin VA , Vinhadelli WS , Moreira MA , Stavely JR , de Barros EG . 2001 . Resistance of bean lines to four races of Uromyces appenduculata isolated in the state of Minas Gerais . Fitopatologia Brasileiara 26 , 77 – 80 . Google Scholar CrossRef Search ADS Fiorani F , Schurr U . 2013 . Future scenarios for plant phenotyping . Annual Review of Plant Biology 64 , 267 – 291 . Google Scholar CrossRef Search ADS Foyer CH , Lam HM , Nguyen HT et al. 2016 . Neglecting legumes has compromised human health and sustainable food production . Nature Plants 2 , 16112 . Google Scholar CrossRef Search ADS García-Ruiz A , Cole JB , VanRaden PM et al. 2016 . Changes in genetic selection differentials and generation intervals in US Holstein dairy cattle as a result of genomic selection . Proceedings of the National Academy of Sciences, USA 113 , E3995 – E4004 . Google Scholar CrossRef Search ADS Ghafoor A , and McPhee K . 2012 . Marker assisted selection (MAS) for developing powdery mildew resistant pea cultivars . Euphytica 186 , 593 – 607 . Google Scholar CrossRef Search ADS Ghanem ME , Marrou H , Biradar C , Sinclair TR . 2015a. Production potential of lentil (Lens culinaris Medik.) in East Africa . Agricultural Systems 137 , 24 – 38 . Google Scholar CrossRef Search ADS Ghanem ME , Marrou H , Soltani A , Kumar S , Sinclair TR . 2015b. Lentil variation in phenology and yield evaluated with a model . Agronomy Journal 107 , 1967 – 77 . Google Scholar CrossRef Search ADS Gioia T , Galinski A , Lenz H et al. 2016 . GrowScreen-PaGe, a non-invasive, high-throughput phenotyping system based on germination paper to quantify crop phenotypic diversity and plasticity of root traits under varying nutrient supply . Functional Plant Biology 44 , 76 – 93 . Google Scholar CrossRef Search ADS Granier C , Aguirrezabal L , Chenu K et al. 2006 . PHENOPSIS, an automated platform for reproducible phenotyping of plant responses to soil water deficit in Arabidopsis thaliana permitted the identification of an accession with low sensitivity to soil water deficit . New Phytologist 169 , 623 – 635 . Google Scholar CrossRef Search ADS Granqvist B . 2006 . Is quality declared seed production an effective way to address seed and food security in Africa ?http://knowledge.cta.int/Dossiers/S-T-Policy/ACP-agricultural-S-T-dialogue/Demanding-Innovation/Feature-articles/Is-Quality-Declared-Seed-Production-an-effective-and-sustainable-way-to-address-Seed-and-Food-Security-in-Africa. Grisley W . 1993 . Seed for bean production in sub-Saharan Africa: issues, problems, and possible solutions . Agricultural Systems 43 , 19 – 33 . Google Scholar CrossRef Search ADS Gupta S , Nawaz K , Parween S , Roy R , Sahu K , Kumar Pole A , Khandal H , Srivastava R , Kumar Parida S , Chattopadhyay D . 2017 . Draft genome sequence of Cicer reticulatum L., the wild progenitor of chickpea provides a resource for agronomic trait improvement . DNA Research 24 , 1 – 10 . Gupta PK , Varshney RK . 2000 . The development and use of microsatellite markers for genetics and plant breeding with emphasis on bread wheat . Euphytica 113 , 163 – 185 . Google Scholar CrossRef Search ADS Hammer GL , Kropff MJ , Sinclair TR , Porter JR . 2002 . Future contributions of crop modelling—from heuristics and supporting decision making to understanding genetic regulation and aiding crop improvement . European Journal of Agronomy 18 , 15 – 31 . Google Scholar CrossRef Search ADS Hickey JM , Chiurugwi T , Mackay I , Powell W , Implementing Genomic Selection in CGIAR Breeding Programs Workshop Participants . 2017 . Genomic prediction unifies animal and plant breeding programs to form platforms for biological discovery . Nature Genetics 49 , 1297 – 1303 . Google Scholar CrossRef Search ADS Holzworth DP , Huth NI , Zurcher EJ et al. 2014 . APSIM – evolution towards a new generation of agricultural systems simulation . Environmental Modelling & Software 62 , 327 – 350 . Google Scholar CrossRef Search ADS Hoogenboom G , Porter CH , Shelia V et al. 2017 . Decision Support System for Agrotechnology Transfer (DSSAT) Version 4.7 , https://DSSAT.net. Gainesville, FL, USA: DSSAT Foundation . Huth NI , Thorburn PJ , Radford BJ , Thornton CM . 2010 . Impacts of fertilisers and legumes on N2O and CO2 emissions from soils in subtropical agricultural systems: a simulation study . Agriculture Ecosystems & Environment 136 , 351 – 357 . Google Scholar CrossRef Search ADS Hyten DL , Song Q , Choi IY , Yoon MS , Specht JE , Matukumalli LK , Nelson RL , Shoemaker RC , Young ND , Cregan PB . 2008 . High-throughput genotyping with the GoldenGate assay in the complex genome of soybean . Theoretical and Applied Genetics 116 , 945 – 952 . Google Scholar CrossRef Search ADS Isidro J , Jannink JL , Akdemir D , Poland J , Heslot N , Sorrells ME . 2015 . Training set optimization under population structure in genomic selection . Theoretical and Applied Genetics 128 , 145 – 158 . Google Scholar CrossRef Search ADS Iversen CM , Murphy MT , Allen MF , Childs J , Eissenstat DM , Lilleskov EA , Sarjala TM , Sloan VL , Sullivan PF . 2011 . Advancing the use of minirhizotrons in wetlands . Plant Soil 352 , 23 – 39 . Google Scholar CrossRef Search ADS Jaganathan D , Thudi M , Kale S , Azam S , Roorkiwal M , Gaur PM , Kishor PB , Nguyen H , Sutton T , Varshney RK . 2015 . Genotyping-by-sequencing based intra-specific genetic map refines a “QTL-hotspot” region for drought tolerance in chickpea . Molecular Genetics and Genomics 290 , 559 – 571 . Google Scholar CrossRef Search ADS Janila P , Pandey MK , Shasidhar Y et al. 2016 . Molecular breeding for introgression of fatty acid desaturase mutant alleles (ahFAD2A and ahFAD2B) enhances oil quality in high and low oil containing peanut genotypes . Plant Science 242 , 203 – 213 . Google Scholar CrossRef Search ADS Jannink JL , Lorenz AJ , Iwata H . 2010 . Genomic selection in plant breeding: from theory to practice . Briefings in Functional Genomics 9 , 166 – 177 . Google Scholar CrossRef Search ADS Jarquín D , Kocak K , Posadas L , Hyma K , Jedlicka J , Graef G , Lorenz A . 2014 . Genotyping by sequencing for genomic prediction in a soybean breeding population . BMC Genomics 15 , 740 . Google Scholar CrossRef Search ADS Jeudy C , Adrian M , Baussard C et al. 2016 . RhizoTubes as a new tool for high throughput imaging of plant root development and architecture: test, comparison with pot grown plants and validation . Plant Methods 12 , 31 . Google Scholar CrossRef Search ADS Johansen C , Siddique K . 2017 . Grain legumes in integrated crop . In: Shivasankar S , Bergvinson D , Gaur P , Kumar S , Beebe S , Tamo M , eds. Achieving sustainable cultivation of grain legumes. Chapter 10, Grain legumes in integrated crop management systems . Sawston, UK : Burleigh Dodds Science Publishing Limited . Jones JW , Hoogenboom G , Porter CH , Boote KJ , Batchelor WD , Hunt LA , Wilkens PW , Singh U , Gijsman AJ , Ritchie JT . 2003 . DSSAT cropping system model . European Journal of Agronomy 18 , 235 – 265 . Google Scholar CrossRef Search ADS Kale SM , Jaganathan D , Ruperao P et al. 2015 . Prioritization of candidate genes in “QTL-hotspot” region for drought tolerance in chickpea (Cicer arietinum L.) . Scientific Reports 5 , 15296 . Google Scholar CrossRef Search ADS Kamanga BCG , Waddington SR , Whitbread A , Almekinders CJM , Giller K . 2013 . Improving the efficiency of use of small amounts of nitrogen and phosphorus fertiliser on smallholder maize in central Malawi . Experimental Agriculture 50 , 229 – 249 . Google Scholar CrossRef Search ADS Kamburova VS , Nikitina EV , Shermatov SE , Buriev ZT , Kumpatla SP , Emani C , Abdurakhmonov IY . 2017 . Genome editing in plants: an overview of tools and applications . International Journal of Agronomy 2017 , 7315351 . Google Scholar CrossRef Search ADS Kang YJ , Kim SK , Kim MY et al. 2014 . Genome sequence of mungbean and insights into evolution within Vigna species . Nature Communications 5 , 5443 . Google Scholar CrossRef Search ADS Kang YJ , Satyawan D , Shim S et al. 2015 . Draft genome sequence of adzuki bean, Vigna angularis . Scientific Reports 5 , 8069 . Google Scholar CrossRef Search ADS Keating BA , Carberry PS , Hammer GL et al. 2003 . An overview of APSIM, a model designed for farming systems simulation . European Journal of Agronomy 18 , 267 – 288 . Google Scholar CrossRef Search ADS Khanh T , Anh T , Buu B et al. 2013 . Applying molecular breeding to improve soybean rust resistance in Vietnamese elite soybean . American Journal of Plant Sciences 4 , 1 – 6 . Google Scholar CrossRef Search ADS Kholová J , Murugesan T , Kaliamoorthy S et al. 2014 . Modelling the effect of plant water use traits on yield and stay-green expression in sorghum . Functional Plant Biology 41 , 1019 – 1034 . Google Scholar CrossRef Search ADS Kim HK , van Oosterom E , Dingkuhn M , Luquet D , Hammer G . 2010 . Regulation of tillering in sorghum: environmental effects . Annals of Botany 106 , 57 – 67 . Google Scholar CrossRef Search ADS Kim JS . 2016 . Genome editing comes of age . Nature Protocols 11 , 1573 – 1578 . Google Scholar CrossRef Search ADS Koester RP , Skoneczka JA , Cary TR , Diers BW , Ainsworth EA . 2014 . Historical gains in soybean (Glycine max Merr.) seed yield are driven by linear increases in light interception, energy conversion, and partitioning efficiencies . Journal of Experimental Botany 65 , 3311 – 3321 . Google Scholar CrossRef Search ADS Kuijken RC , van Eeuwijk FA , Marcelis LF , Bouwmeester HJ . 2015 . Root phenotyping: from component trait in the lab to breeding . Journal of Experimental Botany 66 , 5389 – 5401 . Google Scholar CrossRef Search ADS Kumar SR , Hammer GL , Broad I , Harland P , McLean G . 2009 . Modelling environmental effects on phenology and canopy development of diverse sorghum genotypes . Field Crops Research 111 , 157 – 165 . Google Scholar CrossRef Search ADS Kumar V , Khan AW , Saxena RK , Garg V , Varshney RK . 2016 . First-generation HapMap in Cajanus spp. reveals untapped variations in parental lines of mapping 1 populations . Plant Biotechnology Journal 14 , 1673 – 1681 . Google Scholar CrossRef Search ADS Lam HM , Xu X , Liu X et al. 2010 . Resequencing of 31 wild and cultivated soybean genomes identifies patterns of genetic diversity and selection . Nature Genetics 42 , 1053 – 1059 . Google Scholar CrossRef Search ADS Landau-Ellis D , Pantalone VR . 2009 . Marker-assisted backcrossing to incorporate two low phytate alleles into the Tennessee soybean cultivar 5601T . In: Shu GY , ed. Induced plant mutations in the genomics era . Rome : Food and Agriculture Organization of the United Nations (FAO) , 316 – 318 . Lavaud C , Lesné A , Piriou C , Le Roy G , Boutet G , Moussart A , Poncet C , Delourme R , Baranger A , Pilet-Nayel ML . 2015 . Validation of QTL for resistance to Aphanomyces euteiches in different pea genetic backgrounds using near-isogenic lines . Theoretical and Applied Genetics 128 , 2273 – 2288 . Google Scholar CrossRef Search ADS Le Marié C , Kirchgessner N , Flütsch P , Pfeifer J , Walter A , Hund A . 2016 . RADIX: rhizoslide platform allowing high throughput digital image analysis of root system expansion . Plant Methods 12 , 40 . Google Scholar CrossRef Search ADS Le Marié C , Kirchgessner N , Marschall D , Walter A , Hund A . 2014 . Rhizoslides: paper-based growth system for non-destructive, high throughput phenotyping of root development by means of image analysis . Plant Methods 10 , 13 . Google Scholar CrossRef Search ADS Leamy LJ , Zhang H , Li C , Chen CY , Song BH . 2017 . A genome-wide association study of seed composition traits in wild soybean (Glycine soja) . BMC Genomics 18 , 18 . Google Scholar CrossRef Search ADS Lee YG , Jeong N , Kim JH et al. 2015 . Development, validation and genetic analysis of a large soybean SNP genotyping array . The Plant Journal 81 , 625 – 636 . Google Scholar CrossRef Search ADS Li Z , Liu ZB , Xing A , Moon BP , Koellhoffer JP , Huang L , Ward RT , Clifton E , Falco SC , Cigan AM . 2015 . Cas9-Guide RNA directed genome editing in soybean . Plant Physiology 169 , 960 – 970 . Google Scholar CrossRef Search ADS Lobell DB , Asseng S . 2017 . Comparing estimates of climate change impacts from process based and statistical crop models . Environment Research Letters 12 , 015001 . Google Scholar CrossRef Search ADS Lopes MS , Reynolds MP , Manes Y , Singh RP , Crossa J , Braun HJ . 2012 . Genetic yield gains and changes in associated traits of CIMMYT spring bread wheat in an historic set representing 30 years of breeding . Crop Science 52 , 1123 . Google Scholar CrossRef Search ADS Lorenz AJ , Chao S , Asoro FG , Heffner EL . 2011 . Genomic selection in plant breeding: knowledge and prospects . Advances in Agronomy 110 , 77 – 123 . Google Scholar CrossRef Search ADS Lucas MR , Huynh BL , Roberts PA , Close TJ . 2015 . Introgression of a rare haplotype from Southeastern Africa to breed California blackeyes with larger seeds . Frontiers in Plant Science 6 , 126 . Ma Y , Reif JC , Jiang Y et al. 2016 . Potential of marker selection to increase prediction accuracy of genomic selection in soybean (Glycine max L.) . Molecular Breeding 36 , 113 . Google Scholar CrossRef Search ADS Maeght JL , Rewald B , Pierret A . 2013 . How to study deep roots—and why it matters . Frontiers in Plant Science 4 , 299 . Google Scholar CrossRef Search ADS Maereka EK , Rubyogo JC . 2015 . Integrated bean seed systems in Africa: implications for community seed production . In: Ojiewo CO , Kugbei S , Bishaw Z , Rubyogo JC , eds. Community seed production . Workshop proceedings , December 2013 . Addis Ababa : ICRISAT and Rome : FAO , 9 – 11 . Maldonado dos Santos JV , Valliyodan B , Joshi T et al. 2016 . Evaluation of genetic variation among Brazilian soybean cultivars through genome resequencing . BMC Genomics 17 , 110 . Google Scholar CrossRef Search ADS Mairhofer S , Zappala S , Tracy SR , Sturrock C , Bennett M , Mooney SJ , Pridmore T . 2012 . RooTrak: automated recovery of three-dimensional plant root architecture in soil from x-ray microcomputed tomography images using visual tracking . Plant Physiology 158 , 561 – 569 . Google Scholar CrossRef Search ADS Marrou H , Sinclair TR , Metral R . 2014 . Assessment of irrigation scenarios to improve performances of Lingot bean (Phaseolus vulgaris) in southwest France . European Journal of Agronomy 59 , 5922 – 5928 . Google Scholar CrossRef Search ADS Massman JM , Gordillo A , Lorenzana RE , Bernardo R . 2013 . Genomewide predictions from maize single-cross data . Theoretical and Applied Genetics 126 , 13 – 22 . Google Scholar CrossRef Search ADS Mathieu L , Lobet G , Tocquin P , Périlleux C . 2015 . “Rhizoponics”: a novel hydroponic rhizotron for root system analyses on mature Arabidopsis thaliana plants . Plant Methods 11 , 3 . Google Scholar CrossRef Search ADS McGuire S , Sperling L . 2016 . Seed systems smallholder farmers use . Food Security 8 , 179 – 195 . Google Scholar CrossRef Search ADS Meng Y , Hou Y , Wang H , Ji R , Liu B , Wen J , Niu L , Lin H . 2017 . Targeted mutagenesis by CRISPR/Cas9 system in the model legume Medicago truncatula . Plant Cell Reports 36 , 371 – 374 . Google Scholar CrossRef Search ADS Messina CD , Podlich D , Dong Z , Samples M , Cooper M . 2011 . Yield-trait performance landscapes: from theory to application in breeding maize for drought tolerance . Journal of Experimental Botany 62 , 855 – 868 . Google Scholar CrossRef Search ADS Meuwissen TH , Hayes BJ , Goddard ME . 2001 . Prediction of total genetic value using genome-wide dense marker maps . Genetics 157 , 1819 – 1829 . Michno JM , Wang X , Liu J , Curtin SJ , Kono TJ , Stupar RM . 2015 . CRISPR/Cas mutagenesis of soybean and Medicago truncatula using a new web-tool and a modified Cas9 enzyme . GM Crops & Food 6 , 243 – 252 . Google Scholar CrossRef Search ADS Miklas PN , Kelly JD , Beebe SE , Blair MW . 2006 . Common bean breeding for resistance against biotic and abiotic stresses: from classical to MAS breeding . Euphytica 147 , 105 – 131 . Google Scholar CrossRef Search ADS Miklas PN , Smith JR , Riley R , Grafton KF , Singh SP , Jung G , Coyne DP . 2000 . Marker-assisted breeding for pyramided resistance to common bacterial blight in common bean . Annual Reports Bean Improvement Cooperatives 43 , 39 – 40 . Monyo ES , Varshney RK . 2016 . Seven seasons of learning and engaging smallholder farmers in the drought-prone areas of sub-Saharan Africa and South Asia through Tropical Legumes, 2007–2014 . Patancheru, Telangana, India : International Crops Research Institute for the Semi-Arid Tropics . Mooney S , Pridmore T , Helliwell J , Bennett MJ . 2012 . Developing X-ray computed tomography to non-invasively image 3-D root systems architecture in soil . Plant Soil 352 , 1 – 22 . Google Scholar CrossRef Search ADS Muchero W , Diop NN , Bhat PR et al. 2009 . A consensus genetic map of cowpea [Vigna unguiculata (L) Walp.] and synteny based on EST-derived SNPs . Proceedings of the National Academy of Sciences, USA 106 , 18159 – 18164 . Google Scholar CrossRef Search ADS Muñoz-Amatriaín M , Mirebrahim H , Xu P et al. 2017 . Genome resources for climate-resilient cowpea, an essential crop for food security . The Plant Journal 89 , 1042 – 1054 . Google Scholar CrossRef Search ADS Mutlu N , Miklas PN , Reiser J et al. 2005 . Backcross breeding for improved resistance to common bacterial blight in pinto bean (Phaseolus vulgaris L.) . Plant Breeding 124 , 282 – 287 . Google Scholar CrossRef Search ADS Nagy ED , Guo Y , Tang S et al. 2012 . A high-density genetic map of Arachis duranensis, a diploid ancestor of cultivated peanut . BMC Genomics 13 , 469 . Google Scholar CrossRef Search ADS Neumann K , Klukas C , Friedel S , Rischbeck P , Chen D , Entzian A , Stein N , Graner A , Kilian B . 2015 . Dissecting spatiotemporal biomass accumulation in barley under different water regimes using high-throughput image analysis . Plant, Cell & Environment 38 , 1980 – 1996 . Google Scholar CrossRef Search ADS O’Boyle PD , Kelly JD . 2007 . Use of marker-assisted selection to breed for resistance to common bacterial blight in common bean . Journal of American Society of Horticultural Science 132 , 381 – 386 . Oliveira LK , Melo LC , Brondani C , Peloso MJ , Brondani RP . 2008 . Backcross assisted by microsatellite markers in common bean . Genetics and Molecular Research 7 , 1000 – 1010 . Google Scholar CrossRef Search ADS Pandey MK , Roorkiwal M , Singh VK , Ramalingam A , Kudapa H , Thudi M , Chitikineni A , Rathore A , Varshney RK . 2016 . Emerging genomic tools for legume breeding: current status and future prospects . Frontiers in Plant Science 7 , 455 . Pandey MK , Agarwal G , Kale SM et al. 2017a. Development and evaluation of a high density genotyping ‘Axiom_Arachis’ array with 58K SNPs for accelerating genetics and breeding in groundnut . Scientific Reports 7 , 40577 . Google Scholar CrossRef Search ADS Pandey MK , Khan AW , Singh VK et al. 2017b. QTL-seq approach identified genomic regions and diagnostic markers for rust and late leaf spot resistance in groundnut (Arachis hypogaea L.) . Plant Biotechnology Journal 15 , 927 – 941 . Google Scholar CrossRef Search ADS Parent B , Tardieu F . 2014 . Can current crop models be used in the phenotyping era for predicting the genetic variability of yield of plants subjected to drought or high temperature ? Journal of Experimental Botany 65 , 6179 – 6189 . Google Scholar CrossRef Search ADS Parhe SD , Chimote VP , Deshmukh MP , Chandra K , Akash M . 2017 . Marker-assisted pyramiding of four QTL/genes for Asian rust (Phakopsora pachyrhizi) resistance in soybean . Journal of Crop Improvement 31 , 689 – 711 . Google Scholar CrossRef Search ADS Parween S , Nawaz K , Roy R et al. 2015 . An advanced draft genome assembly of a desi type chickpea (Cicer arietinum L.) . Scientific Reports 5 , 12806 . Google Scholar CrossRef Search ADS Pazhamala LT , Purohit S , Saxena RK , Garg V , Krishnamurthy L , Verdier J , Varshney RK . 2017 . Gene expression atlas of pigeonpea and its application to gain insights into genes associated with pollen fertility implicated in seed formation . Journal of Experimental Botany 68 , 2037 – 2054 . Google Scholar CrossRef Search ADS Pereyra-Irujo GA , Gasco ED , Peirone LS , Aguirrezabal L . 2012 . GlyPh: a low-cost platform for phenotyping plant growth and water use . Functional Plant Biology 39 , 905 – 913 . Google Scholar CrossRef Search ADS Plant Production and Protection Division . 2006 . Quality declared seed system . FAO Plant Production and Protection Paper 185. Rome : Food and Agriculture Organization of the United Nations . PubMed PubMed Pratap A , Chaturvedi SK , Tomar R , Rajan N , Malviya N , Thudi M , Saabale PR , Prajapati U , Varshney RK , Singh NP . 2017 . Marker-assisted introgression of resistance to fusarium wilt race 2 in Pusa 256, an elite cultivar of desi chickpea . Molecular Genetics and Genomics 292 , 1237 – 1245 . Google Scholar CrossRef Search ADS Price AL , Zaitlen NA , Reich D , Patterson N . 2010 . New approaches to population stratification in genome-wide association studies . Nature Reviews Genetics 11 , 459 – 463 . Google Scholar CrossRef Search ADS Prince SJ , Murphy M , Mutava RN , Durnell LA , Valliyodan B , Shannon JG , Nguyen HT . 2017 . Root xylem plasticity to improve water use and yield in water-stressed soybean . Journal of Experimental Botany 68 , 2027 – 2036 . Rellán-Álvarez R , Lobet G , Lindner H et al. 2015 . GLO-Roots: an imaging platform enabling multidimensional characterization of soil-grown root systems . eLife 4 , 07597 . Google Scholar CrossRef Search ADS Reynolds M , Langridge P . 2016 . Physiological breeding . Current Opinion in Plant Biology 31 , 162 – 171 . Google Scholar CrossRef Search ADS Richard CA , Hickey LT , Fletcher S , Jennings R , Chenu K , Christopher JT . 2015 . High-throughput phenotyping of seminal root traits in wheat . Plant Methods 11 , 13 . Google Scholar CrossRef Search ADS Riedelsheimer C , Endelman JB , Stange M , Sorrells ME , Jannink JL , Melchinger AE . 2013 . Genomic predictability of interconnected biparental maize populations . Genetics 194 , 493 – 503 . Google Scholar CrossRef Search ADS Roorkiwal M , Jain A , Kale SM , Doddamani D , Chitikineni A , Thudi M , Varshney RK . 2017 . Development and evaluation of high density SNP array (Axiom® CicerSNP Array) for high resolution genetic mapping and breeding applications in chickpea . Plant Biotechnology Journal 16 , 890 – 901 . Google Scholar CrossRef Search ADS Roorkiwal M , Rathore A , Das RR et al. 2016 . Genome-enabled prediction models for yield related traits in Chickpea . Frontiers in Plant Science 7 , 1666 . Google Scholar CrossRef Search ADS Roorkiwal M , Sawargaonkar SL , Chitikineni A , Thudi M , Saxena RK , Upadhyaya HD , Vales MI , Riera-Lizarazu O , Varshney RK . 2013 . Single nucleotide polymorphism genotyping for breeding and genetics applications in chickpea and pigeonpea using the BeadXpress platform . The Plant Genome 6 , doi: 10.3835/plantgenome2013.05.0017 . Rubyogo JC , Myer GM , Ajeigbe H et al. 2016 . Integrated seed systems delivering on the promise: experiences from Tropical Legumes II . In: Monyo ES , Varshney RK . Seven seasons of learning and engaging smallholder farmers in the drought-prone areas of sub-Saharan Africa and South Asia through Tropical Legumes, 2007–2014 . Patancheru, Telangana, India : International Crops Research Institute for the Semi-Arid Tropics , 167 – 178 . Rubyogo JC , Sperling JC , Muthoni R , Buruchara R . 2010 . Bean seed delivery for small farmers in sub-saharan Africa: the power of partnerships . Society & Natural Resources 4 , 285 – 302 . Google Scholar CrossRef Search ADS Saatchi M , McClure MC , McKay SD et al. 2011 . Accuracies of genomic breeding values in American Angus beef cattle using K-means clustering for cross-validation . Genetics, Selection, Evolution 43 , 40 . Google Scholar CrossRef Search ADS Saghai-Maroof MA , Jeong SC , Gunduz I , Tucker DM , Buss GR , Tolin S . 2008 . Pyramiding of soybean mosaic virus resistance genes by marker assisted selection . Crop Science 48 , 517 – 526 . Google Scholar CrossRef Search ADS Sanginga N , Dashiell K , Diels J et al. 2003 . Sustainable resource management coupled to resilient germplasm to provide new intensive cereal–grain–legume–livestock systems in the dry Savanna . Agriculture, Ecosystems and Environment 100 , 305 – 314 . Google Scholar CrossRef Search ADS Saxena K , Saxena RK , Varshney RK . 2017 . Use of immature seed germination and single seed descent for rapid genetic gains in pigeonpea . Plant Breeding 136 , 954 – 957 . Google Scholar CrossRef Search ADS Schmutz J , Cannon SB , Schlueter J et al. 2010 . Genome sequence of the palaeopolyploid soybean . Nature 463 , 178 – 183 . Google Scholar CrossRef Search ADS Schmutz J , McClean PE , Mamidi S et al. 2014 . A reference genome for common bean and genome-wide analysis of dual domestications . Nature Genetics 46 , 707 – 713 . Google Scholar CrossRef Search ADS Schopp P , Müller D , Technow F , Melchinger AE . 2017 . Accuracy of genomic prediction in synthetic populations depending on the number of parents, relatedness, and ancestral linkage disequilibrium . Genetics 205 , 441 – 454 . Google Scholar CrossRef Search ADS Sennhenn A , Njarui DMG , Maass BL , Whitbread AM . 2017 . Exploring niches for short-season grain legumes in semi-arid Eastern Kenya—coping with the impacts of climate variability . Frontiers in Plant Science 8 , 699 . Google Scholar CrossRef Search ADS Septiningsih EM , Collard BC , Heuer S , Bailey-Serres J , Ismail AM , Mackill DJ . 2013 . Applying genomics tools for breeding submergence tolerance in rice . In: Varshney RK , Tuberosa R , eds. Translational genomics for crop breeding, Vol II, Abiotic stress, yield and quality . John Wiley & Sons, Inc ., 9 – 30 . Google Scholar CrossRef Search ADS Shi A , Chen P , Li D , Zheng C , Zhang B , Hou A . 2009 . Pyramiding multiple genes for resistance to soybean mosaic virus in soybean using molecular markers . Molecular Breeding 23 , 113 – 124 . Google Scholar CrossRef Search ADS Siddique KH , Johansen C , Turner NC , Jeuffroy M-H , Hashem A , Sakar D , Gan Y , Alghamdi S . 2012 . Innovations in agronomy for food legumes. A review . Agronomy for Sustainable Development 32 , 45 – 64 . Google Scholar CrossRef Search ADS Sinclair TR , Marrou H , Soltani A , Vadez V , Chandolu KC . 2014 . Soybean production potential in Africa . Global Food Security 3 , 31 – 40 . Google Scholar CrossRef Search ADS Sinclair TR , Messina CD , Beatty A , Samples M . 2010 . Assessment across the United States of the benefits of altered soybean drought traits . Agronomy Journal 102 , 475 – 482 . Google Scholar CrossRef Search ADS Sinclair TR , Vadez V . 2012 . The future of grain legumes in cropping systems . Crop and Pasture Science 63 , 501 – 512 . Google Scholar CrossRef Search ADS Singh VK , Khan AW , Jaganathan D et al. 2016 . QTL-seq for rapid identification of candidate genes for 100-seed weight and root/total plant dry weight ratio under rainfed conditions in chickpea . Plant Biotechnology Journal 14 , 2110 – 2119 . Google Scholar CrossRef Search ADS Singh VK , Khan AW , Saxena RK et al. 2017 . Indel-seq: a fast-forward genetics approach for identification of trait-associated putative candidate genomic regions and its application in pigeonpea (Cajanus cajan) . Plant Biotechnology Journal 15 , 906 – 914 . Google Scholar CrossRef Search ADS Singh T , Sharma A . 2016 . Early generation selection for yield and its related traits in soybean [Glycine max (L.) Merrill.] . Legume Research 39 , 343 – 348 . Sita K , Sehgal A , HanumanthaRao B et al. 2017 . Food legumes and rising temperatures: effects, adaptive functional mechanisms specific to reproductive growth stage and strategies to improve heat tolerance . Frontiers in Plant Science 8 , 1658 . Google Scholar CrossRef Search ADS Skirycz A , Vandenbroucke K , Clauw P et al. 2011 . Survival and growth of Arabidopsis plants given limited water are not equal . Nature Biotechnology 29 , 212 – 214 . Google Scholar CrossRef Search ADS Smyth SJ . 2017 . Canadian regulatory perspectives on genome engineered crops . GM Crops & Food 8 , 35 – 43 . Google Scholar CrossRef Search ADS Soltani A , Sinclair TR . 2011 . A simple model for chickpea development, growth and yield . Field Crops Research 124 , 252 – 260 . Google Scholar CrossRef Search ADS Soltani A , Sinclair TR . 2012 . Modeling physiology of crop development, growth and yield . Wallingford, UK : CABI . Google Scholar CrossRef Search ADS Song Q , Hyten DL , Jia G , Quigley CV , Fickus EW , Nelson RL , Cregan PB . 2013 . Development and evaluation of SoySNP50K, a high-density genotyping array for soybean . PLoS One 8 , e54985 . Google Scholar CrossRef Search ADS Song Q , Jia G , Hyten DL et al. 2015 . SNP assay development for linkage map construction, anchoring whole-genome sequence, and other genetic and genomic applications in common bean . G3: Genes, Genomes, Genetics 5 , 2285 – 2290 . Sperling L , McGuire S . 2010 . Understanding and strengthening informal seed markets . Experimental Agriculture 46 , 119 – 136 . Google Scholar CrossRef Search ADS Sprink T , Eriksson D , Schiemann J , Hartung F . 2016 . Regulatory hurdles for genome editing: process- vs. product-based approaches in different regulatory contexts . Plant Cell Reports 35 , 1493 – 1506 . Google Scholar CrossRef Search ADS Stagnari F , Maggio A , Galieni A , Pisante M . 2017 . Multiple benefits of legumes for agriculture sustainability: an overview . Chemical and Biological Technologies in Agriculture 4 , 2 . Google Scholar CrossRef Search ADS Stavely JR . 2000 . Pyramiding rust and viral resistance genes using traditional and marker techniques in common bean . Annual Reports Bean Improvement Cooperatives 43 , 1 – 4 . Sun X , Hu Z , Chen R , Jiang Q , Song G , Zhang H , Xi Y . 2015 . Targeted mutagenesis in soybean using the CRISPR-Cas9 system . Scientific Reports 5 , 10342 . Google Scholar CrossRef Search ADS Taran B , Buchwaldt L , Tullu A , Banniza S , Warkentin TD , Vandenberg A . 2003 . Using molecular markers to pyramid genes for resistance to ascochyta blight and anthracnose in lentil (Lens culinaris Medik.) . Euphytica 134 , 223 – 230 . Google Scholar CrossRef Search ADS Tardieu F , Cabrera-Bosquet L , Pridmore T , Bennett M . 2017 . Plant phenomics, from sensors to knowledge . Current Biology 27 , R770 – R783 . Google Scholar CrossRef Search ADS Tardieu F , Tuberosa R . 2010 . Dissection and modelling of abiotic stress tolerance in plants . Current Opinion in Plant Biology 13 , 206 – 212 . Google Scholar CrossRef Search ADS Tayeh N , Aluome C , Falque M et al. 2015a. Development of two major resources for pea genomics: the GenoPea 13.2K SNP array and a high-density, high-resolution consensus genetic map . The Plant Journal 84 , 1257 – 1273 . Google Scholar CrossRef Search ADS Tayeh N , Aubert G , Pilet-Nayel ML , Lejeune-Hénaut I , Warkentin TD , Burstin J . 2015b. Genomic tools in pea breeding programs: status and perspectives . Frontiers in Plant Science 6 , 1037 . Tayeh N , Klein A , Le Paslier MC et al. 2015c. Genomic prediction in pea: effect of marker density and training population size and composition on prediction accuracy . Frontiers in Plant Science 6 , 941 . Teixeira EI , Brown HE , Sharp J , Meenken ED , Ewert F . 2015 . Evaluating methods to simulate crop rotations for climate impact assessments—a case study on the Canterbury plains of New Zealand . Environmental Modelling & Software 72 , 304 – 313 . Google Scholar CrossRef Search ADS Tekwa IJ , Ijabula ST , Maijama’a NP . 2010 . Effect of herbicides, seed dressing chemicals and spray regimes on germination, insect infestation and yield of cowpea (Vigna unguiculata (L) Walp) . Australian Journal of Agricultural Engineering 1 , 14 – 17 . Thudi M , Chitikineni A , Liu X et al. 2016a. Recent breeding programs enhanced genetic diversity in both desi and kabuli varieties of chickpea (Cicer arietinum L.) . Scientific Reports 6 , 38636 . Google Scholar CrossRef Search ADS Thudi M , Khan AW , Kumar V , Gaur PM , Katta K , Garg V , Roorkiwal M , Samineni S , Varshney RK . 2016b. Whole genome re-sequencing reveals genome-wide variations among parental lines of 16 mapping populations in chickpea (Cicer arietinum L.) . BMC Plant Biology 16 Suppl 1 , 10 . Google Scholar CrossRef Search ADS Tisné S , Serrand Y , Bach L et al. 2013 . Phenoscope: an automated large-scale phenotyping platform offering high spatial homogeneity . The Plant Journal 74 , 534 – 544 . Google Scholar CrossRef Search ADS Trachsel S , Kaeppler SM , Brown KM , Lynch JP . 2011 . Shovelomics: High throughput phenotyping of maize (Zea mays L.) root architecture in the field . Plant Soil 341 , 75 – 87 . Google Scholar CrossRef Search ADS Vadez V . 2014 . Root hydraulics: the forgotten side of root in drought adaptation . Field Crops Research 165 , 15 – 24 . Google Scholar CrossRef Search ADS Vadez V , Halilou O , Hissene HM , Sibiry-Traore P , Sinclair TR , Soltani A . 2017 . Mapping water stress incidence and intensity, optimal plant populations, and cultivar duration for African groundnut productivity enhancement . Frontiers in Plant Science 8 , 432 . Google Scholar CrossRef Search ADS Vadez V , Kholová J , Hummel G , Zhokhavets U , Gupta SK , Hash CT . 2015 . LeasyScan: a novel concept combining 3D imaging and lysimetry for high-throughput phenotyping of traits controlling plant water budget . Journal of Experimental Botany 66 , 5581 – 5593 . Google Scholar CrossRef Search ADS Vadez V , Kholova J , Zaman-Allah M , Belko N . 2013a. Water: the most important ‘molecular’ component of water stress tolerance research . Functional Plant Biology 40 , 1310 – 1322 . Google Scholar CrossRef Search ADS Vadez V , Soltani A , Sinclair TR . 2012 . Modelling possible benefits of root related traits to enhance terminal drought adaptation of chickpea . Field Crops Research 137 , 108 – 115 . Google Scholar CrossRef Search ADS Vadez V , Soltani A , Sinclair TR . 2013b. Crop simulation analysis of phenological adaptation of chickpea to different latitudes of India . Field Crops Research 146 , 1 – 9 . Google Scholar CrossRef Search ADS Valdisser PAMR , Pereira WJ , Almeida Filho JE et al. 2017 . In-depth genome characterization of a Brazilian common bean core collection using DArTseq high-density SNP genotyping . BMC Genomics 18 , 423 . Google Scholar CrossRef Search ADS Valliyodan B , Dan Qiu , Patil G et al. 2016 . Landscape of genomic diversity and trait discovery in soybean . Scientific Reports 6 , 23598 . Google Scholar CrossRef Search ADS Van Raden PM , Van Tassell CP , Wiggans GR , Sonstegard TS , Schnabel RD , Taylor JF , Schenkel FS . 2009 . Reliability of genomic predictions for North American Holstein bulls . Journal of Dairy Science 92 , 16 – 24 . Google Scholar CrossRef Search ADS Varshney RK . 2016 . Exciting journey of 10 years from genomes to fields and markets: some success stories of genomics-assisted breeding in chickpea, pigeonpea and groundnut . Plant Science 242 , 98 – 107 . Google Scholar CrossRef Search ADS Varshney RK , Chen W , Li Y et al. 2012a. Draft genome sequence of pigeonpea (Cajanus cajan), an orphan legume crop of resource-poor farmers . Nature Biotechnology 30 , 83 – 89 . Google Scholar CrossRef Search ADS Varshney RK , Gaur PM , Chamarthi SK , Krishnamurthy L , Tripathi S , Kashiwagi J , Samineni S , Singh VK , Thudi M , Jaganathan D . 2013a. Fast-track introgression of “QTL-hotspot” for root traits and other drought tolerance traits in JG11, an elite and leading variety of chickpea . The Plant Genome 6 , 1 – 9 . Varshney RK , Glaszmann JC , Leung H , Ribaut JM . 2010 . More genomic resources for less-studied crops . Trends in Biotechnology 28 , 452 – 460 . Google Scholar CrossRef Search ADS Varshney RK , Graner A , Sorrells ME . 2005 . Genomics-assisted breeding for crop improvement . Trends in Plant Science 10 , 621 – 630 . Google Scholar CrossRef Search ADS Varshney RK , Hoisington DA , Tyagi AK . 2006 . Advances in cereal genomics and applications in crop breeding . Trends in Biotechnology 24 , 490 – 499 . Google Scholar CrossRef Search ADS Varshney RK , Kudapa H , Pazhamala L et al. 2015 . Translational genomics in agriculture: some examples in grain legumes . Critical Reviews in Plant Sciences 34 , 169 – 194 . Google Scholar CrossRef Search ADS Varshney RK , Mohan SM , Gaur PM et al. 2013b. Marker-assisted backcrossing to introgress resistance to Fusarium Wilt Race 1 and Ascochyta blight in C 214, an elite cultivar of chickpea . The Plant Genome 7 , 1 – 11 . Varshney RK , Pandey MK , Janila P , Nigam SN , Sudini H , Gowda MV , Sriswathi M , Radhakrishnan T , Manohar SS , Nagesh P . 2014 . Marker-assisted introgression of a QTL region to improve rust resistance in three elite and popular varieties of peanut (Arachis hypogaea L.) . Theoretical and Applied Genetics 127 , 1771 – 1781 . Google Scholar CrossRef Search ADS Varshney RK , Ribaut JM , Buckler ES , Tuberosa R , Rafalski JA , Langridge P . 2012b. Can genomics boost productivity of orphan crops ? Nature Biotechnology 30 , 1172 – 1176 . Google Scholar CrossRef Search ADS Varshney RK , Roorkiwal M , Nguyen HT . 2013c. Legume genomics: from genomic resources to molecular breeding . The Plant Genome 6 , 1 – 7 . Varshney RK , Saxena RK , Upadhyaya HD et al. 2017 . Whole-genome resequencing of 292 pigeonpea accessions identifies genomic regions associated with domestication and agronomic traits . Nature Genetics 49 , 1082 – 1088 . Google Scholar CrossRef Search ADS Varshney RK , Singh VK , Hickey J , Xun X , Marshall DF , Wang J , Edwards D , Ribaut J . 2016 . Analytical and decision support tools for genomics-assisted breeding . Trends in Plant Science 15 , S1360 – S1385 . Varshney RK , Song C , Saxena RK et al. 2013d. Draft genome sequence of chickpea (Cicer arietinum) provides a resource for trait improvement . Nature Biotechnology 31 , 240 – 246 . Google Scholar CrossRef Search ADS Visscher PM , Brown MA , McCarthy MI , Yang J . 2012 . Five years of GWAS discovery . American Journal of Human Genetics 90 , 7 – 24 . Google Scholar CrossRef Search ADS Waliyar F , Kumar PL , Ntare BR , Diallo AT . 2007 . A century of research on groundnut rosette disease and its management. Information Bulletin no. 75 . Patancheru, Andhra Pradesh, India : International Crops Research Institute for the Semi-Arid Tropics . Wang L , Wang L , Tan Q , Fan Q , Zhu H , Hong Z , Zhang Z , Duanmu D . 2016 . Efficient inactivation of symbiotic nitrogen fixation related genes in Lotus japonicus using CRISPR-Cas9 . Frontiers in Plant Science 7 , 1333 . Wang LL , Wang LX , Zhou Y , Duanmu D . 2017 . Use of CRISPR/Cas9 for symbiotic nitrogen fixation research in legumes . Progress in Molecular Biology and Translational Science 150 , 1 – 495 . Google Scholar CrossRef Search ADS Wasson AP , Rebetzke GJ , Kirkegaard JA , Christopher J , Richards RA , Watt M . 2014 . Soil coring at multiple field environments can directly quantify variation in deep root traits to select wheat genotypes for breeding . Journal of Experimental Botany 65 , 6231 – 6249 . Google Scholar CrossRef Search ADS Wojciechowski MF , Lavin M , Sanderson MJ . 2004 . A phylogeny of legumes (Leguminosae) based on analysis of the plastid matK gene resolves many well-supported subclades within the family . American Journal of Botany 91 , 1846 – 1862 . Google Scholar CrossRef Search ADS Wray NR , Yang J , Hayes BJ , Price AL , Goddard ME , Visscher PM . 2013 . Pitfalls of predicting complex traits from SNPs . Nature Reviews Genetics 14 , 507 – 515 . Google Scholar CrossRef Search ADS Yakubu H , Buji IB , Sandabe MK . 2011 . Effects of seed-dressing fungicides on germination, nodulation, N2-fixation and yields of two groundnut varieties in semi-arid region of Nigeria . Sandabe International Journal of Applied Agricultural Research 6 , 121 – 129 . Yang K , Tianb Z , Chenc C et al. 2015 . Genome sequencing of adzuki bean (Vigna angularis) provides insight into high starch and low fat accumulation and domestication . Proceedings of the National Academy of Sciences, USA 112 , 3213 – 18 . Zaman-Allah M , Jenkinson DM , Vadez V . 2011 . A conservative pattern of water use, rather than deep or profuse rooting, is critical for the terminal drought tolerance of chickpea . Journal of Experimental Botany 62 , 4239 – 4252 . Google Scholar CrossRef Search ADS Zhang C , Tar’an B , Warkentin T , Tullu A , Bett KE , Vandenberg B , Somers D . 2006 . Selection for lodging resistance in early generations of field pea by molecular markers . Crop Science 46 , 321 – 329 . Google Scholar CrossRef Search ADS Zhang X , Pérez-Rodríguez P , Burgueño J , Olsen M , Buckler E , Atlin G , Prasanna BM , Vargas M , San Vicente F , Crossa J . 2017 . Rapid cycling genomic selection in a multiparental tropical maize population . G3 7 , 2315 – 2326 . Google Scholar CrossRef Search ADS Zhou Z , Jiang Y , Wang Z et al. 2015a. Resequencing 302 wild and cultivated accessions identifies genes related to domestication and improvement in soybean . Nature Biotechnology 33 , 408 – 414 . Google Scholar CrossRef Search ADS Zhou L , Wang SB , Jian J , et al. 2015b. Identification of domestication-related loci associated with flowering time and seed size in soybean with the RAD-seq genotyping method . Scientific Reports 5 , 9350 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Experimental BotanyOxford University Press

Published: Mar 5, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area