Background: For more than 25 years, the golden mussel, Limnoperna fortunei, has aggressively invaded South American freshwaters, having travelled more than 5000 km upstream across 5 countries. Along the way, the golden mussel has outcompeted native species and economically harmed aquaculture, hydroelectric powers, and ship transit. We have sequenced the complete genome of the golden mussel to understand the molecular basis of its invasiveness and search for ways to control it. Findings: We assembled the 1.6-Gb genome into 20 548 scaffolds with an N50 length of 312 Kb using a hybrid and hierarchical assembly strategy from short and long DNA reads and transcriptomes. A total of 60 717 coding genes were inferred from a customized transcriptome-trained AUGUSTUS run. We also compared predicted protein sets with those of complete molluscan genomes, revealing an exacerbation of protein-binding domains in L. fortunei. Conclusions: We built one of the best bivalve genome assemblies available using a cost-effective approach using Illumina paired-end, mate-paired, and PacBio long reads. We expect that the continuous and careful annotation of L. fortunei’s Received: 13 July 2017; Revised: 5 November 2017; Accepted: 11 December 2017 The Author(s) 2017. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 2 Uliano-Silva et al. genome will contribute to the investigation of bivalve genetics, evolution, and invasiveness, as well as to the development of biotechnological tools for aquatic pest control. Keywords: Amazon; binding domain; bivalves; genomics; TLR; transposon Data Description Brazil . In this campaign, we were able to raise around USD$20 000.00 at the same time as we promoted scientific ed- The golden mussel Limnoperna fortunei is an Asian bivalve that ar- ucation and awareness in Brazil. rived in the southern part of South America about 25 years ago Here we present the first complete genome dataset for the . Research suggests that L. fortunei was introduced in South invasive bivalve Limnoperna fortunei, assembled from short and America through ballast water of ships coming from Hong Kong long DNA reads and using a hybrid and hierarchical assembly or Korea . It was found for the first time in the estuary of the strategy. This high-quality reference genome represents a sub- La Plata River in 1991 . Since then, it has moved ∼5000 km, stantial resource for further studies of genetics and evolution of invading upstream continental waters and reaching northern mussels, as well as for the development of new tools for plague parts of the continent , leaving behind a track of great eco- control. nomic impact and environmental degradation . The latest in- festation was reported in 2016 in the Sao ˜ Francisco River, one of the main rivers in the northeast of Brazil, with a 2700-km Genome sequencing in short Illumina and long PacBio riverbed that provides water to more than 14 million people. At reads Paulo Afonso, one of the main hydroelectric power plants in the Limnoperna fortunei mussels were collected from the Jacui Sao ˜ Francisco River, maintenance due to clogging of pipelines River, Porto Alegre, Rio Grande do Sul, Brazil (29 59 29.3 S and corrosion caused by the golden mussel is estimated to cost 51 16 24.0 W). Voucher specimens were housed at the zoolog- U$700 000 per year (personal communication, Mizael Gusma, ˜ ical collection (specimen number: 19 643) of the Biology Insti- ´ ˜ Chief Maintenance Engineer for Centrais Hidreletricas do Sao tute at the Universidade Federal do Rio de Janeiro, Brazil. For the Francisco [CHESF]). genome assembly, a total of 3 individuals were sampled for DNA A recent review has shown that, before arriving in South extraction from gills and to produce the 3 types of DNA libraries America, L. fortunei was already an invader in China. Originally used in this study. DNA was extracted using DNeasy Blood and from the Pearl River Basin, the golden mussel has traveled 1500 Tissue Kit (Qiagen, Hilden, Germany) to prepare libraries for Illu- km into the Yang Tse and Yellow River basins, being limited mina Nextera paired-end reads, with ∼180-bp and ∼500-bp in- further north only by the extreme natural barriers of Northern sert sizes, (ii) Illumina Nextera mate-paired reads with insert China . Today, L. fortunei is found in the Paraguaizinho River, sizes ranging from 3 to 15 Kb, and (iii) Pacific Biosciences long located only 150 km from the Teles-Pires River that belongs to reads (Table 1). Illumina libraries were sequenced, respectively, the Alto Tapajos ´ River Basin and is the first to directly connect in a HiScanSQ or HiSeq 1500 machine, and Pacific Biosciences with the Amazon River Basin . Due to its fast dispersion rates, reads were produced with the P4C6 chemistry and sequenced it is very likely that L. fortunei will reach the Amazon River Basin in 10-SMRT Cells. All Illumina reads were submitted to qual- in the near future. ity analysis with FastQC (FastQC, RRID:SCR 014583) followed by The reason why some freshwater bivalves, such as L. fortunei, trimming with Trimmomatic (Trimmomatic, RRID:SCR 011848) Dreissena polymorpha,and Corbicula uminea fl , are aggressive in- . Pacific Biosciences adaptor-free subread sequences were vaders is not fully understood. These bivalves present charac- used as input data for the genome assembly. teristics such as (i) tolerance to a wide range of environmental For transcriptome sequencing, RNA was sampled from 4 variables, (ii) short life span, (iii) early sexual maturation, and tissues (gills, adductor muscle, digestive gland, and foot) of 3 (iv) high reproductive rates that allow them to reach densities −2 different golden mussel specimens. RNA was extracted using as high as 150 000 ind.m over a year [7, 8] that may explain the the NEXTflex Rapid Directional RNA-Seq Kit (Bio Scientifics, aggressive behavior. On the other hand, these traits are not ex- TX, USA) and 12 barcodes from NEXTflex Barcodes compatible clusive to invasive freshwater bivalves and do not explain how with Illumina NexSeq Machine. Resulting reads (Supplementary they outcompete native species and disperse so widely. Table S1) were submitted to FastQC quality analysis and To the best of our knowledge, there are no reports of strate- trimmed with Trimmomatic for all NEXTflex adaptors and bar- gies successful at controlling the expansion of mussel invasion codes. A total of 3 sets of de novo assembled transcriptomes were in industrial facilities. Bivalves can sense chemicals in the wa- generated using Trinity (Trinity, RRID:SCR 013048)(Table 2); 1 set ter and close their valves as a defensive response , making for each specimen was a pool of the 4 tissue samples to avoid as- them tolerant to a wide range of chemical substances, including sembly bias due to intraspecific polymorphism [ 16]. strong oxidants like chlorine . Microencapsulated chemicals have shown better results in controlling mussel populations in closed environments [10, 11], but it is unlikely they would work Genome assembly using a hybrid and hierarchical in the wild. Currently, there is no effective and efficient approach strategy to control the invasion by L. fortunei. The genome sequence is one of the most relevant and in- Jellyfish software (Jellyfish, RRID:SCR 005491) was used to formative descriptions of species biology. The genetic substrate count and determine the distribution frequency of lengths 25 of invasive populations, upon which natural selection operates, and 31 kmers (Fig. 1) for the Illumina DNA paired-end and mate- can be of primary importance to understanding and controlling paired reads (Table 1). The genome size was estimated to be a biological invader [12, 13]. 1.6 Gb by using the 25-kmer distribution plot as total kmer We have partially funded the golden mussel genome number and then subtracting erroneous reads (starting kmer sequencing through a pioneer crowdfunding initiative in counts from ×12 coverage) to further divide by the homozygous Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Limnoperna fortunei hybrid genome 3 Table 1: DNA reads produced for L. fortunei genome assembly Library technology Raw data Trimmed data Reads insert size Pairs Number of reads Number of bases Number of reads Number of bases Illumina Nextera Paired-end – 180 bp R1 209 542 721 21 060 365 702 209 036 571 21 001 101 404 R2 209 542 721 21 049 308 698 209 036 571 20 991 650 008 Paired-end – 500 bp R1 153 948 902 15 472 966 961 153 482 290 15 423 123 500 R2 153 948 902 15 462 883 157 153 482 290 15 414 813 589 Mate-paired 3 – 12 Kb R1 178 392 944 18 017 687 344 58 157 933 5 822 572 152 R2 178 392 944 18 017 687 344 58 157 933 5 811 310 412 Pacific Biosciences P4C – 10/SMTRC Subreads 1 663 730 11 171 487 485 Trimmomatic parameters for Illumina reads—ILLUMINACLIP: NexteraPE-PE.fa:2:30:10 SLIDINGWINDOW:4:2 LEADING:10 TRAILING:10 CROP:101 HEADCROP:0 MINLEN:80. Table 2: Trinity assembled transcripts used in the assembly and annotation of L. fortunei genome Number of reads Number of Number of Average GC Sample Pooled tissues prior assembly trinity transcripts trinity genes contig length % Mussel 1 Gills, mantle, digestive gland, foot 406 589 144 433 197 303 172 854 34 Mussel 2 Gills, mantle, digestive gland, foot 376 577 660 435 054 298 117 824 34 Mussel 3 Gills, mantle, digestive gland, foot 334 316 116 499 392 351 649 844 34 the resulting contigs were assembled into scaffolds using Pacific Biosciences long subreads data and the PacBio- correction-free assembly algorithm DBG2OLC withpa- rameters LD1 0 k 17 KmerCovTh 10 MinOverlap 20 AdaptiveTh 0.01. Finally, (iii) resulting scaffolds were submitted to 6 iter- ative runs of the program L RNA Scaffolder , which uses exon distance information from de novo assembled transcripts (Table 2) to fill gaps and connect scaffolds whenever appropri- ate. At the end, (iv) the final genome scaffolds were corrected for Illumina and Pacific Biosciences sequencing errors with the software PILON : All DNA and RNA short Illumina reads were re-aligned back to the genome with BWA aligner (BWA, RRID:SCR 010910), and resulting SAM files were BAM-converted, sorted, and indexed with the SAMTOOLs package (SAMTOOLS, RRID:SCR 002105). Pilon identifies Figure 1: Kmer distribution of Limnoperna fortunei Illumina DNA reads (Table 1). INDELS and mismatches by coverage of reads and yields a final corrected genome draft. Pilon was run with parameters –diploid –duplicates. coverage-peak depth (×45 coverage), as performed by Li et al. The final genome was assembled in 20 548 scaffolds, with an (2010) . A double-peak kmer distribution was used as evi- N50 of 312 Kb and a total assembly length of 1.6 Gb (Table 3). denceofgenomediploidy(Fig. 1) and high heterozygosity. The The golden mussel genome presents 81% of all Benchmark- rate of heterozygosity was estimated to be 2.3%, and it was cal- ing Universal Single Copy Orthologs (BUSCO version 3.3 analysis culated as described by Vij et al. (2016) , using as input data with Metazoa database; BUSCO, RRID:SCR 015008)(Table 4)and, the 17-kmer distribution plot for reads from 1 unique specimen. compared with the mollusk genomes currently available [29–36], Initially, we attempted to assemble the golden mus- it represents one of the best assemblies of molluscan genomes sel genome using only short Illumina reads of different so far also in terms of scaffold N50 and contiguity (Table 5). insert sizes (paired-end and mate-paired) (Table 1) using One main challenges of assembling bivalve genomes lies traditional de novo assembly software such as ALLPATHS in the high heterozygosity and amount of repetitive elements (ALLPATHS-LG, RRID:SCR 010742), SOAPdenovo (SOAP- these organisms present: (i) the mussels L. fortunei and Modiolus denovo, RRID:SCR 010752), and MaSuRCA (MaSuRCA, philippinarum and the oyster Crassostrea gigas genomes were es- RRID:SCR 010691). All these attempts resulted in very timated to have heterozygosity rates of 2.3%, 2.02%, and 1.95%, fragmented genome drafts, with an N50 no higher than 5 Kb respectively, which are substantially higher than other animal and a total of 4 million scaffolds. To reduce fragmentation, genomes , and (ii) repetitive elements correspond to at least we further sequenced additional long reads (10 PacBio SMTR 30% of the genomes of all studied bivalves so far (Table 3)[29–32, Cells) (Table 1) and performed a hybrid and hierarchical de novo 34–36]. Also, retroelements might be active in some species such assembly, described below and depicted in Fig. 2. as L. fortunei (refer to the “Retroelements” section of this pa- First, (i) trimmed paired-end and mate-paired DNA Illumina per) and C. gigas , allowing genome rearrangements that may reads (Table 1) were assembled into contigs using the software hinder genome assembly. One exception seems to be the deep- Sparse Assembler  with parameters LD 0 NodeCovTh 1 sea mussel B. platifrons, which has lower heterozygosity rates EdgeCovTh 0 k 31 g 15 PathCovTh 100 GS 1 800 000 000. Next, (ii) Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 4 Uliano-Silva et al. Figure 2: Hierarchical assembly strategy employed for the golden mussel genome assembly. Trimmed Illumina reads were assembled to the level of contigs with the Sparse Assembler algorithm (Step 1). Then, Illumina contigs and PacBio reads were used to build scaffolds with the DBG2OLC assembler, which anchors Illumina contigs to erroneous PacBio subreads, correcting them and building longer scaffolds (Step 2), followed by transcriptome joining scaffolds using L RNA scaffolder (Step 3). Final scaffolds were corrected by re-aligning all Illumina DNA and RNA-seq reads back to them and calling consensus with Pilon software (Step 4). Inboldisthe bioinformatics software used in each step. Red blocks indicate PacBio errors, which are represented by insertions and/or deletions, found in approximately 12% of PacBio subreads. Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Limnoperna fortunei hybrid genome 5 Table 3: Assembly statistics for Limnoperna fortunei’s genome Parameter Value Estimated genome size by kmer analysis, Gb 1.6 Total size of assembled genome, Gb 1.673 Number of scaffolds 20 548 Number of contigs 61 093 Scaffold N50, Kb 312 Maximum scaffold length, Mb 2.72 Percentage of genome in scaffolds >50 Kb 82.55 Masked percentage of total genome 33 Mapping percentage of Illumina reads back to scaffolds 91 Table 4: Summary statistics of BUSCO analysis for L. fortunei genome run for Metazoans Categories Number of genes Percentage Total BUSCO groups searched 978 – Complete BUSCOs 801 81.9 Complete and single-copy BUSCOs 769 78.62 Complete and duplicated BUSCOs 32 3.27 Fragmented BUSCOs 72 7.36 Missing BUSCOs 105 10.73 compared with other bivalves . Sun et al.  suggested that it might be due to recurrent population bottlenecks that hap- pened after events of population extinction and recolonization in the extreme environment . Nevertheless, most of the bi- valve genome projects relying only on short Illumina reads are likely to present fragmented initial drafts [29, 31]. PacBio long reads allowed us to increase the N50 to 32 Kb and to reduce the number of scaffolds from millions to 61 102, using the DBG2OLC  assembler. Finally, interactive runs of L RNA scaffolder  using the transcriptomes (Table 2) rendered the final result of N50 312 Kb in 20 548 scaffolds. It is important to note that as- sembly statistics can perform better for genomes assembled with reads generated with DNA extracted from 1 unique individ- ual. This, however, was not possible for L. fortunei’s genome due to the high amount of high-quality DNA necessary to produce Illumina mate-pairs and PacBio long reads. In this study, the challenge of assembling the high polymorphic regions between haplotypes was enhanced by the difficulties of assembling reads that originated from highly polymorphic regions across indi- viduals. However, the golden mussel assembly presented here shows that the use of Illumina contigs, low coverage of PacBio long reads, and transcriptome and Illumina re-mapping for final correction (Fig. 2) represent an option for cost-efficient assem- bly of highly heterozygous genomes of nonmodel species such as bivalves. Around 10% of repetitive elements are transposons Initial masking of L. fortunei genome was done using the Re- peatMasker program (RepeatMasker, RRID:SCR 012954)with the parameter -species bivalves and masked 3.4% of the total genome. This content was much lower than the masked por- tion of other molluscan genomes, 34% in C. gigas  and 36% in M. galloprovincialis , suggesting that the fast evolution of interspersed elements limits the use of repeat libraries from di- vergent taxa . Thus, we generated a de novo repeat library for L. fortunei using the program RepeatModeler (RepeatModeler, RRID:SCR 015027) and its integrated tools RECON , TRF Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Table 5: Comparison of genome assembly statistics for molluscan genomes Haliotis Lottia Aplysia Ruditapes Patinopecten Crassostrea Mytillus Bathymodiolus Modiolus Limnoperna discus hannai gigantea californica philippinarum yessoensis gigas Pinctadafucata galloprovincialis platifrons philippinarum fortunei Estimated genome size 1.65 Gb 359.5 Mb 1.8 Gb 1.37 Gb 1.43 Gb 545 Mb 1.15 Gb 1.6 Gb 1.64 Gb 2.38 Gb 1.6 Gb Number of scaffolds 80 032 4475 8766 223 851 82 731 11 969 7997 1746 447 65 664 74 575 20 548 Total size of scaffolds 1 865 475 499 359 512 207 715 791 924 2 561 070 351 987 685 017 558 601 156 915 721 316 1 599 211 957 1 659 280 971 2 629 649 654 1 673 125 894 Longest scaffold 2 207 537 9 386 848 1 784 514 572 939 7 498 238 1 964 558 5 897 787 67 529 2 790 175 715 382 2 720 304 Shortest scaffold 854 1000 5001 500 200 100 1807 100 292 205 558 Number of scaffolds >1 79 923 (99.9) 4471 (99.9) 8766 (100) 138 771 (61.9) 16 004 (19.3) 5788 (48.4) 7997 (100) 393 685 (22.5) 38 704 (58.9) 44 921 (60.2) 20 547 (100) Knt(%) Number of scaffolds >1 67 (0.1) 98 (2.2) 27 (0.3) 0 (0.0) 248 (0.3) 60 (0.5) 27 (0.3) 0 (0.0) 164 (0.2) 0 (0) 95 (0.5) Mnt(%) Mean scaffold size 23 309 80 338 81 655 11 441 11 939 46 671 114 508 916 25 269 35 262 81 425 Median scaffold size 1697 3622 13 763 1327 362 824 14 683 258 1284 13 722 22 134 N50 scaffold length 200 099 1 870 055 264 327 48 447 803 631 401 319 345 846 2651 343 373 100 161 312 020 Sequencing coverage ×322 ×8.87 ×11 ×39.7 ×297 ×155 ×234 ×32 ×319 ×209.5 ×60 Sequencing Technology Illumina + PacBio Sanger Sanger Illumina Illumina Illumina Illumina + BACs Illumina Illumina Illumina Illumina + PacBio 6 Uliano-Silva et al. Table 6: Summary of gene annotation against various databases for , and RepeatScout . This de novo repeat library was the L. fortunei whole-genome-predicted genes input to RepeatMasker, together with the first masked genome draft of L. fortunei, and resulted in a final masking of 33.4% of Total number of genes 60 717 the genome. Even though more than 90% of the repeats were Total number of exons 220 058 not classified by RepeatMasker (Supplementary Table S2), 8.85% Total number of proteins 60 717 of the repeats were classified as LINEs, Class I transposable ele- Average protein size, aa 304 ments. In addition, large numbers of reverse-transcriptases (824 Number of protein BLAST hits with Uniprot 26 198 counts, Pfam RVT 1 PF00078), transposases (177 counts, Pfam ∗ Number of protein BLAST hits with NR NCBI (no 14 810 HTH Tnp Tc3 2 PF01498), integrases (501 counts, Pfam Retrovi- hits with Uniprot) ral integrase core domain PF00665), and other related elements Number of protein HMMER hits with Pfam.A 24 513 were detected; more than 98% of these had detectable tran- Number with proteins with KO assigned by KEGG 8387 scripts. Number of proteins with BLAST hits with EggNOG 36 868 All considered hits had a minimum e-value of 1e-05. More than 30 000 sequences were identified by gene prediction and automated annotation analyzed. A total of 6337 ortholog groups are shared among the To annotate the golden mussel genome, we sequenced a num- 5 bivalve species. ber of transcriptomes (Table S1), de novo assembled (Table 2)and Of all the orthologs found for the total 10 species, 44 groups aligned these transcriptomes to the genome scaffolds, and cre- are composed of single-copy orthologs containing 1 represen- ated gene models with the PASA pipeline . These models tative protein sequence of each species. These sequences were were used to train and run the ab initio gene predictor AUGUS- used to reconstruct a phylogeny: the single-copy ortholog se- TUS (Augustus: Gene Prediction, RRID:SCR 008417) (Supplemen- quences were concatenated and aligned with CLUSTALW , tary Fig. S1) . The complete gene models yielded by PASA with a resulting alignment 30 755 sites in length (Fig. 3B).  were BLASTed (e-value 1e-20) against the Uniprot database ProtTest 3.4.2  was used to estimate the best-fitting substitu- (UniProt, RRID:SCR 002380), and those with 90% or more of their tion model, which was VT+G+I+F. With this alignment and sequences showing in the BLAST hit alignment were consid- model, we reconstructed the phylogeny using PhyML and ered for further analysis. Next, all the necessary filters to run 100 bootstrap repetition; the resulting tree is shown in Fig. 3B. an AUGUSTUS  personalized training were performed: (i) only gene models with more than 3 exons were maintained, Protein domain analysis shows expansion of binding domain in L. (ii) sequences with 90% or more overlap were withdrawn and fortunei only the longest sequences were retained, and (iii) only gene We performed a quantitative comparison of protein domains models free of repeat regions, as indicated by BLASTN similar- predicted from whole-genome projects of 10 molluscan species. ity searches with de novo library of repeats, were maintained. The complete protein sets of M. galloprovincialis, M. philippinarum, These curated data yielded a final set of 1721 gene models on B. platifrons, Ruditapes philippinarum, Patinopecten yessoensis, C. gi- which AUGUSTUS  was trained in order to predict genes in gas, Pinctada fucata, Lottia gigantean, and Haliotis discus hannai the genome using the default AUGUSTUS  parameters. Once (Supplementary Table S3) were submitted to domain annota- the gene models were predicted, a final step was performed tion using HMMER against the Pfam-A database (e-value 1e-05). by using the PASA pipeline  once again in the update mode Protein expansions in L. fortunei were rendered using the nor- (parameters -c -A -g -t). This final step compared the 55 638 malized Pfam count value (average) obtained from the other 9 gene models predicted by AUGUSTUS  with the 40 780 ini- mollusks, according to a model based on the Poisson cumula- tial transcript-based gene-structure models from PASA to tive distribution. Bonferroni correction (P ≤ 0.05) was applied for generate the final set of 60 717 gene models for L. fortunei. Of false discovery, and absolute frequencies of Pfam-assigned do- those, 58% had transcriptional evidence based on RNA Illumina mains were initially normalized by the total count number of reads (Table S2) re-mapping, a rate that was expected as our Pfam-assigned domains found in L. fortunei to compensate for RNA-Seq libraries were constructed for only 4 tissues of adult discrepancies in genome size and annotation bias. golden mussel specimens without any environmental stress in- For L. fortunei, the annotation against Pfam-A classified duction (Table 2). Therefore, these libraries lack transcripts for 40 127 domains in 24 513 gene models, of which 83 and 67 developmental stages for some other cell types (i.e., hemocytes) were expanded or contracted, respectively, in comparison with and stress-inducible genes. Finally, 67% of the gene models were the other mollusks (Fig. 4A; Supplementary Table S4 and S5). annotated by homology searches against Uniprot or NCBI NR The 83 overrepresented domains were further analyzed for (Table 6). functional enrichment using domain-centric Gene Ontology (Fig. 4B). The analysis shows a prominent expansion of binding Protein clustering indicates evolutionary proximity among mollusk domains in L. fortunei, such as Thrombospondin (TSP 1), Colla- species gen, Immunoglobulins (Ig, I-set, Izumo-Ig Ig 3), and Ankyrins Gene family relationships were assigned using reciprocal best (Ank 2, Ank 3, and Ank 4). These repeats have a variety of BLAST and OrthoMCL software (version 1.4)  between L. binding properties and are involved in cell-cell, protein-protein, fortunei proteins and the total protein set predicted for 9 and receptor-ligand interactions driving the evolutionary im- other mollusks: the mussels M. galloprovincialis, M. philippinarum, provement of complex tissues and the immune defense system and B. platifrons,the clam Ruditapes philippinarum, the scallop in metazoans [50–54]. An evolutionary pressure toward the Patinopecten yessoensis, the pacific oyster C. gigas, the pearl oyster development of a diversified innate immune system is also Pinctada fucata (genome version from Du et al. ), and the gas- suggested by the high amount of leucine rich repeats (LRR) and tropods Lottia gigantea and Haliotis discus hannai (see Supplemen- Toll/interleukin-1 receptor homology domains (TIR). Death, an- tary Table S3 for detailed information on the comparative data). other over-represented Pfam, is also part of TLR signaling, being Figure 3A presents orthologs relationships for 5 of the bivalves present in several docking proteins such as Myd88, Irak4, and Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Limnoperna fortunei hybrid genome 7 Figure 3: (A) Gene family assigned with OrthoMCL for the total set of proteins predicted from 5 mussel genome projects. Outside the Venn diagram, the species nameis represented, and below it is the number of proteins/number of clustered proteins/number of clusters. (B) Phylogeny of the concatenated dataset using 44 single-copy orthologs extracted from 10 molluscan genomes. The VT model was estimated to be the best-fitting substitution model with ProtTest 3.4.2. We reconstru cted the phylogeny using PhyML and 100 bootstrap repetition. Pelle . Interestingly, BLAST analysis of L. fortunei gene models were funded through a crowdfunding initiative in Brazil. This against Uniprot identified 2 types of TLRs whose prototypical genome contains valuable information for further evolutionary architecture of N-terminal extracellular LRR motifs including studies of bivalves and metazoa in general. Additionally, our either a single or multiple cysteine cluster domain, a C-terminal team will further search for the presence of proteins of biotech- TIR domain spaced by a single transmembrane-spanning do- nology interest such as the adhesive proteins produced by the main , could be correctly identified using the Simple Modular foot gland that we have described elsewhere  or genes related Architecture Research Tool (SMART) . Indeed, we confirmed to the reproductive system that have been shown to be very ef- 141 sequences with similarity to single cysteine clusters TLRs fective for invertebrate plague control . The golden mussel (scc) typical of vertebrates and 29 sequence hits with the genome and the predicted proteins are available for download multiple cysteine cluster TLRs (mcc) typical of Drosophila . in the GigaScience repository, and the scientific community is Phylogenetic analysis of all sequences (using PhyML , model welcome to further curate the gene predictions. JTT) (Supplementary Fig. S2) shows evidence for TLRs clade As the golden mussel advances towards the Amazon River separation in L. fortunei; the scc TLRs exhibit a higher degree of Basin, the information provided in this study may be used to amino acid changes, higher molecular evolution, and diversifi- help develop biotechnological strategies that may control the ex- cation than the mcc TLRs. Overall, the expansion of these gene pansion of this organism in both industrial facilities and open families might suggest an improved resistance to infections. It environment. is, however, equally curious that other immune-related gene families such as Fribinogen C and C1q seem to be contracted Availability of supporting data (Supplementary Table S5). This feature may depend on the evolution-driven, yet random fate of the L. fortunei genome, a Limnoperna fortunei’s genome and transcriptome data are consequence of different specific duplicate genes in other available in the Sequence Read Archive (SRA) as Bio- species. Also, other protein families involved in toxin Project PRJNA330677 and under the accession numbers metabolism, especially glutathione-based processes and SRR5188384, SRR5195098, SRR518800, SRR5195097, SRR5188315, sulfotransferases, are clearly contracted (Table S5). SRR5181514. This Whole Genome Shotgun Project has been deposited in the DDBJ/ENA/GenBank under accession number NFUK00000000. The version described in this paper is version Final considerations NFUK01000000. Supporting data, also including annotations and BUSCO results, are available via the GigaScience repository, Here we have described the first version of the golden mus- GigaDB . sel complete genome and its automated gene prediction, which Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 8 Uliano-Silva et al. Figure 4: Gene family representation analysis in the L. fortunei genome. (A) Pfam hierarchical clustering, heatmap. Features were selected according to a model based on the Poisson cumulative distribution of each Pfam count in the golden mussel genome vs the normalized average values found in the other 9 molluscan genomes (Bonferroni correction, P ≤ 0.05). Transposable elements were included in the analysis. Colors depict the log2 ratio between Pfam counts found in each single genome and the corresponding mean values. The hierarchical clustering used the average dot product for the data matrix and complete linkage for branching. Abbreviations: Bp: Bathymodioulus platifrons; Cg: Crassostrea gigas;Hd: Haliotus discus hannai; Lf: L. fortunei; Lg: Lottia gigantean;Mg: Mytilus galloprovincialis;Mp: Modioulus philippinarum; Pf: Pinctada fucata; Py: Patinopecten yessoensis; Rp: Ruditapes philippinarum. (B) Gene Ontology analysis of expanded gene families, semantic scatter plot. Shown are cluster representatives after redundancy reduction in a 2-dimensional space applying multidimensional scaling to a matrix of semantic similarities of GO terms. Color indicates the GO enrichment level (legend in upper left-hand corner); size indicates the relative frequency of each term in the UNIPROT database (larger bubbles represent less specific processes). Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Limnoperna fortunei hybrid genome 9 to the genes and proteins that we found in the genome to Additional files thank the backers for their support. The name list is available Supplementary Table S1. RNA raw reads sequenced for 3 L. for- in Supplementary Table S6. tunei specimens, 4 tissues each. Supplementary Table S2. RepeatMasker classification of re- peats predicted in the L. fortunei genome. References Supplementary Table S3. Details of the online availability of the data used for ortholog assignment and protein domain ex- 1. Pastorino G, Darrigran G, Maris MS et al. Limnoperna pansion analysis. fortunei (Dunker, 1857) (Mytilidae), nuevo bivalvo inva- Supplementary Table S4. Expanded protein families in the L. sor em aguas Del Rio de la plata. Neotropica 1993;39: fortunei genome. 101–2. Supplementary Table S5. Contracted protein families in the 2. Darrigran G. Potential impact of filter-feeding invaders on L. fortunei genome. temperate inland freshwater environments. Biol Invasions Supplementary Table S6. Fantasy names given to L. fortunei 2002;4(1/2):145–56. genes and proteins from the backers that supported us through 3. Uliano-Silva M, Fernandes F da C, Holanda IBB et al. Inva- crowdfunding (www.catarse.me/genoma). sive species as a threat to biodiversity: the golden mussel Supplementary Figure 1. Steps performed for the prediction Limnoperna fortunei approaching the Amazon River basin. In: and annotation of the L. fortunei genome. Alodi S, ed. Exploring Themes on Aquatic Toxicology. Re- Supplementary Figure 2. Phylogenetic tree of Toll-like (TLRs) search Signpost, India; 2013. receptors found in the L. fortunei genome. 4. Boltovskoy D, Correa N. Ecosystem impacts of the invasive bivalve Limnoperna fortunei (golden mussel) in South America. Hydrobiologia 2015;746(1):81–95. Abbreviations 5. Xu M. Distribution and spread of Limnoperna fortunei in China. BUSCO: Benchmarking Universal Single-Copy Orthologs; KEGG: In: Boltovskoy D, ed. Limnoperna fortunei. Cham, Switzer- Kyoto Encyclopedia of Genes and Genomes; SRA: Sequence Read land: Springer International Publishing; 2015:313–20. Archive. 6. Oliveira M, Hamilton S, Jacobi C. Forecasting the expansion of the invasive golden mussel Limnoperna fortunei in Brazil- ian and North American rivers based on its occurrence in Ethics approval the Paraguay River and Pantanal wetland of Brazil. Aquat In- Limnoperna fortunei specimens used for DNA extraction and vasions 2010;5(1):59–73. sequencing were collected in the Jacu´ıRiver (29 59 29.3 S 7. Karatayev AY, Boltovskoy D, Padilla DK, Burlakova LE. The in- 51 16 24.0 W), southern Brazil. This bivalve is an exotic species vasive bivalves Dreissena polymorpha and Limnoperna fortunei: in Brazil and is not characterized as an endangered or protected parallels, contrasts, potential spread and invasion impacts. J species. Shellfish Res 2007; 26(1):205–13. 8. Orensanz JM (Lobo), Schwindt E, Pastorino G et al. No longer the pristine confines of the world ocean: a survey of exotic Competing interests marine species in the Southwestern Atlantic. Biol Invasions The authors declare that they have no competing interests. 2002;4(1/2):115–43. 9. Claudi R, Mackie GL. Practical manual for zebra mussel moni- toring and control. Boca Raton, FL: Lewis Publishers; 1994:227 Funding 10. Calazans SHC, Americo JA, Fernandes F da C et al. Assess- ment of toxicity of dissolved and microencapsulated bio- This work was supported by the Brazilian Government agen- cies CAPES (PVE 71/2013), FAPERJ APQ1 (2014), and FAPERJ/DFG cides for control of the Golden Mussel Limnoperna fortunei. Mar Environ Res 2013 91:104–8. (39/2014). Also, this work was funded through crowdfunding with the support of 346 people (www.catarse.me/genoma). 11. Aldridge DC, Elliott P, Moggridge GD. Microencapsulated biobullets for the control of biofouling zebra mussels. Env- iron Sci Technol 2006;40(3):975–9. Author contributions 12. Cox GW. Alien species and evolution: the evolutionary ecol- ogy of exotic plants, animals, microbes, and interacting na- Conceived and designed the experiments: M.R., M.U., T.O., tive species. Washington DC: Island Press; 2004:377. C.M., F.D. Performed the experiments: M.U., J.A. Analyzed the 13. Hall MR, Kocot KM, Baughman KW et al. The crown-of- data: M.U., T.O., C.M., F.D., F.P., N.C., I.C., M.R. Contributed thorns starfish genome as a guide for biocontrol of this coral reagents/materials/analysis tools: M.R., F.P., C.M. Wrote the pa- reef pest. Nature 2017;544(7649):231–4. per: M.U., F.D., M.R. All authors read and approved the final 14. www.catarse.me/genoma. manuscript. 15. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trim- mer for Illumina sequence data. Bioinformatics 2014;1;170. Acknowledgements 16. Trapnell C, Roberts A, Goff L et al. Differential gene and transcript expression analysis of RNA-seq experiments with We thank Susan Mbedi and Kirsten Richter from BeGenDiv for TopHat and Cufflinks. Nat Protoc 2012; 7(3):562–78. RNA-Seq library preparation and sequencing. We thank Dr. Loris 17. Marc¸ais G, Kingsford C. A fast, lock-free approach for effi- Bennett for IT support while performing bioinformatics analysis. cient parallel counting of occurrences of Kmers. Bioinfor- We especially want to thank the 346 backers that sup- matics 2011;27(6):764–70. ported the sequencing of the golden mussel through 18. Li R, Fan W, Tian G et al. The sequence and de novo assembly crowdfunding in a 2013 campaign that raised U$20 000.00 of the giant panda genome. Nature 2010;463(7279):311–7. (www.catarse.me/genoma). We decided to give fantasy names Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018 10 Uliano-Silva et al. 19. Vij S, Kuhl H, Kuznetsova IS et al. Chromosomal-level 41. Benson G. Tandem repeats finder: a program to analyze DNA assembly of the asian seabass genome using long se- sequences. Nucleic Acids Res 1999;27(2):573–80. quence reads and multi-layered scaffolding. PLoS Genet 42. Price AL, Jones NC, Pevzner PA. De novo identifica- 2016;12(4):e1005954. tion of repeat families in large genomes. Bioinformatics 20. Gnerre S, Maccallum I, Przybylski D et al. High-quality draft 2005;21(Suppl 1):i351–8. assemblies of mammalian genomes from massively parallel 43. Haas BJ, Delcher AL, Mount SM et al. Improving the Ara- sequence data. Proc Natl Acad Sci U S A 2011;108(4):1513–8. bidopsis genome annotation using maximal transcript 21. Luo R, Liu B, Xie Y et al. SOAPdenovo2: an empirically im- alignment assemblies. Nucleic Acids Res 2003;31(19): proved memory-efficient short-read de novo assembler. Gi- 5654–66. gascience 2012;1(1):18. 44. Stanke M, Diekhans M, Baertsch R et al. Using native 22. Zimin AV, Marc¸ais G, Puiu D et al. The MaSuRCA genome as- and syntenically mapped cDNA alignments to improve sembler. Bioinformatics 2013;29(21):2669–77. de novo gene finding. Bioinforma Oxf Engl 2008; 24(5): 23. Ye C, Ma Z, Cannon CH et al. Exploiting sparseness in de novo 637–44. genome assembly. BMC Bioinformatics 2012;13(Suppl 6):S1. 45. Li L, Stoeckert CJ, Roos DS. OrthoMCL: identification of 24. Ye C, Hill CM, Wu S et al. DBG2OLC: efficient assembly of ortholog groups for eukaryotic genomes. Genome Res large genomes using long erroneous reads of the third gen- 2003;13(9):2178–89. eration sequencing technologies. Sci Rep 2016;6:31900. 46. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving 25. XueW,LiJ-T,Zhu Y-Petal.L˙RNA scaffolder: scaffolding the sensitivity of progressive multiple sequence alignment genomes with transcripts. BMC Genomics 2013;14(1):604. through sequence weighting, position-specific gap penalties 26. Walker BJ, Abeel T, Shea T et al. Pilon: an integrated tool for and weight matrix choice. Nucl Acids Res 1994;22(22):4673– comprehensive microbial variant detection and genome as- 80. sembly improvement. PLoS One 2014;9(11):e112963. 47. Darriba D, Taboada GL, Doallo R et al. ProtTest 3: fast selec- 27. Li H, Durbin R. Fast and accurate short read align- tion of best-fit models of protein evolution. Bioinformatics ment with Burrows-Wheeler transform. Bioinformatics 2011;27(8):1164–5. 2009;25(14):1754–60. 48. Muller ¨ T, Vingron M. Modeling amino acid replacement. J 28. Li H, Handsaker B, Wysoker A et al. The sequence Comput Biol 2000;7(6):761–76. Alignment/Map format and SAMtools. Bioinformatics 49. Guindon S, Gascuel O. A simple, fast, and accurate algorithm 2009;25(16):2078–9. to estimate large phylogenies by maximum likelihood. Syst 29. Murgarella M, Puiu D, Novoa B et al. A first insight into the Biol 2003;52(5):696–704. genome of the filter-feeder mussel Mytilus galloprovincialis. 50. Bjorklund ¨ AK, Ekman D, Elofsson A. Expansion of protein do- PLoS One 2016;11(3):e0151561. main repeats. PLoS Comp Biol 2006;2(8):e114. 30. Zhang G, Fang X, Guo X et al. The oyster genome reveals 51. Rennemeier C, Hammerschmidt S, Niemann S et al. stress adaptation and complexity of shell formation. Nature Thrombospondin-1 promotes cellular adherence of gram- 2012;490(7418):49–54. positive pathogens via recognition of peptidoglycan. FASEB J 31. Takeuchi T, Kawashima T, Koyanagi R et al. Draft genome of 2007;21(12):3118–32. the pearl oyster Pinctada fucata: a platform for understanding 52. Schmucker D, Chen B. Dscam and DSCAM: complex genes in bivalve biology. DNA Res 2012;19(2):117–30. simple animals, complex animals yet simple genes. Genes 32. Sun J, Zhang Y, Xu T et al. Adaptation to deep-sea chemosyn- Devel 2009;23(2):147–56. thetic environments as revealed by mussel genomes. Nat 53. Pancer Z, Amemiya CT, Ehrhardt GRA et al. Somatic diver- Ecol Evol 2017;1(5):0121 sification of variable lymphocyte receptors in the agnathan 33. Nam B-H, Kwak W, Kim Y-O et al. Genome sequence of pa- sea lamprey. Nature 2004;430(6996):174–80. cific abalone ( Haliotis discus hannai): the first draft genome in 54. Tucker RP. The thrombospondin type 1 repeat superfamily. family Haliotidae. Gigascience 2017;6(5):1–8. Int J Biochem Cell Biol 2004;36(6):969–74. 34. Wang S, Zhang J, Jiao W et al. Scallop genome provides in- 55. Park HH, Lo YC, Lin SC et al. The death domain superfam- sights into evolution of bilaterian karyotype and develop- ily in intracellular signaling of apoptosis and inflammation. ment. Nat Ecol Evol 2017;1(5):0120. Annu Rev Immunol 2007;25(1):561–86. 35. Mun S, Kim Y-J, Markkandan K et al. The whole-genome and 56. Leulier F, Lemaitre B. Toll-like receptors — taking an evolu- transcriptome of the manila clam (Ruditapes philippinarum). tionary approach. Nat Rev Genet 2008;3(3):165–78. Genome Biol Evol 2017;9(6):1487–98 57. Schultz J, Milpetz F, Bork P et al. SMART, a simple modular ar- 36. Du X, Fan G, Jiao Y et al. The pearl oyster Pinctada fucata chitecture research tool: identification of signaling domains. martensii genome and multi-omic analyses provide insights Proc Natl Acad Sci U S A 1998;95(11):5857–64 into biomineralization. Gigascience 2017;6(8):1–12 58. Uliano-Silva M, Americo JA, Brindeiro R et al. Gene discovery 37. Smit AF, Hubley R, Green PJ. RepeatMasker Open-3.0. 1996. through transcriptome sequencing for the invasive mussel 2010. Limnoperna fortunei. PLoS One 2014;9(7):e10297. 38. Fu H, Dooner HK. Intraspecific violation of genetic colinear- 59. Hammond A, Galizi R, Kyrou K et al. A CRISPR-Cas9 ity and its implications in maize. Proc Natl Acad Sci U S A gene drive system targeting female reproduction in the 2002;99(14):9573–8. malaria mosquito vector Anopheles gambiae. Nat Biotechnol 39. Smith AFA, Hubley R. RepeatModeler Open-1.0. 2014. 2015;34(1):78–83. http://www.repeatmasker.org. Accessed June 2016. 60. Uliano-Silva M, Dondero F, Otto TD et al. A hybrid- 40. Bao Z, Eddy SR. Automated de novo identification of re- hierarchical genome assembly strategy to sequence the peat sequence families in sequenced genomes. Genome Res invasive golden mussel, Limnoperna fortunei. GigaScience 2002;12(8):1269–76. Database 2017. http://dx.doi.org/10.5524/100386. Downloaded from https://academic.oup.com/gigascience/article-abstract/7/2/1/4750781 by Ed 'DeepDyve' Gillespie user on 16 March 2018
GigaScience – Oxford University Press
Published: Feb 1, 2018
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
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
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.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera