Abstract Background Both genome-wide association studies (GWAS) and epigenome-wide association studies (EWAS) are aiming to discover molecular signs for diseases, which possibly can be helpful for future therapeutic intervention strategies. The most prominently used tissue in association studies on humans is venous blood. In contrast to the unchangeable genotype, epigenetic DNA methylation is more variable. Methylation is affected not only by a subject’s constitution such as age, gender, ethnicity, genotype, lifestyle and health status, but is also determined by tissue-specific cell types. Methods PubMed, published before 2017, was researched, documenting the importance of epigenetic analyses on single cell types instead of whole blood in EWAS. Results Initial studies documented that stressor-induced, mostly marginal, DNA methylation changes in whole-blood samples (< 5% methylation difference) may rely not on uniform distribution of that methylation shift among each blood cell type, but on strongly altered methylation (> 20%) in single cell types. The effect size in single cell types enables the performance of epigenome-wide studies on replicated smaller cohorts, in contrast to the requirement of larger international consortium-based approaches. Conclusions Therefore, the identification of a specific cell type that is responsible for association between DNA methylation in whole blood with the phenotype of interest, has to be a prioritized experimental approach in association studies. This is a key prerequisite for constructive interpretation of epigenetic signs in the context of diverse biological function of the tissue blood, for detection of causality link between methylation and phenotype and for establishment of valuable clinical biomarkers and therapeutic targets. EWAS, blood, cell type, DNA methylation, causality, biomarker Key Messages Stressor-induced epigenetic DNA methylation changes found in frequently used tissue, such as whole blood, in EWAS are predominantly suggested to be cell type specific. Identification of cell type’s origin of altered DNA methylation is highly beneficial in finding clinical biomarkers or in disclosure of physiological/pathological processes induced by changed conditions. By the larger effect size of altered epigenetic DNA methylation in single cell types, association studies may enroll fewer subjects to reach meaningful findings. Introduction Cellular rearrangement during the differentiation process is accompanied or caused by dynamic chromatin changes that influence phenotype. In an orchestrated action, epigenetic marks such as DNA methylation at CpG residues, histone tail modifications, nucleosome remodelling and open versus closed chromatin packing are involved in these changes.1 Due to the loss of high epigenetic plasticity and ductility during the differentiation process, each terminally differentiated cell type in normal tissue is characterized by specific epigenetic signatures2,3 that are responsible for determining cell-type-specific functioning.4 Depending on cell type, these differences are more or less pronounced. In blood, DNA methylation between most frequent cell types differed from an extent of 3.5% up to 42% when data from about 450 000 pre-selected CpGs were analysed.3 However, in spite of epigenetic rigidity, dynamic chromatin changes in adaptive response to a changing microenvironment is retained in adult cells5 and may account for inter-individual variability in cell-type-specific DNA methylation.2 Moreover, due to specialized function of differentiated cell types, dynamic methylation changes in adaptation to changing external or internal conditions are expectedly more cell-type-specific than common.6–8 DNA methylation is determined by the opposing processes of methylation or demethylation. Genome-wide, changed methylations, in terms of hypomethylation, are found more frequently, especially at transcription-determining regulatory DNA sites.9 This hypomethylation basically relies on the replication-independent process of active demethylation that might influence a methylation pattern prompt within hours after induction.10 Emphasizing the cell-type-specific adaptation to a stressor, this review discusses epigenetic changes in response to external exposures and provides strong expectation for the future. For clinical purposes, the focus was directed on DNA methylation because, currently, it is inexpensive and more accurate than histone tail modification assays.11 Biological samples for EWAS Blood samples are commonly used in epidemiological studies and clinical research to obtain standardized specimens for genome-wide studies, including epigenetic DNA methylation studies, using a minimally invasive approach. It contains information about DNA methylation for blood cell types. Additionally, it may contain DNA of cells of other entities such as rare cancer cells or circulating DNA in plasma or serum from solid tumours. The identification of the accompanying foreign tissue DNA with an aberrant methylation pattern served, in particular, as an approach to provide diagnostic biomarkers for cancer.12,13 It should be emphasized that without the knowledge of the foreign cell type’s condition-specific DNA methylation pattern, the clinical diagnosis cannot be manifested. This leads to the presumption that altered DNA methylation found in whole blood, which is a response to any changed condition, can similarly be caused by methylation changes specifically altered in a single blood cell type. With respect to the tissue blood, it must be taken into account that this represents a fast renewing tissue in which the majority of different cell types stay less than 1 week in the blood stream.14 Therefore, methylation changes may have disappeared after a very short transient time. Alternatively, condition-triggered methylation changes may be present for a longer transient time or permanently. This can occur: (i) by permanent presence of the provoking condition; or (ii) as a result of stable methylation changes in long-lived specialized blood cells, such as memory lymphocytes of the adaptive immune system or ‘trained’ leukocytes of the innate immune system.15,16 Additionally, a condition may have an effect at the level of progenitor cell differentiation or may alter regulation of common lineage-specific transcription factors.17 Another practical, non-invasively obtainable, donor specimen for epidemiological studies is buccal swabs containing mainly epithelial cells from the oral mucous membrane and, to a lesser extent, leukocytes.18,19 To obtain single cell types from composed blood or buccal specimens, there are to date promising cell-separation techniques that guarantee high purity of separated cell types.20 However, due to the extended experimental duration and sample manipulation until DNA becomes extracted, cell separation techniques are not indicated for investigation of an immediate effect by acute exposure. Indeed, this is a very rare constellation for EWAS. In general, cell separation might potentially affect methylation pattern or alter methylation caused by technical differences. However, this unproven concern for influence on methylation pattern by cell separation procedure21 has not been supported by this review. Using well-validated cell separation protocols, even small methylation differences could be genuine rather than caused by technical artefacts.22 Benefit of altered epigenetic DNA methylation The concomitant plasticity and rigidity of methylation pattern, reflected by transient to permanent methylation changes in response to changing conditions, enable the discovery of valuable biomarkers or new causal relationships between provoking conditions and altered phenotype. Epigenetic biomarkers can be used to indicate time-resolved past or present exposures and present health status, or can serve to predict adverse health effects in the future. Additionally, the causal responsibility of a DNA methylation for an altered phenotype can offer the identification of new therapeutic targets. Use as a biomarker From a suggested 30 million real CpG sites in the human genome, there is a small minority of CpG sites considered to date to be valuable biomarkers established from blood samples. They apply mostly to diagnostic purposes, especially for cancer, by using mainly plasma or serum specimens12 unlike buffy coat or single cell types of the blood. There is a great challenge to robustly validate methylation at single CpG sites, associated with any human condition/disease, as a biomarker. In the EWAS literature, there is frequently an inappropriate use of the term ‘biomarker’ which does not take into consideration its statistical requirements.23–28 To fulfill criteria for a good clinically used biomarker, it is proposed: to have (i) high sensitivity; and (ii) high specificity, of at least 90% each;29 and to be (iii) accurate; and (iv) reproducible.30 A laboratory test with 95% sensitivity and specificity has been classified as an excellent test.31 However, the integrity of both specificity and sensitivity strongly depends on the prevalence of condition of interest. For instance, the possible implementation of the analyses of methylation in SEPT9 (pooled sensitivity and specificity of 0.66 and 0.91),32 a member of GTP-binding protein family, in blood samples as a predictive diagnostic biomarker assay for colorectal cancer (CRC, prevalence of 1.13% for inhabitants at age of 50 in Germany),33,34 would cause about 92% false-positive probands making use of medical health check-ups for CRC at the age of 50. Thus, the two criteria for biomarker sensitivity and specificity alone, without considering the prevalence of a condition/disease, must be treated with adequate caution. Besides sensitivity and specificity, other quality values used for a biomarker assay, such as accuracy or the positive predictive value (PPV), are all similarly influenced by the prevalence of the condition/disease and should be used with caution.31 Examples of putative epigenetic biomarkers for lifestyle stressors In the past 5 years, a great number of EWAS have been performed dealing with methylation changes in adult or cord blood by active tobacco smoking, in adulthood or by the child’s mother during pregnancy. There was seemingly no single CpG found to be valuable as a putative biomarker indicating chronic smoking behaviour. Thus, combinations of different CpGs were checked for use as a biomarker. In one study, a combination of four CpG sites using receiver-operator characteristic (ROC) curves released a sensitivity of 69% and specificity of 90% in a test sample set, and a sensitivity of 71% and specificity of 80% in a validation set.35 Another approach, based on a score composed of 29 CpG sites associated with smoking, identified current smokers in two small ethnic cohorts of men, with 100% sensitivity and 97% specificity in Europeans (n = 16) and, differently, with 80% sensitivity and 95% specificity in South Asians (n = 20), in comparison with former smokers (n = 14 and 10, respectively) and non-smokers (n = 65 and 64, respectively).36 In addition to reports on adult smokers, through the use of a methylation score in newborns for sustained maternal smoking during pregnancy, the detection of false-negative newborns increased from 20% in a training set (n = 1057) to 42% in a smaller test set (n = 221).27 Altogether, these studies on whole blood could not establish a convincing set of CpGs as a general biomarker for smoking. In contrast, by analysing methylation changes at the single CpG site cg05575921 (AHRR) in granulocytes, chronic smokers could be identified with a specificity and sensitivity both of 100%.6 Thus, identification of a biomarker seems to be more successful when considering cell-type-specific methylation changes. Examples of candidate diagnostic epigenetic biomarkers for cancer in blood The biopsy-free search for an epigenetic biomarker of cancer encompasses the search for unique epigenetic variants of the trait/cancer in blood. Progress remains modest, despite much effort in identifying prognostic clinical biomarkers for early detection of cancer. Most prominently investigated CpG sites in plasma/serum DNA are annotated to the genes SEPT9 (sensitivity, 0.66; specificity, 0,91)32 and SDC2 (sensitivity, 87.0%; specificity, 95.2%)37 for colorectal cancer, SHOX2 (sensitivity, 60%; specificity, 90%)38 for lung cancer and in cell-based DNA RASSF1A, APC, RARB, GSTP139 for prostate cancer. It remains to be seen if methylation analyses in blood cell types can be as successful in releasing a strong biomarker for cancer as for smoking. Uncovering causality of physiological/pathological processes To uncover the causal impact of condition-triggered methylation changes in blood on phenotype, cell type origin of methylation must be identified. The following depicts at least five suggested distribution patterns of methylation changes among different cell types of blood (Figure 1), illustrating different reasons for mainly minor methylation changes reported from EWAS so far on whole blood. Figure 1 View largeDownload slide Diverse reasons for altered methylation in whole blood. (A) Methylation changes are exclusively established to a single cell type. These altered methylations may be exclusively found in cases, or the frequency of cells with altered methylation differs between cases and controls. (B) The proportion of cells with a specific methylation mark (i.e. GPR15-expressing lymphoid T cells and B cells according to)7 is changed. (C) Both the proportion of a cell subtype and proportion of cells with a specific methylation mark are changed in cases. (D) Methylation changes are found commonly in all cell types. (E) Changed methylation changes are first randomly distributed and might become over-represented in progeny cell populations. G, E, N, M, T, B and NK, granulocytes, eosinophil granulocyte, neutrophil granulocyte, monocytes, T cells, B cells and natural killer cells, as surrogates for the much more diverse cell types of the blood; Me, changed methylation marks (i.e. CpG sites). Figure 1 View largeDownload slide Diverse reasons for altered methylation in whole blood. (A) Methylation changes are exclusively established to a single cell type. These altered methylations may be exclusively found in cases, or the frequency of cells with altered methylation differs between cases and controls. (B) The proportion of cells with a specific methylation mark (i.e. GPR15-expressing lymphoid T cells and B cells according to)7 is changed. (C) Both the proportion of a cell subtype and proportion of cells with a specific methylation mark are changed in cases. (D) Methylation changes are found commonly in all cell types. (E) Changed methylation changes are first randomly distributed and might become over-represented in progeny cell populations. G, E, N, M, T, B and NK, granulocytes, eosinophil granulocyte, neutrophil granulocyte, monocytes, T cells, B cells and natural killer cells, as surrogates for the much more diverse cell types of the blood; Me, changed methylation marks (i.e. CpG sites). First, methylation changes are exclusively found in a single cell type (Figure 1A). A convincing example for this was recently given, showing lineage-specific methylation differences for smoking-associated CpGs.17 Importantly, methylation changes that were detected in less frequent cell types were undetectable in whole blood or peripheral blood mononuclear cells (PBMC). Consequently, EWAS focusing on the effect of smoking on separated blood cell populations such as peripheral blood mononuclear cells23,24,40 differed as expected strongly by differences in methylation from that obtained in whole blood. In addition, the effect size of methylation change is greater in single cell types than in composed whole blood or PBMC. Second, methylation changes are found by expansion of functionally changed cell types (Figure 1B). This possibility was recently confirmed for the small tobacco smoking-induced hypomethylation of about 2% in whole blood at cg19859270. This CpG site, annotated to GPR15, was found to be hypomethylated exclusively in GPR15-expressing lymphoid cells, like T cells and B cells. Chronic tobacco smoking specifically induced the expansion of GPR15+ cells, with T cells as the major source of cg19859270 hypomethylation in whole blood.7 Another example can be given with the smoking-induced hypomethylation at cg05575921 (AHRR). This CpG site was found hypomethylated in myeloid as in lymphoid cells but with different effect size. Compared with about 21% hypomethylation in whole blood of smokers, the effect size was much greater in the myeloid granulocytes (55%) than in lymphoid T cells (< 10%).6 Third, a single cell type is both specifically expanded and/or contains a condition-induced methylation level (Figure 1C). A prominent example was provided recently, showing a partial association with eosinophil numbers for all total serum IgE-associated CpGs, together with the affected methylation level of IgE-associated CpGs, exclusively in eosinophils especially in asthmatic subjects. This suggests an enrichment of a distinctive eosinophil subset in atopic asthma.8 In the same manner, analysing methylation in different blood cell types revealed that methylation changes were dissimilarly distributed among cell subtypes.41,42 Fourth, alteration in methylation by changed condition is common in all cell types (Figure 1D). This possibility seems to be rebutted by showing dissimilar distribution patterns of methylation changes among different blood cell types in response to an adverse environment.6,17 However, common methylation changes can occur when methylation-influencing genetic variations are dissimilarly distributed among comparing groups in EWAS. The genotype accounts for the majority (two-thirds) of inter-individual epigenetic variations.9,43,44 Fortunately, methylation analyses on different cell types have the advantage of distinguishing between genetically dependent or independent CpGs, provided that CpG methylation is commonly affected by genotype in each cell type. Finally, affected methylation changes are randomly distributed among different cell types (Figure 1E). It has been shown that inflammation-induced 5-halogenated or oxidation-induced cytosine damage products induce or prevent the methylation by DNA maintenance methyltransferase DNMT1 at these DNA sites, respectively.45 White blood cells, especially those, possessing peroxidases such as myeloperoxidase in activated neutrophils and monocytes or eosinophil peroxidase, are able to execute transhalogenation of deoxycytidine omitting DNA-binding enzymes.46 Although 99% fidelity of inheritance of methylation pattern has been postulated,47 persistent dysregulation of DNA methylation can be achieved by environmental pollutants affecting the fidelity of methyltransferases.48 Therefore, with both the non-site-specific chemical reaction on DNA and the impairment of fidelity of methyltransferases, a random pattern of altered DNA methylation is conceivable. Since alterations in cytosine methylation pattern are usually observed in human tumours, it can be proposed that rare changes become heritable and over-represented by progenitor cells. Thus, in dependence on the cellular stage by altered cytosine damage products, such methylation changes are undetectable or become detectable in EWAS. Consequences of cell-type-specific altered methylation for EWAS based on whole blood Irrespective of technical problems with design of oligonucleotides used to interrogate individual CpG sites within a given DNA sample, and analysis methods used by the Illumina Infinium HumanMethylation450K BeadChip49,50 as the most widely used platform in EWAS, the most important issue remains how to disentangle the comprehensive methylation datasets into valuable conclusions. Common methylation pattern of major blood cell types are frequently considered by adjustment of datasets for cell composition before analysis on methylation differences, but are often not considered when methylation differences are discussed. Adjustment for cell composition To exclude the possibility that altered methylation was evoked by individual differences in cell composition in blood, epigenetic datasets of EWAS are often adjusted for cell composition.3,51 To date, different reference-based52–57 or reference-free9,58–61 statistical methods for inferring main cell type proportion in whole blood samples has been released for application in EWAS. These methods, in good approximation, mirror the expected proportion of main cell types in blood, especially in great cohorts. However the use of adjustment methods can drastically change the conclusion made in EWAS.62 When comparing methylation profiles obtained from Illumina Infinium HumanMethylation450K BeadChip with different methods of adjustment for cell-type mixture confounding, no adjustment method was found whose performance was uniformly the best and, in some cases, the unadjusted results were quite comparable to the best adjusted results. For datasets with a large number of features, such as sequencing data, reference-free methods are even rather impractical owing to computational complexity.63,64 At subject level, estimated cell type proportion by DNA methylation may be less accurate and have a strong partial deviation from expected cell type proportion, especially for non-abundant cell types with a proportion of less than 10% in whole blood.57,65 Therefore, it can be suggested that the adjustment for cell composition per se must not be an absolute requirement before data analysis in EWAS. The scientific discussion about adjustment for cell composition becomes insignificant when investigation of methylation pattern in single cell types of blood would be performed. Data discussion Consistently, the majority of condition-triggered CpG methylation changes are not accompanied by CpG-annotated gene expression.66–68 Less than 1% of smoking-induced candidate CpG sites (5 out of 751) were associated with the transcript related to the same gene, analysing genome-wide DNA methylation by Illumina HumanMethylation450K and transcriptomic profiling focusing on about 9000 transcripts. Similar conclusions can be made comparing meta-analyses on the effect of smoking on gene expression69 and CpG methylation70 in blood. Only six of the top 25 ranked differentially expressed genes were found among the top 90 ranked CpG-annotated genes. However, top ranked altered CpG sites are frequently discussed in the functional context of their CpG-annotated genes or translated proteins with condition of interest in EWAS dealing with whole blood, omitting experimental confirmation of gene or protein expression. For instance, the smoking-induced methylation change at CpG site annotated to F2RL3 was set in the context of detrimental cardiovascular function of its protein,71 although a gene expression of F2RL3 in blood was not found.69 Thus, whether this methylation change is present at sites of smoking-induced cardiovascular pathology, such as platelets, vessel intima or macrophages, and associated with F2RL3 protein expression, cannot be postulated solely from methylation data on whole blood nor on single blood cell types. Therefore, cell-type-specific methylation pattern and weak correlation between epigenomic and transcriptomic changes all together weaken the importance of gene ontology pathway analysis based on methylation data, but should be considered to omit biased conclusions from datasets yielded from whole blood. Future study strategies Upon analysing DNA methylation, there are at least three detrimental conditions. The small effect size of altered methylation from studies on whole blood is a serious disadvantage in extracting different single CpG sites or regions. By increasing the number of analysed CpG sites, the more frequent implementation of whole-genome bisulphite sequencing (WGBS) approaches in near future is accompanied by an increase in expected false-positive results (increase in type I error), since WGBS analyse approximately 100-fold more CpGs in comparison with the yet more frequently used Illumina 450K arrays. Thereby the significance threshold after statistical multiple testing decreases from about 1.1 x 10−7 for 450K arrays to about 2.0 x 10−9 for WGBS. Grouping adjacent single CpGs into differentially methylated regions (DMR) becomes an alternative strategy reducing, to some extent, this type I error in WGBS. In contrast to the analysis of a fixed genotype (SNP) in GWAS, DNA methylation itself can be affected by multiple known and possibly yet unknown variable conditions. It has been shown that DNA methylation in whole blood is affected: constitutionally by age, chronological72–74 or gestational,75,76 circadian rhythm,77–80 genotype/ethnicity,44,81–83 gender,49,84,85 cell composition3,65,67,86–88 and physical health;89,90 by lifestyle factors such as active or passive smoking23,35,36,66,68,71,91–97 and nutritional behaviour determining body mass index;98–102 and additionally by diverse environmental stressors such as mother’s smoking during pregnancy,6,40,103–107 air pollution108,109 or metals110 as surrogates. This interference of DNA methylation requires a more stringent stratification of cohorts in EWAS. An enlargement of the number of subjects in cohorts for association studies would take the three detrimental conditions into account. However: (i) larger cohorts can indeed release more differential DNA methylation sites but more often with much smaller methylation differences, which does not improve the search for biomarkers; (ii) cohorts must dramatically increase in order to address the large number of human CpGs; and (iii) it remains impossible to consider all single DNA methylation-influencing conditions or their combinations as confounders. Thus the best alternative strategy would be the detection of altered DNA methylation in separated cell types (Figure 2). Due to the great number of different major cell types and their subtypes in blood, harbouring cell-type-specific or condition-triggered methylation patterns, the identification of a cell type’s origin of epigenetic changes in blood might be composed of consecutive cell separation rounds. In a hypothesis-free approach, a first cell separation round might be focused on granulocytes and PBMC, which may cost-effectively be separated by a simple density centrifugation step, followed by a second round with more specific separation techniques for cell subtypes of composed blood specimens, until the cell type’s origin of altered methylation has been found. For large cohorts with first datasets from whole blood, deeper methylation analysis in single cell types might be performed on a much smaller case-control subcohort. Figure 2 View largeDownload slide Proposed strategy in association studies to identify cell type origin of epigenetic changes in blood. Because of the great number of different major cell types and their minor subtypes in blood, harbouring cell-type-specific or condition-triggered methylation patterns, the identification process might be composed of consecutive cell separation rounds. GWAS, genome-wide association study; EWAS, epigenome-wide association study. Figure 2 View largeDownload slide Proposed strategy in association studies to identify cell type origin of epigenetic changes in blood. Because of the great number of different major cell types and their minor subtypes in blood, harbouring cell-type-specific or condition-triggered methylation patterns, the identification process might be composed of consecutive cell separation rounds. GWAS, genome-wide association study; EWAS, epigenome-wide association study. It should be emphasized that focusing primarily on specific cell types in blood for a given phenotype in a hypothesis-based approach, a general response of the tissue blood should not be neglected, since the discovery of novel cell types of the innate or adaptive immune system and their function is ongoing.111,112 For instance, in separated CD4+ and CD8+ T cells from blood of co-twins with/without psoriasis, differentially methylated genes were not found, although psoriasis plaques in skin are mediated by CD4+ and CD8+ T cells.22 Nevertheless, it can not be excluded that there are adverse constitutions in blood or other organs still not yet reflected by changed DNA methylation in whole blood. For heavy alcohol consumption, replicated altered CpG sites have not been currently constituted dealing with whole blood, and it remains speculative if such can be found extending investigation on different cell types. On the other side, the association of methylation profile in smoking-induced AHRR of blood monocytes, a cell type involved in atherogenesis, with carotid plaque scores113 must not be an exclusive biological link to atherosclerosis because there are other blood cell types of the myeloid and lymphoid lineage possessing altered methylation changes.17 In summary, it is strongly recommended to implement cell-type-composed blood specimens when specialized cell types are the primary scientific focus and vice versa, to extend investigations to single cell types when the focus is the whole blood. Summary Advantages of methylation analysis in blood cell types With strong reduction or elimination of inter-individual variance in cell composition, adjustment for cell composition becomes insignificant. By the methylation pattern in different cell types of blood, the following are enabled: identification of condition-provoked biological processes; identification of altered methylation in minor cell types; identification of cell type’s origin of altered methylation; assumption of genotype-influenced CpG site; and reduction in publication bias. By the expected higher effect size in single cell types compared with whole blood, the following are enabled: identification of biomarkers; reduction in number of participants in cohort studies; and identification of putative functional important CpG sites. Disadvantages of methylation analysis in blood cell types Disadvantages are negligible in comparison to the advantages: more comprehensive biosample handling (additional cell type separation) increase in analyses per subject investigations of immediate responses to changing conditions are restricted. Conclusion Although investigations on epigenetic DNA methylation are widespread among the research community, the success in disclosing causal relationship toward health-related outcomes to identify therapeutic targets is, in reality, still discouraging. There are currently more speculations than robust and replicated data in this field.114 It remains questionable whether more frequent implementation of whole-genome bisulphite sequencing will dramatically improve this situation. Thus, the present report intended to attract attention away from the mere statistical evaluation of epigenetic datasets and more towards the greater awareness of the metrics (effect size) in relation to biological plausibility.115,116 For this reason, the essential biological importance of cell type’s origin of altered DNA methylation in blood was illuminated and the detection of cell type’s origin is suggested to be the only meaningful strategy to uncover biomarkers and/or biological processes. Cell type separation has, amongs others things, the particular advantage of obtaining a greater effect size, and thereby fewer subjects must be enrolled in EWAS to discover meaningful findings. Conflict of interest: There is no conflict of interest. References 1 Naumova AK , Greenwood CMT (eds). Epigenetics and Complex Traits . New York, NY : Springer Science+Business Media , 2013 . Google Scholar CrossRef Search ADS 2 Novak P , Stampfer MR , Munoz-Rodriguez JL et al. Cell-type specific DNA methylation pattern define human breast cellular identity . PLoS One 2012 ; 7: e52299 . Google Scholar CrossRef Search ADS PubMed 3 Reinius LE , Acevedo N , Joerink M et al. . Differential DNA methylation in purified human blood cells: implications for cell lineage and studies on disease susceptibility . PLoS One 2012 ; 7: e41361 . Google Scholar CrossRef Search ADS PubMed 4 Bird A. DNA methylation pattern and epigenetic memory . Genes Dev 2002 ; 16: 6 – 21 . Google Scholar CrossRef Search ADS PubMed 5 Vizoso M , Esteller M. DNA methylation plasticity contributes to the natural history of metastasis . Cell Cycle 2015 ; 14: 2863 – 64 . Google Scholar CrossRef Search ADS PubMed 6 Bauer M , Fink B , Thurmann L , Eszlinger M , Herberth G , Lehmann I. Tobacco smoking differently influences cell types of the innate and adaptive immune system - indications from CpG site methylation . Clin Epigenetics 2015 ; 7: 83 . Google Scholar CrossRef Search ADS PubMed 7 Bauer M , Linsel G , Fink B et al. . A varying T cell subtype explains apparent tobacco smoking induced single CpG hypomethylation in whole blood . Clin Epigenetics 2015 ; 7: 81 . Google Scholar CrossRef Search ADS PubMed 8 Liang L , Willis-Owen SA , Laprise C et al. An epigenome-wide association study of total serum immunoglobulin E concentration . Nature 2015 ; 520: 670 – 74 . Google Scholar CrossRef Search ADS PubMed 9 Bauer T , Trump S , Ishaque N et al. Environment-induced epigenetic reprogramming in genomic regulatory elements in smoking mothers and their children . Mol Syst Biol 2016 ; 12: 861 . Google Scholar CrossRef Search ADS PubMed 10 Hassan HM, , Kolendowski B, , Isovic M et al. Regulation of Active DNA Demethylation through RAR-Mediated Recruitment of a TET/TDG Complex . Cell Rep 2017 ; 19: 1685 – 97 . Google Scholar CrossRef Search ADS PubMed 11 Andersen AM , Dogan MV , Beach SR , Philibert RA. Current and future prospects for epigenetic biomarkers of substance use disorders . Genes (Basel) 2015 ; 6: 991 – 1022 . Google Scholar CrossRef Search ADS PubMed 12 Han X , Wang J , Sun Y. Circulating Tumor DNA as Biomarkers for Cancer Detection . Genom Proteom Bioinformatics 2017 ; 15: 59 – 72 . Google Scholar CrossRef Search ADS 13 Wu T , Cheng B , Fu L. Clinical applications of circulating tumor cells in pharmacotherapy: challenges and perspectives . Mol Pharmacol 2017 ; 92: 232 – 39 . Google Scholar CrossRef Search ADS PubMed 14 Daniels VG , Wheater PR , Burkit HG. Functional Histology: A Text and Colour Atlas . Edinburgh, UK : Churchill Livingstone , 1979 . 15 Bekkering S , Joosten LA , van der Meer JW , Netea MG , Riksen NP. Trained innate immunity and atherosclerosis . Curr Opin Lipidol 2013 ; 24: 917 927 . Google Scholar CrossRef Search ADS 16 Narni-Mancinelli E , Soudja SM , Crozat K et al. . Inflammatory monocytes and neutrophils are licensed to kill during memory responses in vivo . PLoS Pathog 2011 ; 7: e1002457 . Google Scholar CrossRef Search ADS PubMed 17 Su D , Wang X , Campbell MR et al. Distinct epigenetic effects of tobacco smoking in whole blood and among leukocyte subtypes . PLoS One 2016 ; 11: e0166486 . Google Scholar CrossRef Search ADS PubMed 18 Lowe R , Gemma C , Beyan H et al. Buccals are likely to be a more informative surrogate tissue than blood for epigenome-wide association studies . Epigenetics 2013 ; 8: 445 – 54 . Google Scholar CrossRef Search ADS PubMed 19 Teschendorff AE , Yang Z , Wong A et al. Correlation of smoking-associated DNA methylation changes in buccal cells with DNA methylation changes in epithelial cancer . JAMA Oncol 2015 ; 1: 476 – 85 . Google Scholar CrossRef Search ADS PubMed 20 Tomlinson MJ , Tomlinson S , Yang XB , Kirkham J. Cell separation: Terminology and practical considerations . J Tissue Eng 2013 ; 4: 2041731412472690 . Google Scholar CrossRef Search ADS PubMed 21 Liang L , Cookson WO. Grasping nettles: cellular heterogeneity and other confounders in epigenome-wide association studies . Hum Mol Genet 2014 ; 23: R83 – 8 . Google Scholar CrossRef Search ADS PubMed 22 Gervin K , Vigeland MD , Mattingsdal M et al. DNA methylation and gene expression changes in monozygotic twins discordant for psoriasis: identification of epigenetically dysregulated genes . PLoS Genet 2012 ; 8: e1002454 . Google Scholar CrossRef Search ADS PubMed 23 Dogan MV , Shields B , Cutrona C et al. . The effect of smoking on DNA methylation of peripheral blood mononuclear cells from African American women . BMC Genomics 2014 ; 15: 151 . Google Scholar CrossRef Search ADS PubMed 24 Philibert RA , Beach SR , Brody GH. Demethylation of the aryl hydrocarbon receptor repressor as a biomarker for nascent smokers . Epigenetics 2012 ; 7: 1331 – 38 . Google Scholar CrossRef Search ADS PubMed 25 Zhang Y , Schottker B , Florath I et al. Smoking-associated DNA methylation biomarkers and their predictive value for all-cause and cardiovascular mortality . Environ Health Perspect 2016 ; 124: 67 – 74 . Google Scholar CrossRef Search ADS PubMed 26 Zhang Y , Yang R , Burwinkel B et al. F2RL3 methylation in blood DNA is a strong predictor of mortality . Int J Epidemiol 2014 ; 43: 1215 – 25 . Google Scholar CrossRef Search ADS PubMed 27 Reese SE , Zhao S , Wu MC et al. DNA methylation score as a biomarker in newborns for sustained maternal smoking during pregnancy . Environ Health Perspect 2016 ; 125: 760 – 66. Google Scholar CrossRef Search ADS PubMed 28 Ladd-Acosta C , Shu C , Lee BK et al. Presence of an epigenetic signature of prenatal cigarette smoke exposure in childhood . Environ Res 2016 ; 144(Pt A): 139 – 48 . Google Scholar CrossRef Search ADS 29 Brower V. Biomarkers: Portents of malignancy . Nature 2011 ; 471: S19 – 21 . Google Scholar CrossRef Search ADS PubMed 30 Strimbu K , Tavel JA. What are biomarkers? Curr Opin HIV AIDS 2010 ; 5: 463 – 66 . Google Scholar CrossRef Search ADS PubMed 31 Wians FH. Clinical laboratory tests: which, why and what do the results mean? Lab Med 2009 ; 40 : 105 – 13 . Google Scholar CrossRef Search ADS 32 Yan S , Liu Z , Yu S , Bao Y. Diagnostic value of methylated septin9 for colorectal cancer screening: a meta-analysis . Med Sci Monit 2016 ; 22: 3409 – 18 . Google Scholar CrossRef Search ADS PubMed 33 Robert Koch Institut . Cancer in Germany, Colon and Rectum. 2010 . http://www.gbe-bund.de/pdf/Darm_C18_21.pdf (February 2018, date last accessed). 34 Statistische Bundesamt (Federal Statistical Office). Periodic update of German population. https://www-genesis.destatis.de/genesis/online/data;jsessionid = DD6DC999FF30B4952CCB3D1A44AA0139.tomcat_GO_1_1?operation = abruftabelleBearbeiten&levelindex = 2&levelid = 1476714342359&auswahloperation = abruftabelleAuspraegungAuswaehlen&auswahlverzeichnis = ordnungsstruktur&auswahlziel = werteabruf&selectionname = 12411-0005&auswahltext = &werteabruf = starten (February 2018, date last accessed). 35 Shenker NS , Ueland PM , Polidoro S et al. DNA methylation as a long-term biomarker of exposure to tobacco smoke . Epidemiology . 2013 ; 24: 712 – 16 . Google Scholar CrossRef Search ADS PubMed 36 Elliott HR , Tillin T , McArdle WL et al. . Differences in smoking associated DNA methylation pattern in South Asians and Europeans . Clin Epigenetics 2014 ; 6: 4 . Google Scholar CrossRef Search ADS PubMed 37 Oh T , Kim N , Moon Y et al. Genome-wide identification and validation of a novel methylation biomarker, SDC2, for blood-based detection of colorectal cancer . J Mol Diagn 2013 ; 15: 498 – 507 . Google Scholar CrossRef Search ADS PubMed 38 Kneip C , Schmidt B , Seegebarth A et al. SHOX2 DNA methylation is a biomarker for the diagnosis of lung cancer in plasma . J Thorac Oncol 2011 ; 6: 1632 – 38 . Google Scholar CrossRef Search ADS PubMed 39 Roupret M , Hupertan V , Catto JW et al. Promoter hypermethylation in circulating blood cells identifies prostate cancer progression . Int J Cancer 2008 ; 122: 952 – 56 . Google Scholar CrossRef Search ADS PubMed 40 Novakovic B , Ryan J , Pereira N , Boughton B , Craig JM , Saffery R. Postnatal stability, tissue, and time specific effects of AHRR methylation change in response to maternal smoking in pregnancy . Epigenetics 2014 ; 9: 377 – 86 . Google Scholar CrossRef Search ADS PubMed 41 Simar D , Versteyhe S , Donkin I et al. DNA methylation is altered in B and NK lymphocytes in obese and type 2 diabetic human . Metabolism 2014 ; 63: 1188 – 97 . Google Scholar CrossRef Search ADS PubMed 42 Absher DM , Li X , Waite LL et al. Genome-wide DNA methylation analysis of systemic lupus erythematosus reveals persistent hypomethylation of interferon genes and compositional changes to CD4+ T-cell populations . PLoS Genet 2013 ; 9: e1003678 . Google Scholar CrossRef Search ADS PubMed 43 Chen L , Ge B , Casale FP et al. Genetic drivers of epigenetic and transcriptional variation in human immune cells . Cell 2016 ; 167: 1398 – 414 e24 . Google Scholar CrossRef Search ADS PubMed 44 Heyn H , Moran S , Hernando-Herraez I et al. DNA methylation contributes to natural human variation . Genome Res 2013 ; 23 : 1363 – 72 . Google Scholar CrossRef Search ADS PubMed 45 Valinluck V , Sowers LC. Endogenous cytosine damage products alter the site selectivity of human DNA maintenance methyltransferase DNMT1 . Cancer Res 2007 ; 67 : 946 – 50 . Google Scholar CrossRef Search ADS PubMed 46 Henderson JP , Byun J , Williams MV , Mueller DM , McCormick ML , Heinecke JW. Production of brominating intermediates by myeloperoxidase. A transhalogenation pathway for generating mutagenic nucleobases during inflammation . J Biol Chem 2001 ; 276 : 7867 – 75 . Google Scholar CrossRef Search ADS PubMed 47 Damelin M , Bestor TH. Biological functions of DNA methyltransferase 1 require its methyltransferase activity . Mol Cell Biol 2007 ; 27: 3891 – 99 . Google Scholar CrossRef Search ADS PubMed 48 Mauro M , Caradonna F , Klein CB. Dysregulation of DNA methylation induced by past arsenic treatment causes persistent genomic instability in mammalian cells . Environ Mol Mutagen 2016 ; 57: 137 – 50 . Google Scholar CrossRef Search ADS PubMed 49 Chen YA , Lemire M , Choufani S et al. Discovery of cross-reactive probes and polymorphic CpGs in the Illumina Infinium HumanMethylation450 microarray . Epigenetics 2013 ; 8: 203 – 09 . Google Scholar CrossRef Search ADS PubMed 50 Harper KN , Peters BA , Gamble MV. Batch effects and pathway analysis: two potential perils in cancer studies involving DNA methylation array analysis . Cancer Epidemiol Biomarkers Prev 2013 ; 22: 1052 – 60 . Google Scholar CrossRef Search ADS PubMed 51 Adalsteinsson BT , Gudnason H , Aspelund T et al. Heterogeneity in white blood cells has potential to confound DNA methylation measurements . PLoS One 2012 ; 7: e46705 . Google Scholar CrossRef Search ADS PubMed 52 Houseman EA , Accomando WP , Koestler DC et al. DNA methylation arrays as surrogate measures of cell mixture distribution . BMC Bioinformatics 2012 ; 13: 86 . Google Scholar CrossRef Search ADS PubMed 53 Jaffe AE , Irizarry RA. Accounting for cellular heterogeneity is critical in epigenome-wide association studies . Genome Biol 2014 ; 15: R31 . Google Scholar CrossRef Search ADS PubMed 54 Cardenas A , Allard C , Doyon M et al. Validation of a DNA methylation reference panel for the estimation of nucleated cells types in cord blood . Epigenetics 2016 ; 11: 773 – 79 . Google Scholar CrossRef Search ADS PubMed 55 Bakulski KM , Feinberg JI , Andrews SV et al. DNA methylation of cord blood cell types: Applications for mixed cell birth studies . Epigenetics 2016 ; 11: 354 – 62 . Google Scholar CrossRef Search ADS PubMed 56 Waite LL , Weaver B , Day K et al. Estimation of cell-type composition including T and B cell subtypes for whole blood methylation microarray data . Front Genet 2016 ; 7: 23 . Google Scholar CrossRef Search ADS PubMed 57 Accomando WP , Wiencke JK , Houseman EA , Nelson HH , Kelsey KT. Quantitative reconstruction of leukocyte subsets using DNA methylation . Genome Biol 2014 ; 15: R50 . Google Scholar CrossRef Search ADS PubMed 58 Houseman EA , Molitor J , Marsit CJ. Reference-free cell mixture adjustments in analysis of DNA methylation data . Bioinformatics 2014 ; 30: 1431 – 39 . Google Scholar CrossRef Search ADS PubMed 59 Zou J , Lippert C , Heckerman D , Aryee M , Listgarten J. Epigenome-wide association studies without the need for cell-type composition . Nat Methods 2014 ; 11 : 309 – 11 . Google Scholar CrossRef Search ADS PubMed 60 Houseman EA , Kile ML , Christiani DC , Ince TA , Kelsey KT , Marsit CJ. Reference-free deconvolution of DNA methylation data and mediation by cell composition effects . BMC Bioinformatics 2016 ; 17: 259 . Google Scholar CrossRef Search ADS PubMed 61 Heiss JA , Breitling LP , Lehne B , Kooner JS , Chambers JC , Brenner H. Training a model for estimating leukocyte composition using whole-blood DNA methylation and cell counts as reference . Epigenomics 2017 ; 9: 13 – 20 . Google Scholar CrossRef Search ADS PubMed 62 McGregor K , Bernatsky S , Colmegna I et al. . An evaluation of methods correcting for cell-type heterogeneity in DNA methylation studies . Genome Biol 2016 ; 17: 84 . Google Scholar CrossRef Search ADS PubMed 63 McGregor K , Labbe A , Greenwood CM. Response to: Correcting for cell-type effects in DNA methylation studies: reference-based method outperforms latent variable approaches in empirical studies . Genome Biol 2017 ; 18: 25 . Google Scholar CrossRef Search ADS PubMed 64 Hattab MW , Shabalin AA , Clark SL et al. Correcting for cell-type effects in DNA methylation studies: reference-based method outperforms latent variable approaches in empirical studies . Genome Biol 2017 ; 18: 24 . Google Scholar CrossRef Search ADS PubMed 65 Koestler DC , Christensen B , Karagas MR et al. Blood-based profiles of DNA methylation predict the underlying distribution of cell types: a validation analysis . Epigenetics 2013 ; 8: 816 – 26 . Google Scholar CrossRef Search ADS PubMed 66 Tsaprouni LG , Yang TP , Bell J et al. Cigarette smoking reduces DNA methylation levels at multiple genomic loci but the effect is partially reversible upon cessation . Epigenetics 2014 ; 9: 1382 – 96 . Google Scholar CrossRef Search ADS PubMed 67 Rask-Andersen M , Bringeland N , Nilsson EK et al. Postprandial alterations in whole-blood DNA methylation are mediated by changes in white blood cell composition . Am J Clin Nutr 2016 ; 104: 518 – 25 . Google Scholar CrossRef Search ADS PubMed 68 Guida F , Sandanger TM , Castagne R et al. Dynamics of smoking-induced genome-wide methylation changes with time since smoking cessation . Hum Mol Genet 2015 ; 24 : 2349 – 59 . Google Scholar CrossRef Search ADS PubMed 69 Huan T , Joehanes R , Schurmann C et al. A whole-blood transcriptome meta-analysis identifies gene expression signatures of cigarette smoking . Hum Mol Genet 2016 ; 25: 4611 – 23 . Google Scholar PubMed 70 Joehanes R , Just AC , Marioni RE et al. Epigenetic signatures of cigarette smoking . Circ Cardiovasc Genet 2016 ; 9: 436 – 47 . Google Scholar CrossRef Search ADS PubMed 71 Breitling LP , Yang R , Korn B , Burwinkel B , Brenner H. Tobacco-smoking-related differential DNA methylation: 27K discovery and replication . Am J Hum Genet 2011 ; 88: 450 – 57 . Google Scholar CrossRef Search ADS PubMed 72 Hannum G , Guinney J , Zhao L et al. Genome-wide methylation profiles reveal quantitative views of human aging rates . Mol Cell 2013 ; 49: 359 – 67 . Google Scholar CrossRef Search ADS PubMed 73 Horvath S. DNA methylation age of human tissues and cell types . Genome Biol 2013 ; 14: R115 . Google Scholar CrossRef Search ADS PubMed 74 Weidner CI , Lin Q , Koch CM et al. Aging of blood can be tracked by DNA methylation changes at just three CpG sites . Genome Biol 2014 ; 15: R24 . Google Scholar CrossRef Search ADS PubMed 75 Knight AK , Craig JM , Theda C et al. . An epigenetic clock for gestational age at birth based on blood methylation data . Genome Biol 2016 ; 17: 206 . Google Scholar CrossRef Search ADS PubMed 76 Simpkin AJ , Suderman M , Gaunt TR et al. Longitudinal analysis of DNA methylation associated with birth weight and gestational age . Hum Mol Genet 2015 ; 24: 3752 – 63 . Google Scholar CrossRef Search ADS PubMed 77 Wong CC , Parsons MJ , Lester KJ et al. Epigenome-wide DNA methylation analysis of monozygotic twins discordant for diurnal preference . Twin Res Hum Genet 2015 ; 18: 662 – 69 . Google Scholar CrossRef Search ADS PubMed 78 Zhu Y , Stevens RG , Hoffman AE et al. Epigenetic impact of long-term shiftwork: pilot evidence from circadian genes and whole-genome methylation analysis . Chronobiol Int 2011 ; 28: 852 – 61 . Google Scholar CrossRef Search ADS PubMed 79 Bonsch D , Hothorn T , Krieglstein C et al. Daily variations of homocysteine concentration may influence methylation of DNA in normal healthy individuals . Chronobiol Int 2007 ; 24: 315 – 26 . Google Scholar CrossRef Search ADS PubMed 80 Skuladottir GV , Nilsson EK , Mwinyi J , Schioth HB. One-night sleep deprivation induces changes in the DNA methylation and serum activity indices of stearoyl-CoA desaturase in young healthy men . Lipids Health Dis 2016 ; 15: 137 . Google Scholar CrossRef Search ADS PubMed 81 Bell JT , Pai AA , Pickrell JK et al. DNA methylation pattern associate with genetic and gene expression variation in HapMap cell lines . Genome Biol 2011 ; 12: R10 . Google Scholar CrossRef Search ADS PubMed 82 Gibbs JR , van der Brug MP , Hernandez DG et al. Abundant quantitative trait loci exist for DNA methylation and gene expression in human brain . PLoS Genet 2010 ; 6: e1000952 . Google Scholar CrossRef Search ADS PubMed 83 Zhang D , Cheng L , Badner JA et al. Genetic control of individual differences in gene-specific methylation in human brain . Am J Hum Genet 2010 ; 8: 411 – 19 . Google Scholar CrossRef Search ADS 84 van Dongen J , Nivard MG , Willemsen G et al. Genetic and environmental influences interact with age and sex in shaping the human methylome . Nat Commun 2016 ; 7: 11115 . Google Scholar CrossRef Search ADS PubMed 85 Mamrut S , Avidan N , Staun-Ram E et al. Integrative analysis of methylome and transcriptome in human blood identifies extensive sex- and immune cell-specific differentially methylated regions . Epigenetics 2015 ; 10: 943 – 57 . Google Scholar CrossRef Search ADS PubMed 86 Lam LL , Emberly E , Fraser HB et al. Factors underlying variable DNA methylation in a human community cohort . Proc Natl Acad Sci U S A 2012 ; 109(Suppl 2): 17253 – 60 . Google Scholar CrossRef Search ADS PubMed 87 Lowe R , Rakyan VK. Correcting for cell-type composition bias in epigenome-wide association studies . Genome Med 2014 ; 6: 23 . Google Scholar CrossRef Search ADS PubMed 88 Sehouli J , Loddenkemper C , Cornu T et al. Epigenetic quantification of tumor-infiltrating T-lymphocytes . Epigenetics 2011 ; 6: 236 – 46 . Google Scholar CrossRef Search ADS PubMed 89 Chambers JC , Loh M , Lehne B et al. Epigenome-wide association of DNA methylation markers in peripheral blood from Indian Asians and Europeans with incident type 2 diabetes: a nested case-control study . Lancet Diabetes Endocrinol 2015 ; 3: 526 – 34 . Google Scholar CrossRef Search ADS PubMed 90 Gomez-Uriz AM , Milagro FI , Mansego ML et al. Obesity and ischemic stroke modulate the methylation levels of KCNQ1 in white blood cells . Hum Mol Genet 2015 ; 24: 1432 – 40 . Google Scholar CrossRef Search ADS PubMed 91 Ambatipudi S , Cuenin C , Hernandez-Vargas H et al. Tobacco smoking-associated genome-wide DNA methylation changes in the EPIC study . Epigenomics 2016 ; 8: 599 – 618 . Google Scholar CrossRef Search ADS PubMed 92 Harlid S , Xu Z , Panduri V , Sandler DP , Taylor JA. CpG sites associated with cigarette smoking: analysis of epigenome-wide data from the Sister Study . Environ Health Perspect 2014 ; 122 : 673 – 78 . Google Scholar PubMed 93 Lee MK , Hong Y , Kim SY , London SJ , Kim WJ. DNA methylation and smoking in Korean adults: epigenome-wide association study . Clin Epigenetics 2016 ; 8: 103 . Google Scholar CrossRef Search ADS PubMed 94 Sun YV , Smith AK , Conneely KN et al. Epigenomic association analysis identifies smoking-related DNA methylation sites in African Americans . Hum Genet 2013 ; 132: 1027 – 37 . Google Scholar CrossRef Search ADS PubMed 95 Wan ES , Qiu W , Baccarelli A et al. Cigarette smoking behaviors and time since quitting are associated with differential DNA methylation across the human genome . Hum Mol Genet 2012 ; 21: 3073 – 82 . Google Scholar CrossRef Search ADS PubMed 96 Zeilinger S , Kuhnel B , Klopp N et al. . Tobacco smoking leads to extensive genome-wide changes in DNA methylation . PLoS One 2013 ; 8: e63812 . Google Scholar CrossRef Search ADS PubMed 97 Reynolds LM , Magid HS , Chi GC et al. Secondhand tobacco smoke exposure associations with DNA methylation of the aryl hydrocarbon receptor repressor . Nicotine Tob Res 2017 ; 19: 442 – 51 . Google Scholar PubMed 98 Wilson LE , Harlid S , Xu Z , Sandler DP , Taylor JA. An epigenome-wide study of body mass index and DNA methylation in blood using participants from the Sister Study cohort . Int J Obes (Lond) 2017 ; 41: 194 – 99 . Google Scholar CrossRef Search ADS PubMed 99 Burris HH , Baccarelli AA , Byun HM et al. Offspring DNA methylation of the aryl-hydrocarbon receptor repressor gene is associated with maternal BMI, gestational age, and birth weight . Epigenetics 2015 ; 10 : 913 – 21 . Google Scholar CrossRef Search ADS PubMed 100 Aslibekyan S , Demerath EW , Mendelson M et al. Epigenome-wide study identifies novel methylation loci associated with body mass index and waist circumference . Obesity (Silver Spring) 2015 ; 23: 1493 – 501 . Google Scholar CrossRef Search ADS PubMed 101 Demerath EW , Guan W , Grove ML et al. Epigenome-wide association study (EWAS) of BMI, BMI change and waist circumference in African American adults identifies multiple replicated loci . Hum Mol Genet 2015 ; 2: 4464 – 79 . Google Scholar CrossRef Search ADS 102 Dick KJ , Nelson CP , Tsaprouni L et al. DNA methylation and body-mass index: a genome-wide analysis . Lancet 2014 ; 383: 1990 – 98 . Google Scholar CrossRef Search ADS PubMed 103 Wang IJ , Chen SL , Lu TP , Chuang EY , Chen PC. Prenatal smoke exposure, DNA methylation, and childhood atopic dermatitis . Clin Exp Allergy 2013 ; 43: 535 – 43 . Google Scholar CrossRef Search ADS PubMed 104 Joubert BR , Haberg SE , Bell DA et al. Maternal smoking and DNA methylation in newborns: in utero effect or epigenetic inheritance? Cancer Epidemiol Biomarkers Prev 2014 ; 23: 1007 – 17 . Google Scholar CrossRef Search ADS PubMed 105 Markunas CA , Xu Z , Harlid S et al. Identification of DNA methylation changes in newborns related to maternal smoking during pregnancy . Environ Health Perspect 2014 ; 122 : 1147 – 53 . Google Scholar PubMed 106 Ivorra C , Fraga MF , Bayon GF et al. DNA methylation pattern in newborns exposed to tobacco in utero . J Transl Med 2015 ; 13: 25 . Google Scholar CrossRef Search ADS PubMed 107 Kupers LK , Xu X , Jankipersadsing SA et al. DNA methylation mediates the effect of maternal smoking during pregnancy on birthweight of the offspring . Int J Epidemiol 2015 ; 44: 1224 – 37 . Google Scholar CrossRef Search ADS PubMed 108 Gruzieva O , Xu CJ , Breton CV et al. Epigenome-wide meta-analysis of methylation in children related to prenatal NO2 air pollution exposure . Environ Health Perspect 2017 ; 125: 104 – 10 . Google Scholar CrossRef Search ADS PubMed 109 Panni T , Mehta AJ , Schwartz JD et al. Genome-wide analysis of DNA methylation and fine particulate matter air pollution in three study populations: KORA F3, KORA F4, and the Normative Aging Study . Environ Health Perspect 2016 ; 124: 983 – 90 . Google Scholar CrossRef Search ADS PubMed 110 Martin EM , Fry RC. A cross-study analysis of prenatal exposures to environmental contaminants and the epigenome: support for stress-responsive transcription factor occupancy as a mediator of gene-specific CpG methylation patterning . Environ Epigenet 2016 , Jan. pii: dvv011. 111 Lepore M , Kalinichenko A , Calogero S et al. Functionally diverse human T cells recognize non-microbial antigens presented by MR1 . Elife 2017 , Jun 20. doi: 10.7554/eLife.29743. 112 Petersen BC , Budelsky AL , Baptist AP , Schaller MA , Lukacs NW. Interleukin-25 induces type 2 cytokine production in a steroid-resistant interleukin-17RB+ myeloid population that exacerbates asthmatic pathology . Nat Med 2012 ; 18 : 751 – 58 . Google Scholar CrossRef Search ADS PubMed 113 Reynolds LM , Wan M , Ding J et al. DNA methylation of the aryl hydrocarbon receptor repressor associations with cigarette smoking and subclinical atherosclerosis . Circ Cardiovasc Genet 2015 ; 8 : 707 – 16 . Google Scholar CrossRef Search ADS PubMed 114 Relton CL , Davey Smith G. Two-step epigenetic Mendelian randomization: a strategy for establishing the causal role of epigenetic processes in pathways to disease . Int J Epidemiol 2012 ; 41 : 161 – 76 . Google Scholar CrossRef Search ADS PubMed 115 Fedak KM , Bernal A , Capshaw ZA , Gross S. Applying the Bradford Hill criteria in the 21st century: how data integration has changed causal inference in molecular epidemiology . Emerg Themes Epidemiol 2015 ; 12: 14 . Google Scholar CrossRef Search ADS PubMed 116 Claridge-Chang A , Assam PN. Estimation statistics should replace significance testing . Nat Methods 2016 ; 13 : 108 – 09 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association 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)
International Journal of Epidemiology – Oxford University Press
Published: Mar 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