Skip to main content
NCBI home page
UKPMC Funders Author Manuscripts logoLink to UKPMC Funders Author Manuscripts
. Author manuscript; available in PMC: 2019 May 8.
Published in final edited form as: New Phytol. 2018 Aug 19;221(2):738–742. doi: 10.1111/nph.15388

Opportunities and limitations of reduced representation bisulfite sequencing in plant ecological epigenomics

Ovidiu Paun 1, Koen JF Verhoeven 2, Christina L Richards 3
PMCID: PMC6504643  EMSID: EMS82368  PMID: 30121954

Summary

Investigating the features and implications of epigenetic mechanisms across the breadth of organisms and ecosystems is important for understanding the ecological relevance of epigenetics. Several cost-effective reduced representation bisulfite sequencing approaches (RRBS) have been recently developed and applied to different organisms that lack a well-annotated reference genome. These new approaches improve the assessment of epigenetic diversity in ecological settings and may provide functional insights. We asses here the opportunities and limitations of RRBS in non-model plant species. Well-thought out experimental designs that include complementary gene expression studies, and the improvement of genomics resources for the target group promise to maximize impact of future RRBS studies.

Keywords: adaptation, bisulfite sequencing, bsRADseq, DNA methylation, ecological genomics, epiGBS, response to environment, RRBS

I. Introduction

Ecological epigenetics aims to understand the unique contributions of epigenetic mechanisms to ecological and evolutionary processes (Bossdorf et al., 2008; Richards et al., 2017). Several studies have shown that DNA methylation variation correlates with ecological factors, suggesting a potential role for epigenetics in adaptation (Verhoeven et al., 2016; Richards et al., 2017). However, the low genomic resolution of many studies on non-model organisms precludes pinpointing causality of epigenetic effects, and the peculiarities of the few plant and animal model organisms may impede generalizations (e.g., Lyko, 2001; Alonso et al., 2015). Recently, we discussed how high resolution epigenomics tools must be combined with ecologically-relevant experimental settings to explore the importance of epigenetic mechanisms (Richards et al., 2017). We also argued that in order to evaluate the relevance of epigenetic mechanisms to ecology and evolution, these approaches need to be applied across a diversity of organisms and conditions. However, powerful epigenomic methodologies have largely been accessible for only a few model species. The ‘gold standard’ of bisulfite sequencing applied to whole genomes (WGBS) evaluates methylation status for essentially every cytosine in a genome (Frommer et al., 1992; Harris et al., 2010; Li & Tollefsbol, 2011; Olova et al., 2018), but this approach is still limited in application to species with a high-quality reference genome (e.g. Niederhuth et al., 2016). Furthermore, this approach is prohibitively expensive for large experimental designs, particularly for average to large plant genomes.

Several recently developed approaches based on reduced-representation bisulfite sequencing methods (RRBS) provide cost-effective alternatives to WGBS, by interrogating only a representative fraction of the genome (Gu et al., 2011; Wang et al., 2015; Trucchi et al., 2016; van Gurp et al., 2016). RRBS approaches are scalable for ecological experimental designs, can be used on organisms without reference genomes, and allow for greater resolution than previous marker-based approaches. RRBS reads can also be mapped to a reference or draft reference genome when available (Gugger et al., 2016; Lea et al., 2016; Trucchi et al., 2016; Weyrich et al., 2016), and may provide functional insights. For the species that lack a reference genome, RRBS studies can generate accession-specific references for only the loci that are interrogated, either by including non-bisulfite-converted samples in the sequencing (e.g. for bsRADseq, Trucchi et al., 2016) or by inferring the un-converted reference from the bisulfite treated reads themselves (e.g. epiGBS, van Gurp et al., 2016). These attributes combined with cost-efficiency make RRBS methods attractive options for studies that address ecological and evolutionary epigenetics in essentially any organism. However, it is important to evaluate their strengths along with any limitations to identify which research questions can be fruitfully addressed.

II. RRBS loci as genome-wide epigenetic markers

RRBS methods are more powerful than previous methods (such as methylation-sensitive amplified fragment length polymorphism or MS-AFLP/MSAP; see e.g., Schrey et al., 2013) for exploring patterns of DNA methylation variation at a genome-wide scale. The increase in power is due to the fact that RRBS typically reviews multiple cytosine positions within each of tens of thousands of fragments. Unlike earlier methods, RRBS data is quantitative at nucleotide resolution, which reflects the biology of DNA methylation across cells in a tissue sample. RRBS methods provide methylation information at different sequence contexts (CG, CHG, CHH). Methylation in these different DNA contexts is correlated to distinct functions, regulated by different molecular pathways, and known to have different levels of heritability and environmental-sensitivity, thus providing further tools for dissecting natural patterns of DNA methylation (Dubin et al., 2015; Niederhuth et al., 2016; Trucchi et al., 2016; Niederhuth & Schmitz, 2017). RRBS data can be integrated to quantify overall similarity among individuals in DNA methylation, which supplies a basis for describing patterns of epigenetic variation within and among natural populations.

Such descriptive population epigenetic measures provide useful insights about the distribution of epigenetic variation. For instance, to what extent is natural epigenetic population structure correlated to underlying genetic population structure, and to what extent is it correlated with environmental variation? High correlation between epigenetic and genetic structure has been observed in several studies of natural populations of Arabidopsis thaliana (Dubin et al., 2015; Kawakatsu et al., 2016), experimental populations of maize (Eichten et al., 2013), and recombinant inbred lines (RILs) of soybean (Schmitz et al., 2013), suggesting that a significant portion of epigenetic variation is under genetic control. These studies rely on detecting associations between differentially methylated regions (DMR) and neighboring genetic polymorphisms to indicate genetic control in cis over DNA methylation. Associations in trans have also been detected, for example between CHH methylation and genetic variation in the chromomethyltransferase CMT2 (Dubin et al., 2015). However, cis and trans correlations can also arise in the absence of genetic control, when a spontaneous epimutation becomes stably inherited (Taudt et al., 2016). In addition, some studies have reported that a component of epigenetic population structure was not explained by genetic structure, and was instead associated with ecological conditions (e.g. Richards et al., 2012; Dubin et al., 2015; Foust et al., 2016; Kawakatsu et al., 2016). This suggests that epigenetics can contribute directly or indirectly to adaptation: either i) because epigenetic changes are induced by environmental conditions, and provide a capacity for phenotypic plasticity or longer-term heritable responses, and/or ii) because the epigenetic variation that does not correlate to overall (neutral) genetic population structure can be under control by individual genetic loci.

In addition to characterizing natural DNA methylation variation, the high heritability of cytosine methylation in plants suggests that RRBS-loci could be used as markers for linkage mapping purposes in crosses where genetic variation is limited or absent (Johannes et al., 2009; Cortijo et al., 2014; Dapp et al., 2015; Hofmeister et al., 2017). Also, general characteristics of DNA methylation dynamics, such as transgenerational stability, epimutation and reversal rates, and response to experimental or field conditions of interest, can be investigated by looking at many, but not necessarily all, loci in the genome, and may thus be addressed with RRBS approaches.

III. Exploiting functional annotation of RRBS loci

The DNA sequence contained in RRBS fragments holds promise for providing information about the function of the genomic regions that show DNA methylation differences (Fig. 1). In particular, epigenetic changes at enhancer elements can drive gene expression alterations and have functional consequences (Taudt et al., 2016). RRBS methods were initially developed for studies in mammals where a large fraction of gene promoters are associated with GC-rich areas (CpG islands) and methylation status of these areas has functional consequences for gene regulation (Meissner et al., 2005). Using restriction enzymes that bias toward GC-rich sites (such as MspI), RRBS in mammals evaluates methylation in 80-90% of CpG islands by targeting only a small portion of the total genome (Smith et al., 2009). Typical plant genomes lack CpG islands and MspI-based reduced representation of genomes does not result in a bias towards gene promoters (Hsu et al., 2017). However, bias to either gene bodies or gene promoters can be achieved in plants by selecting different restriction enzymes, and increasing overall genome representation. For instance, RRBS in maize using the frequent cutter MseI provides a representation of >80% of the annotated gene promoters when coverage is increased to ~19% of the total genome. With a different enzyme (CviQI), >80% of gene bodies were included by sequencing ~15% of the genome (Hsu et al., 2017). The enrichment for functional components of the genome may not be as extreme as in mammals, but it does allow for functional insights. For instance, Hsu et al. (2017) identified genes with tissue-specific expression associated with specific promoter DNA methylation profiles. In addition, complementary expression analyses with RNAseq or qPCR will complement RRBS datasets, potentially uncovering gene expression alterations associated with DNA methylation variation.

Figure 1.

Figure 1

Open in a new tab

Typically, studies in ecological epigenetics will be interested in methylation differences that are associated with two different environmental conditions as shown in this cartoon. Reduced representation bisulfite sequencing (RRBS) data can provide functional insight in non-model species (e.g. the left most RRBS locus), but detecting the genomic context of RRBS fragments is difficult in the absence of a reference genome. In addition, the number of cytosines contained per RBBS locus may be insufficient to call differentially methylated regions (DMRs), hence RRBS analyses will be limited to calling differentially methylated positions (DMPs). TE - transposable element.

IV. Limitations of RRBS methods for non-model species

Although RRBS can target functional regions of a genome, identification of the genomic context of RRBS loci in non-model species is typically restricted to fragments that have homology to annotated species (Fig. 1). This approach has known limitations since genes and their function evolve over time, which may prevent proper interpretation of the information obtained in non-model species (Pavey et al., 2012; Alvarez et al., 2015). In the absence of a reference genome, a reference transcriptome could help improve functional predictions. In practice however, even if a large portion of the RRBS fragments overlap with a well annotated transcriptome (e.g., ca 10,000 of 36,000 RRBS fragments in Spartina, Alvarez et al., unpublished; >40% of bsRADseq in Heliosperma; Trucchi et al., 2016) or can be annotated with known plant genes (10-15% of epiGBS loci in six grassland species, Van Moorsel et al., 2018), only some of the fragments actually overlap the 5’ end of genes (Alvarez et al., unpublished) where increased DNA methylation has been correlated with gene silencing (Niederhuth et al., 2016; Niederhuth & Schmitz, 2017). The functional relevance of gene body methylation outside of the 5’ end in plants appears to vary by context and across taxa (Niederhuth & Schmitz, 2017), and often gene body methylation is not correlated or only weakly correlated to gene expression (Bewick & Schmitz, 2017). While methylation of the promoter region is highly correlated with silencing, without a well-annotated reference genome of the target species or a close relative, it is difficult or even impossible to identify those RRBS fragments that overlap promoter regions. Further, obtaining sufficient coverage of the genome may be prohibitive for plants with average to large genomes.

RRBS approaches allow for more power than earlier methods to detect outlier loci or identify loci with high FST (e.g., Platt et al., 2015), which suggests selection on ecologically important genomic regions (see also McKinney et al., 2017 for a discussion of such an approach). Unlike DNA sequence information that is generally organised in large (e.g. several megabases) linkage blocks shaped by recombination, co-inheritance of epigenetic signals may be less stable and only locally restricted to one or a few neighboring genomic elements that may be missed in reduced representation methods. Similarly, for traits that are controlled by relatively few functional loci, these may simply be missed in RRBS. This problem is amplified in species with large and complex genomes.

A further difficulty for all RRBS approaches is the identification of differentially methylated regions (DMRs). Studies in a variety of species have found that methylation changes across large chromosomal stretches (DMRs) are more likely to influence transcriptional activity at nearby loci, and contribute to phenotypic change than at single cytosines (DMPs) (reviewed in Richards et al., 2017). However, DMRs are still difficult to define even in whole genome studies. The short fragments interrogated with RRBS methods (typically less than 500 bp) will contain only a few cytosine positions, and calling DMRs with statistical confidence with such data will be difficult in most cases (Fig. 1). Hence, most RRBS studies will be limited to calling DMPs, with their inherent stochasticity.

The evolutionary history of all plant genomes has been shaped by polyploidization and hybridization, which have played a major role in plant diversification and adaptation (e.g., Van de Peer et al., 2017). Sequencing approaches generally will require more sequencing depth to identify all copies of duplicated loci, particularly for those with high levels of heterozygosity, and this problem may be exacerbated with short fragments that have less discriminatory power. This means that the number of polymorphisms that define different copies at a given locus will grade into the number of polymorphisms that define different locations within the genome. This level of information about the genomic landscape is limited to very few non-model species, but could have profound impacts on understanding the importance of both sequence level and methylation variation for ecology and evolution.

Finally, unravelling the extent to which epigenetic variants are genetically controlled by or at least partly autonomous from genetic variants in different systems and environmental settings remains a key challenge for understanding the ecological and evolutionary role of epigenetic variation (Richards et al., 2017). In principle, this issue is better addressed with whole-genome data than with a reduced representation data. Reduced representation data provide adequate information to estimate overall genetic and epigenetic population structure, and thus to evaluate the overall correlation between the two. But in absence of full genome data RRBS is limited in helping pin point the genetic determinants of individual epigenetic variants.

V. Maximising the impact of RRBS in plants

RRBS approaches allow for great power to detect patterns of variation in the epigenomic landscape, either as heritable signals or as plastic responses to environmental conditions or experimental treatments. However, confirming local adaptation requires detecting specific patterns of organismal response in reciprocal transplant studies in the field or in experiments in controlled environments, ideally by controlling for any genetic influence (Richards et al., 2010). A careful design of experimental conditions is particularly important for assessing epigenetic variation, which has fractions that are environmentally or developmentally labile (Richards et al., 2010, 2017). Measuring epigenetic mechanisms with a proper experimental design can provide information about changes that underlie phenotypic plasticity, and identify changes that are correlated to genetic differences or habitat of origin. Analyzing progeny across at least three generations will also better assess transgenerational heritability of the markers surveyed, but careful consideration of the environment in which to grow these offspring is required.

The flexibility of the RRBS protocols allows for several technical ways to improve impact. For instance, the choice of restriction enzymes determines what portions of the genome are sampled. Methylation-sensitive enzymes will target areas in plant genomes that are biased away from heavily methylated repetitive regions, and thus enrich towards coding regions. But using these enzymes will increase the probability of missing data, particularly within DMRs since individuals that are methylated at the recognition site will not be represented. An enzyme with a GC-rich restriction site will also bias against repetitive regions since several transposable elements (TEs) are associated with regions of the genome that are more AT-rich (Le et al., 2000). Furthermore, more frequent cutters (with shorter recognition sequences) will increase representation of the genome regardless of genomic context. Improvements can also come from using platforms that produce longer reads (e.g. MiSeq instead of HiSeq) or by optimizing libraries to get information across larger regions with paired-end sequencing approaches (e.g., up to 800 bp for bsRADseq using Illumina). This increases the likelihood of annotation and calling of DMRs, and potentially identifying promoters.

DNA methylation analyses require some additional considerations since it can be difficult to distinguish among substitutions and epiallele changes in bisulfite sequencing. In particular, mutations of either C or G to an A or T can appear to be an epimutation since bisulfite treatment deaminates un-methylated cytosines to uracil, which then pairs with thymine through PCR amplification. A single reference genome does not provide the information needed to discriminate these possibilities. Because of the cost advantage of RRBS compared to WGBS, unconverted references for each accession can be sequenced even for large experimental designs for differentiating genetic variation from methylation variation among individuals.

VI. Conclusions

In the past few years, several RRBS methodologies have emerged, increasing the potential and broadening the scope for epigenetics studies in non-model species. While these methodologies improve resolution for marker-based approaches, they have limitations for functional conclusions in species that lack a good reference genome (Fig. 1). These limitations result from the difficulty of targeting the RRBS fragments towards the most functionally relevant contexts for DNA methylation: the promoter regions and the 5’ end of transcribed regions. Although approaches exist to enrich for these specific portions of the genome, generating a draft reference genome will be imperative to locate the promoter regions, and allow for better exploitation of RRBS data. Improving genomics resources in a variety of organisms is an essential next step for understanding the importance of epigenetic mechanisms in ecology and evolution.

Acknowledgements

We would like to acknowledge the organisers and attendees of the 40th New Phytologist Symposium on “Plant epigenetics: from mechanisms to ecological relevance” for seeding the present review. We are grateful to Mariano Alvarez, Marta Robertson, Emiliano Trucchi, Thomas van Gurp and Cornelis A. M. Wagemaker for their contribution to the ideas presented here. This work was supported by an Austrian Science Fund (FWF) project (Y661-B16) to O.P., funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 764965 to K.J.F.V., and funding from the National Science Foundation (U.S.A.) IOS-1556820 to C.L.R.

References

  1. Alonso C, Perez R, Bazaga P, Herrera CM. Global DNA cytosine methylation as an evolving trait: phylogenetic signal and correlated evolution with genome size in angiosperms. Frontiers in Genetics. 2015;6:4. doi: 10.3389/fgene.2015.00004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alvarez M, Schrey AW, Richards CL. Ten years of transcriptomics in wild populations: what have we learned about their ecology and evolution? Molecular Ecology. 2015;24:710–725. doi: 10.1111/mec.13055. [DOI] [PubMed] [Google Scholar]
  3. Bewick AJ, Schmitz RJ. Gene body DNA methylation in plants. Current Opinion in Plant Biology. 2017;36:103–110. doi: 10.1016/j.pbi.2016.12.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bossdorf O, Richards CL, Pigliucci M. Epigenetics for ecologists. Ecology Letters. 2008;1:106–115. doi: 10.1111/j.1461-0248.2007.01130.x. [DOI] [PubMed] [Google Scholar]
  5. Cortijo S, Wardenaar R, Colomé-Tatché M, Gilly A, Etcheverry M, Labadie K, Caillieux E, Hospital F, Aury JM, Wincker P, et al. Mapping the epigenetic basis of complex traits. Science. 2014;343:1145–1148. doi: 10.1126/science.1248127. [DOI] [PubMed] [Google Scholar]
  6. Dapp M, Reinders J, Bédiée A, Balsera C, Bucher E, Theiler G, Granier C, Paszkowski J. Heterosis and inbreeding depression of epigenetic Arabidopsis hybrids. Nature Plants. 2015;1 doi: 10.1038/nplants.2015.92. 15092. [DOI] [PubMed] [Google Scholar]
  7. Dubin MJ, Zhang P, Meng D, Remigereau MS, Osborne EJ, Paolo Casale F, Drewe P, Kahles A, Jean G, Vilhjálmsson B, et al. DNA methylation in Arabidopsis has a genetic basis and shows evidence of local adaptation. eLife. 2015;4:e05255. doi: 10.7554/eLife.05255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Eichten SR, Briskine R, Song J, Li Q, Swanson-Wagner R, Hermanson PJ, Waters AJ, Starr E, West PT, Tiffin P, et al. Epigenetic and genetic influences on DNA methylation variation in maize populations. Plant Cell. 2013;25:2783–2797. doi: 10.1105/tpc.113.114793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Foust CM, Preite V, Schrey AW, Verhoeven KJF, Richards CL. Genetic and epigenetic differences associated with environmental gradients in replicate populations of two salt marsh perennials. Molecular Ecology. 2016;25:1639–1652. doi: 10.1111/mec.13522. [DOI] [PubMed] [Google Scholar]
  10. Frommer M, McDonald LE, Millar DS, Collis CM, Watt F, Grigg GW, Molloy PL, Paul CL. A genomic sequencing protocol that yields a positive display of 5-methylcytosine residues in individual DNA strands. Proceedings of the National Academy of Sciences. 1992;89:1827–1831. doi: 10.1073/pnas.89.5.1827. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gu H, Smith ZD, Bock C, Boyle P, Gnirke A, Meissner A. Preparation of reduced representation bisulfite sequencing libraries for genome-scale DNA methylation profiling. Nature Protocols. 2011;6:468. doi: 10.1038/nprot.2010.190. [DOI] [PubMed] [Google Scholar]
  12. Gugger PF, Fitz-Gibbon S, PellEgrini M, Sork VL. Species-wide patterns of DNA methylation variation in Quercus lobata and their association with climate gradients. Molecular Ecology. 2016;25:1665–1680. doi: 10.1111/mec.13563. [DOI] [PubMed] [Google Scholar]
  13. Harris RA, Wang T, Coarfa C, Nagarajan RP, Hong C, Downey SL, Johnson BE, Fouse SD, Delaney A, Zhao Y, et al. Comparison of sequencing-based methods to profile DNA methylation and identification of monoallelic epigenetic modifications. Nature Biotechnology. 2010;28:1097–105. doi: 10.1038/nbt.1682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hofmeister BT, Lee K, Rohr NA, Hall DW, Schmitz RJ. Stable inheritance of DNA methylation allows creation of epigenotype maps and the study of epiallele inheritance patterns in the absence of genetic variation. Genome Biology. 2017;18:155. doi: 10.1186/s13059-017-1288-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hsu FM, Yen MR, Wang CT, Lin CY, Wang CJR, Chen PY. Optimized reduced representation bisulfite sequencing reveals tissue-specific mCHH islands in maize. Epigenetics & Chromatin. 2017;10:42. doi: 10.1186/s13072-017-0148-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Johannes F, Porcher E, Teixeira F, Saliba-Colombani V, Simon M, Agier N, Bulski A, Albuisson J, Heredia F, Audigier P, et al. Assessing the impact of transgenerational epigenetic variation on complex traits. PLoS Genetics. 2009;5:e1000530. doi: 10.1371/journal.pgen.1000530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kawakatsu T, Huang SC, Jupe F, Sasaki E, Schmitz RJ, Urich MA, Castanon R, Nery JR, Barragan C, He Y, et al. Epigenomic diversity in a global collection of Arabidopsis thaliana accessions. Cell. 2016;166:492–505. doi: 10.1016/j.cell.2016.06.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Le QH, Wright S, Yu Z, Bureau T. Transposon diversity in Arabidopsis thaliana. Proceedings of the National Academy of Sciences, USA. 2000;97:7376–7381. doi: 10.1073/pnas.97.13.7376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lea AJ, Altmann J, Alberts SC, Tung J. Resource base influences genome-wide DNA methylation levels in wild baboons (Papio cynocephalus) Molecular Ecology. 2016;25:1681–1696. doi: 10.1111/mec.13436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Li Y, Tollefsbol TO. DNA methylation detection: bisulfite genomic sequencing analysis. In: Tollefsbol T, editor. Epigenetics Protocols. Methods in Molecular Biology. Vol. 791. Humana Press; Totowa, New Jersey (USA): 2011. pp. 11–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lyko F. DNA methylation learns to fly. Trends in Genetics. 2001;17:169–172. doi: 10.1016/s0168-9525(01)02234-x. [DOI] [PubMed] [Google Scholar]
  22. McKinney GJ, Larson WA, Seeb LW, Seeb JE. RADseq provides unprecedented insights into molecular ecology and evolutionary genetics: comment on Breaking RAD by Lowry et al. (2016) Molecular Ecology Resources. 2017;17:356–361. doi: 10.1111/1755-0998.12649. [DOI] [PubMed] [Google Scholar]
  23. Meissner A, Gnirke A, Bell GW, Ramsahoye B, Lander ES, Jaenisch R. Reduced representation bisulfite sequencing for comparative high-resolution DNA methylation analysis. Nucleic Acids Research. 2005;33:5868–5877. doi: 10.1093/nar/gki901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Niederhuth CE, Bewick AJ, Ji L, Alabday M, Kim KD, Li Q, Rohr NA, Rambani A, Burke JM, Udall JA, et al. Widespread natural variation of DNA methylation within angiosperms. Genome Biology. 2016;17:194. doi: 10.1186/s13059-016-1059-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Niederhuth CE, Schmitz RJ. Putting DNA methylation in context: from genomes to gene expression in plants. Biochimica et Biophysica Acta. 2017;1860:149–156. doi: 10.1016/j.bbagrm.2016.08.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Olova N, Krueger F, Andrews S, Oxley D, Berrens RV, Branco MR, Reik W. Comparison of whole-genome bisulfite sequencing library preparation strategies identifies sources of biases affecting DNA methylation data. Genome Biology. 2018;19:33. doi: 10.1186/s13059-018-1408-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Pavey SA, Bernatchez L, Aubin-Horth N, Landry CR. What is needed for next-generation ecological and evolutionary genomics? Trends Ecology and Evolution. 2012;27:673–678. doi: 10.1016/j.tree.2012.07.014. [DOI] [PubMed] [Google Scholar]
  28. Platt A, Gugger P, Sork VL. Genome-wide signature of local adaptation linked to variable CpG methylation in oak populations. Molecular Ecology. 2015;24:3823–3830. doi: 10.1111/mec.13230. [DOI] [PubMed] [Google Scholar]
  29. Richards CL, Bossdorf O, Verhoeven KJF. Understanding natural epigenetic variation. New Phytologist. 2010;187:562–564. doi: 10.1111/j.1469-8137.2010.03369.x. [DOI] [PubMed] [Google Scholar]
  30. Richards CL, Schrey AW, Pigliucci M. Invasion of diverse habitats by few Japanese knotweed genotypes is correlated with high epigenetic differentiation. Ecology Letters. 2012;15:1016–1025. doi: 10.1111/j.1461-0248.2012.01824.x. [DOI] [PubMed] [Google Scholar]
  31. Richards CL, Alonso C, Becker C, Bossdorf O, Bucher E, Colomé-Tatché M, Durka W, Engelhardt J, Gaspar B, Gogol-Döring A, et al. Ecological plant epigenetics: Evidence from model and non-model species, and the way forward. Ecology Letters. 2017;20:1576–1590. doi: 10.1111/ele.12858. [DOI] [PubMed] [Google Scholar]
  32. Schrey AW, Alvarez M, Foust CM, Kilvitis HJ, Lee JD, Liebl AL, Martin LB, Richards CL, Robertson M. Ecological epigenetics: beyond MS-AFLP. Integrative and Comparative Biology. 2013;53:340–350. doi: 10.1093/icb/ict012. [DOI] [PubMed] [Google Scholar]
  33. Smith ZD, Gu H, Bock C, Gnirke A, Meissner A. High-throughput bisulfite sequencing in mammalian genomes. Methods. 2009;48:226–232. doi: 10.1016/j.ymeth.2009.05.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Schmitz RJ, He Y, Valdés-López O, Khan SM, Joshi T, Urich MA, Nery JR, Diers B, Xu D, Stacey G, Ecker JR. Epigenome-wide inheritance of cytosine methylation variants in a recombinant inbred population. Genome Research. 2013;23:1663–1674. doi: 10.1101/gr.152538.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Taudt A, Colome-Tatche M, Johannes F. Genetic sources of population epigenomic variation. Nature Review in Genetics. 2016;17:319–332. doi: 10.1038/nrg.2016.45. [DOI] [PubMed] [Google Scholar]
  36. Trucchi E, Mazzarella AB, Gilfillan GD, Lorenzo MT, Schönswetter P, Paun O. BsRADseq: screening DNA methylation in natural populations of non-model species. Molecular Ecology. 2016;25:1697–1713. doi: 10.1111/mec.13550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Van de Peer Y, Mizrachi E, Marchal K. The evolutionary significance of polyploidy. Nature Reviews Genetics. 2017;18:411. doi: 10.1038/nrg.2017.26. [DOI] [PubMed] [Google Scholar]
  38. van Gurp TP, Wagemaker NCAM, Wouters B, Vergeer P, Ouborg JNJ, Verhoeven KJF. epiGBS: reference-free reduced representation bisulfite sequencing. Nature Methods. 2016;13:322–324. doi: 10.1038/nmeth.3763. [DOI] [PubMed] [Google Scholar]
  39. Van Moorsel SJ, Schmid MW, Wagemaker CAM, van Gurp TP, Schmid B, Vergeer P. Evidence for rapid evolution in a grassland biodiversity experiment. BioRxiv. 2018 doi: 10.1101/262303. 262303. [DOI] [PubMed] [Google Scholar]
  40. Verhoeven KJF, vonHoldt BM, Sork VL. Epigenetics in ecology and evolution: what we know and what we need to know. Molecular Ecology. 2016;25:1631–1638. doi: 10.1111/mec.13617. [DOI] [PubMed] [Google Scholar]
  41. Wang S, Lv J, Zhang L, Dou J, Sun Y, Li X, Fu X, Dou H, Mao J, Hu X, et al. MethylRAD: a simple and scalable method for genome-wide DNA methylation profiling using methylation-dependent restriction enzymes. Open Biology. 2015;5 doi: 10.1098/rsob.150130. 150130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Weyrich A, Lenz D, Jeschek M, Chung TH, Rübensam K, Göritz F, Jewgenow K, Fickel J. Paternal intergenerational epigenetic response to heat exposure in male Wild guinea pigs. Molecular Ecology. 2016;25:1729–1740. doi: 10.1111/mec.13494. [DOI] [PubMed] [Google Scholar]

RESOURCES