Individuals may delay dispersing from their natal habitat, even after maturation to adulthood. Such delays can have broad consequences from determining population structure to allowing an individual to gain indirect fitness by helping parents rear future offspring. Dispersal in species that use delayed dispersal is largely thought to be opportunistic; however, how individuals, particularly inexperienced juveniles, assess their environments to determine the appropriate time to disperse is unknown. One relatively unexplored possibility is that dispersal decisions are the result of epigenetic mechanisms interacting between a genome and environment during development to generate variable dispersive phenotypes. Here, we tested this using epiRADseq to compare genome-wide levels of DNA methylation of blood in cooperatively breeding chestnut-crowned babblers (Pomatostomus ruficeps). We measured dispersive and philopatric individuals at hatching, before fledging, and at 1 year (following when first year dispersal decisions would be made). We found that individuals that dispersed in their first year had a reduced proportion of methylated loci than philopatric individuals before fledging, but not at hatching or as adults. Further, individuals that dispersed in the first year had a greater number of loci change methylation state (i.e. gain or lose) between hatching and fledging. The existence and timing of these changes indicate some influence of development on epigenetic changes that may influence dispersal behavior. However, further work needs to be done to address exactly how developmental environments may be associated with dispersal decisions and which loci in particular are manipulated to generate such changes.
Citation: Liebl AL, Wesner JS, Russell AF, Schrey AW (2021) Methylation patterns at fledging predict delayed dispersal in a cooperatively breeding bird. PLoS ONE 16(6): e0252227. https://doi.org/10.1371/journal.pone.0252227
Editor: Charles R. Brown, University of Tulsa, UNITED STATES
Received: December 7, 2020; Accepted: May 11, 2021; Published: June 4, 2021
Copyright: © 2021 Liebl et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data and accompanying code are accessible at https://github.com/jswesner/liebl_birds and is also be permanently archived on Zenodo at https://zenodo.org/record/4553650#.YKB3z2NOkb0.
Funding: This project was funded by the Natural Environment Research Council to AFR (NE/K005766), USD’s Arts and Sciences to ALL, the National Science Foundation (1504662) to AWS, and GSU-Armstrong’s Department of Biology and College of Science and Mathematics to AWS. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
How animals move and disperse has broad implications for both the individual (e.g. habitat) and the population (e.g. population structure). Dispersal itself is likely influenced by a myriad of factors relating to individual internal and external environment, including the genome [1–4]. Indeed, research suggests a relationship between dispersal and particular gene types such as lipid metabolism and antigen defense . In some species, family groups are dependent on sexually mature individuals delaying dispersal from their natal habitat (herein referred to as delayed dispersal) to remain with breeding parents and delay their own breeding. Such dispersal decisions have additional implications for the ecology and evolution of the species as a whole- from population dynamics and structure to the evolution of cooperative breeding . Individuals may delay or forego dispersal when the benefits of remaining philopatric outweigh those of dispersing; these benefits might include the indirect fitness gained by helping rear relatives, enhanced foraging efficiency, reduced predation risk, the probability of inheriting a high-quality breeding position, or even parental practice . However, even within a species with delayed dispersal, individuals can vary in how long they postpone independence, and the timing of dispersal has been thought to be largely driven by post-developmental opportunism, in that offspring remain with their natal group until a suitable breeding opportunity arises [8–14]. Therefore, environmental cues likely play a large role in shaping how and when individuals disperse from their natal environments. However, how environmental cues are assessed and synthesized to determine the appropriate timing of dispersal, particularly by inexperienced juveniles, is largely unknown.
Research across taxa suggests that epigenetic processes, such as DNA methylation, are one way environments interact with the genome to generate variable phenotypes [15–22]. Epigenetic changes can strengthen, weaken, or entirely silence gene expression in response to some environmental stimuli, imparting consequences for how a genome is expressed into a phenome. Although some epigenetic changes occur rapidly in response to changing environments (including social environment [19, 23]), others are stable through time, allowing exposure to a single environment to induce an epigenetic change that can dictate a specific phenotypic trajectory throughout life . In fact, differences in developmental environment can induce variable social and parental behavior via epigenetic mechanisms [25–27]. Even the reproductive castes that drive eusociality (an extreme version of cooperative behavior) in honeybees are regulated epigenetically via alterations to patterns of DNA methylation by the consumption of royal jelly early in life . However, more research is needed on the impact of developmental environment on differential methylation and how that relates to variation in dispersal behavior and its consequences.
Here, we test whether or not molecular epigenetic interactions between the environment and the genome influence dispersal decisions in a cooperatively breeding vertebrate. Delayed dispersal is particularly important in cooperatively breeding species [6, 28] as 96% of the bird species in which offspring remain with their parents into adulthood also cooperatively breed . Individuals of cooperatively breeding species have three options on reaching adulthood: disperse to breed; disperse without breeding; or, delaying dispersal, remaining philopatric and helping . If individuals that remain gain indirect fitness benefits by helping their parents raise future offspring, and this fitness benefit outweighs the fitness that may be gained by dispersing, cooperative breeding should evolve [30, 31]. Delaying dispersal to help is most likely to evolve when available high quality habitats are rare  and/or when environmental predictability is low , making breeding without help unlikely to be successful.
Here we use the chestnut-crowned babbler (Pomatostomus ruficeps), a bird endemic to arid and semi-arid Australia, where breeding relies on seasonal rainfall. This species does not rely on limited resources (e.g. nesting holes) for breeding or survival, are slow to colonize vacant habitat, and are not observed alone, outside social groups . Although among individual variation exists in timing of dispersal in this species , there are no obvious socio-ecological predictors , and some individuals never disperse. Breeding groups consist of a single breeding female, between one and three breeding males, and between two and 15 (mean = 6) non-breeding adults that care for offspring by provisioning . Non-breeding helpers are often offspring that have delayed dispersal , highlighting the importance of delayed dispersal in cooperative behavior for this species. As a mechanism that interacts between variable environments and the genome, we expected that differences in genome-wide levels of methylation would predict dispersal behavior in chestnut-crowned babblers. Further, we anticipated that these differences in methylation would arise during development, such that no differences in methylation between individuals that dispersed and those that remained philopatric would be evident at hatching but would be before chicks fledged the nest. Understanding how individuals make decisions on when to disperse has implications for understanding how family groups are formed, which may lend clarity to the evolution of cooperative breeding in this and other species.
Materials and methods
Chestnut-crowned babblers have been studied at Fowlers Gap in New South Wales, Australia, a research station owned by the University of New South Wales, since 2004. Each year, breeding is monitored across the field site, all adults are captured, measured, and bled, and chicks are measured and bled one to three times before fledging. Although population size fluctuates by year, we monitor around 500 individuals per year, on average. For this study, samples were collected during the breeding seasons of 2014 and 2015. This study was conducted in accordance with Animal Behaviour’s guidelines for ethical treatment. The protocol was approved by Macquarie University’s Animal Care and Ethics Committee (license no. 06/40A), the NSW National Parks and Wildlife Service, and the Australian Bat and Bird Banding Scheme.
For this study, we used blood samples from individuals at hatching, just before fledging, and adults. Samples were chosen for individuals who had all necessary samples after adult phenotype (dispersive or philopatric) was known to ensure individuals were not related. All blood samples were stored in 95% ethanol at room temperature until DNA extraction. To sample chicks at hatching, eggs were removed from the nest during the latter half of the incubation period, replaced with model eggs, and artificially incubated in a rocking incubator (Brinsea Octogon 20) at 37.5°C and 42–45% humidity until hatching. Within an hour of hatching, chicks were weighed, had a blood sample taken, and toenails clipped for individual identity. Chicks were then returned to the nest and ringed at 10d, before the toenail clip grew out. A second blood sample and morphological measurements were taken at ~15d. After samples were collected, chicks were returned to the nest and allowed to fledge naturally; chestnut-crowned babbler chicks fledge between 18 and 21d. Finally, adults were captured with mistnets when morphological measurements and blood samples were collected.
DNA was extracted from preserved blood samples using Gentra PureGene DNA (Qiagen) extraction kit according to manufacturer’s instructions. Methylation was determined using an epiRADseq approach on an Ion Torrent Personal Genome Machine (Thermo Fisher Scientific) [36, 37]. epiRADseq is a reduced-representation library based approach to look at genome wide levels of methylation . It uses a combination of next-generation sequencing technology with the methylation sensitive restriction enzyme HpaII (New England Biolabs) to measure DNA methylation by read count variation in resulting fragments. Significant differences in loci recovery indicate differences in methylation across individuals. Because the epiRADseq method is based on methylation-sensitive enzymes that do not cut the genome, and therefore do not produce a sequenced read, when a cut site is methylated, greater sequencing depth at a locus indicates lower methylation at that locus. Although genotypic variation in addition to differences in methylation can account for variation in loci sequencing, as genotypes are consistent within individual over time, any comparison looking at the same individual would not run into these problems. Although we acknowledge that other techniques exist to measure DNA methylation that are better suited for the identification of specific genes or sites of methylation (i.e. bisulfite sequencing based methods), we chose epiRADseq because, at the time of this study, no reference genome was available for chestnut-crowned babbler, thus epiRADseq maximized our efficiency to generate whole-genome levels of methylation to screen differences among and within individuals. After restriction digestion with HpaII, we ligated double stranded barcoded adaptors and y-adaptors of the Ion Torrent IonXpress sequences. We conducted emulsion PCR using the Ion PGM-Hi-Q-View OT2-200 kit following manufacturer’s directions on the Ion Express OneTouch2 platform. We then sequenced resulting fragments with the Ion PGM-Hi-Q-View Sequencing 200 Kit using an Ion 316v2 BC Chip following the manufacturer’s protocol.
We determined genome wide levels of methylation in 57 samples from 24 individuals. Of those, eight individuals (four philopatric individuals and four dispersive individuals) were sequenced at the three time points. Individuals were considered philopatric if they remained with their natal social group as adults and dispersive if they moved to become part of an unrelated group in their first year. All individuals originated from unrelated social groups of between three and 14 adults (mean ± SD = 6.5 ± 3.3 with no difference in social group size between dispersive and philopatric individuals: t = -0.5, p = 0.85). Further, birds used did not vary in mass at fledging (mean ± SD = 38.3 ± 1.1 (philopatric) and 37.2 ± 1.0 (dispersers); t = 2.08, p = 0.13).
We demultiplexed epiRADseq runs and performed quality control (QC) using Torrent Suite version 4.4.3. We then trimmed adaptor sequences (retaining 50–150 bp) and performed a de novo assembly using Geneious v. 11.1.4, thus constructing a pseudo-reference sequence from the generated assembly. We mapped individual fragments with BWA [38, 39] and used featureCounts  to determine read counts of fragments for 50 bp bins. To generate the proportion of loci that were methylated at each time point for each individual, we collapsed read count data into a binary matrix to compare methylated loci (as zero; no fragments observed) with unmethylated loci (as one; at least one fragment observed). To determine how methylation status of each loci changed between ages within an individual (i.e. gained or lost methylation), we analyzed the difference in sequence reads at each fragment and determined how that changed through time.
We compared total proportion of methylated loci to age and dispersal status using a Bayesian generalized linear mixed model (GLMM) with a beta likelihood and a logit link (S1 File). Further, we determined how methylation state changed over time with a GLMM with a binomial likelihood and log link (S1 File). In all models, individual was a random effect to account for repeat sampling of individuals. Prior values for both models were chosen using prior predictive checks to ensure prior predictions generated reasonable upper and lower bounds  (S1 File and S1 Fig). Both models were fitted using the brms package  via Hamiltonian Monte Carlo in rstan . We ran four chains of 2000 iterations each. The first 1000 iterations of each chain were discarded as warmup. We used posterior predictive checks to test model performance  and checked chains for convergence by ensuring r-hats were less than 1.1. Finally, using the larger dataset (n = 24; 11 females, 13 males), we found no difference in the proportion of genome methylation between males and females (S2 Fig) and thus sex was not included in the models.
We determined how methylation state changed over time similarly using a GLMM with a binomial likelihood and log link. In this model, the response variable was the number of loci that changed methylation status (gained or lost methylation) out of the total number of loci examined. The predictor variables were dispersal status, age transition (i.e. from hatchling to fledgling or from fledgling to adult), and their interaction with individual as a random effect. Priors were N(0, 1) for the intercept, N(0, 1) for the betas, and Cauchy(0, 1) for the standard deviation.
As prior specifications become increasingly important in models with small samples sizes, as those used here, we conducted a prior sensitivity analysis  to ensure we were not over-generalizing our results from a small sample size (S1 File, S3 and S4 Figs). All data and code are available at gttps://github.com/jswesner/liebl_birds.
Following QC and trimming, the sequence reads formed a pseudo-reference sequence of 21,335,109 bases. Mapping to the pseudo-reference sequence resulted in 10,138 to 27,817 unique mapped reads per individual (i.e. loci). After binning, we removed all loci with a sequenced fragment in less than 10% of screened individuals generating 40,361 bins, which were treated as loci and were used for subsequent analysis.
Genome wide levels of methylation
At hatching, individuals that ultimately dispersed had similar proportions of genome-wide methylation as individuals that remained with their natal group (mean ± SD of the posterior distribution = 0.75 ± 0.02 and 0.77 ± 0.03 respectively; Fig 1; Table 1). In contrast, at fledging, dispersers had a lower average proportion of methylated loci (0.7 ± 0.03 compared to 0.78 ± 0.02 methylated loci in philopatric individuals; Fig 1; Table 1). This corresponded to a mean difference of 0.08 ± 0.04 (95% CrI: 0.01 to -0.15) with a >96% probability that dispersive individuals had lower genome wide methylation than philopatric individuals at fledging. By the adult stage, the difference between philopatric and dispersive individuals had reversed (0.8 ± 0.02 in dispersive individuals versus 0.74 ± 0.03 for philopatric individuals; Fig 1; Table 1), with a 96% probability that dispersive individuals had higher genome-wide methylation than philopatric individuals as adults.
Boxplots represent summaries of the posterior distribution, lines connect estimated values in each stage across 1000 iterations of the posterior distribution, and data points are the raw data.
Changes in methylation over time
Although most (78%, on average) loci did not gain or lose methylation between sampling (i.e. between hatching and fledging or between fledging and adult), dispersers had a greater number of loci that either changed their methylation by either gaining or losing methylation than philopatric birds, particularly between hatching and fledging (Table 2). Specifically, 24 ± 2% of the loci in dispersers either gained or lost methylation between hatching and fledging, whereas only 20 ± 1% of loci changed in philopatric birds (Fig 2); the probability that this difference was greater than zero was >96%. Interestingly, of the loci that changed between hatching and fledging, 897 loci gained methylation and 375 loci lost methylation across all dispersive individuals, whereas only 79 loci gained and 38 loci lost methylation across all philopatric individuals. Between fledging and 1 year, changes in methylation were more comparable between dispersers and philopatric individuals (24 ± 2% and 23 ± 2% respectively), with only a 60% probability that this difference was greater than zero.
Boxplots represent summaries of the posterior distribution for dispersing and philopatric birds and data points are the raw data.
Epigenetic changes during development are influential in how phenotypes are expressed throughout life. One example, in eusocial insects, has shown that the propensity to cooperate or breed is driven by methylation derived during development . In another cooperative breeder, superb starlings (Lamprotornis superbus), differential methylation driven by differences in developmental environment influenced adult fitness levels in males . However, whether changes in methylation influence delayed dispersal decisions, and by extension, cooperation, is yet unknown. Reduced methylation of the genome, allowing greater flexibility of gene expression, can increase phenotypic plasticity of certain traits [15, 19, 45]; this may be particularly beneficial for individuals with an unpredictable future environment, such as those expected for dispersing individuals. On the other hand, as social, and often physical, environments do not change between fledging and adulthood in birds that remain philopatric, those individuals that remain with their natal group may not benefit as much from increased plasticity as those dispersing into an unknown environment. We show that genome-wide levels of methylation in the blood were lower in offspring that dispersed than those that remained philopatric in their first year. Despite no difference at hatching, methylation status changed at a greater number of loci in dispersers between hatching and fledging resulting in lower levels of overall methylation just before fledging. Further, the number of loci that either gained or lost methylation across all dispersers, but in not philopatric individuals, was higher than those found in philopatric but not dispersive individuals. Thus, the epigenetic changes derived during development may prepare offspring for their future environments- whether they be predictable for philopatric individuals or unpredictable for dispersive individuals. Although we cannot yet determine the specific environmental cues generating these differences, given the amount of change seen during the post-hatching, pre-fledging period, the results suggest this developmental window may be particularly important for individuals in future dispersal decisions.
Research in non-cooperatively breeding species suggests a relationship between dispersal and particular gene types such as lipid metabolism and antigen defense . It is likely that some of the differentially methylated loci identified here may also be related to dispersal. However, the epiRADseq approach used does not quantify differential methylation at particular genes; rather, it measures broad differences in methylation patterns across the whole genome, and has been used to measure general, ecological epigenetic patterns across a range of systems [37, 46, 47]. Despite the clear differences we report between dispersive and philopatric individuals in the genome-wide levels of methylation at fledging, we are unable to elucidate whether and which of the differentially methylated loci are found in functionally relevant coding regions of the genome. Future work should use more targeted approaches such as bisulfite sequencing to test for more specific, gene-level differences in methylation. Such studies are still rare in ecological research, but they are important to identify the specific mechanisms by which methylation can influence phenotypes. Additionally, although the epigenetic patterns derived using blood samples (particularly those primarily from nucleated red blood cells as would be expected from the avian samples collected here) might not be representative of all tissues [48–50], some studies have shown correlations between methylation patterns in the blood and other tissues [51, 52]. Further, using peripheral samples has successfully revealed links between epigenetic variation and behavior in other species (primarily humans) [53–56]. Using these peripheral samples has been shown to be informative in describing the link between the environment and the genome and, perhaps more importantly, allows repeat sampling of the same individual over time to analyze how environments are related to changes in methylation and what the subsequent influence on phenotype is.
Dispersal can be influenced by a myriad of factors relating to individual internal and external environment including the genome [1–4], hormonal state [57, 58], personality , and condition [60–62]. Many environmental cues experienced during development influence behavioral phenotypes throughout life (reviewed in ), including dispersal [3, 64–66]; importantly, many of these phenotypes are derived via epigenetic changes to the genome. Notably, cooperative breeding is more prevalent in species living in harsh and unpredictable environments ; further, harsh environments may be expected to play an outsized role in shaping environmentally influenced phenotypes. Although we show no evidence of sex differences in methylation patterns between males and females here, some evidence exists that developmental environment can differentially influence methylation dependent on sex . As with most avian cooperative systems, female offspring tend to be more dispersive whereas males tend to be more philopatric [8, 34]. Here, although dispersive individuals were split between males and females, the philopatric group included only males. Future work might address the interaction between sex and developmental environment on adult behavior.
One particularly important environmental cue experienced during development is parental care [27, 67]; the amount and quality of parental care can vary both within (with parents additional care dependent on sex, age, size, or other offspring attributes [68, 69]) and among broods. Variation in parental care is especially high in cooperatively breeding species like chestnut-crowned babblers, where offspring are raised by a variable number of adults . Although the links between variable parental care and methylation patterns and adult phenotype is unknown, given the amount of variation in parental care demonstrated across offspring, particularly in cooperatively breeding societies, it is possible that it plays an important role in developmentally-determined phenotypes. However, in addition to parental care during the nestling period, many other developmental environments are experienced by chicks, including (but not limited to) territorial quality, the post-fledging social environment, environmental stability, and climatatic variables during dispersal; all of which may also contribute to dispersal decisions.
If differential methylation is indeed driven by variation in parental care in chestnut-crowned babblers, this could reveal a molecular mechanism by which parents and helpers could adaptively and plastically adjust offspring phenotypes to best match the needs of the offspring as well as the group. Given the numerous, and often conflicting cues individuals experiences aduring development, predicting offspring disperal is not only a challenge for scientists, but also for the animals themselves. Using molecular mechanisms such as epigenetic processes that interact with both the environment and genome may help guide adaptive decions regarding whether or not to disperse.
Ecological epigenetics is a relatively new field of study and thus many of the techniques long available to medical researchers are just becoming accessible to molecular ecologists. Despite this, these results fit into a growing body of evidence indicating that differences in epigenetic markers can lead to predictable phenotypic variation across non-model organisms. Although we do not yet know what specific genes are being epigenetically controlled here, or what specific environmental cues are driving molecular differences between dispersive and philopatric individuals, this research paints a picture showing that epigenetic marks, particularly those derived during nestling development, may be important in determining dispersal decisions in chestnut-crowned babblers. Future research should focus on identifying the specific environmental cues that trigger variation in methylation as well as the specific genes that are differentially methylated between dispersive and philopatric individuals. Because dispersal decisions can alter the fitness landscape of cooperatively breeding species, understanding the molecular mechanisms that drive dispersal, and how those mechanisms are derived may shed light on how such a seemingly paradoxical behavior has evolved.
S1 Fig. Prior and posterior comparisons.
The top panels show the model results before including the data (i.e. generated only from the prior distribution). The bottom panels show the same models, but now including the data (i.e. generated from the posterior distribution. Circles are means and error bars are 95% CrI.
S2 Fig. Proportion of methylated loci in male and female birds.
Boxplots summarize the posterior distribution from a Beta GLMM, and dots are the raw data.
S3 Fig. Parameter estimates from a prior sensitivity analysis.
The model presented in the main text has a prior standard deviation of 1, N(0,1), and is shown with the large green circle. The other models contain parameter estimates after adjusting the prior and re-running the model. Values less than 1 are more restrictive priors. Values greater than 1 are less restrictive priors compared to the prior for the main model. Values for the yellow triangle represent parameter estimates from maximum likelihood using the lme4 package. These results are roughly akin to running a Bayesian model with a standard deviation of infinity–Inf (Max Lik) on all priors.
S4 Fig. Conditional effects from a prior sensitivity analysis.
The model presented in the main text—1 (MS)—has a prior standard deviation of 1 (N(0,1)). The other models show alternative results after adjusting the prior and re-running the model. Values less than 1 are more restrictive priors. Values greater than 1 are less restrictive priors compared to the prior for the main model. Results from maximum likelihood use the lme4 package–Inf (Max Lik). Maximum likelihood results are roughly akin to running a Bayesian model with a standard deviation of infinity on all priors.
We would like to thank Simon Griffith, Keith Leggett, the Dowling family, Tom Harris, Niall Stopford, Alex Cones, Sarah Biddescombe, and Matt Creasey for help in the field and Haley Kodak, Katherine Miller, Alex Ragsdale, Kurt Russel, and Natalie Schrey for assistance in the lab.
- 1. Saastamoinen M, Bocedi G, Cote J, Legrand D, Guillaume F, Wheat CW, et al. Genetics of dispersal. Biological Reviews. 2018;93(1):574–99. pmid:28776950
- 2. Pasinelli G, Schiegg K, Jeffrey . Genetic and Environmental Influences on Natal Dispersal Distance in a Resident Bird Species. The American Naturalist. 2004;164(5):660–9. pmid:15540155
- 3. Sinervo Barry, Calsbeek Ryan, Comendant Tosha, Both Christiaan, Adamopoulou Chloe, Clobert Jean. Genetic and Maternal Determinants of Effective Dispersal: The Effect of Sire Genotype and Size at Birth in Side‐Blotched Lizards. The American Naturalist. 2006;168(1):88–99. pmid:16874616.
- 4. Hansson B, Bensch S, Hasselquist D. Heritability of dispersal in the great reed warbler. Ecology Letters. 2003;6(4):290–4.
- 5. Armenta TC, Cole SW, Geschwind DH, Blumstein DT, Wayne RK. Gene expression shifts in yellow-bellied marmots prior to natal dispersal. Behavioral Ecology. 2019;30(2):267–77. pmid:30971856
- 6. Griesser M, Drobniak SM, Nakagawa S, Botero CA. Family living sets the stage for cooperative breeding and ecological resilience in birds. PLOS Biology. 2017;15(6):e2000483. pmid:28636615
- 7. Koenig WD, Pitelka FA, Carmen WJ, Mumme RL, Staback MT. The evolution of delayed dispersal in cooperative breeders. The Quarterly Review of Biology. 1992;67(2):111–50. pmid:1635976
- 8. Koenig WD, Dickinson JL. Ecology and evolution of cooperative breeding in birds: Cambridge University Press; 2004.
- 9. Pruett-Jones S, Lewis M. Sex ratio and habitat limitation promote delayed dispersal in superb fairy-wrens. Nature. 1990;348(6301):541–2.
- 10. Du Plessis MA. Obligate cavity-roosting as a constraint on dispersal of green (red-billed) woodhoopoes: consequences for philopatry and the likelihood of inbreeding. Oecologia. 1992;90(2):205–11. pmid:28313715
- 11. Walters JR, Copeyon CK, Carter Iii J. Test of the ecological basis of cooperative breeding in red-cockaded woodpeckers. The Auk. 1992;109(1):90–7.
- 12. Heg D, Bachar Z, Brouwer L, Taborsky M. Predation risk is an ecological constraint for helper dispersal in a cooperatively breeding cichlid. Proceedings of the Royal Society of London Series B: Biological Sciences. 2004;271(1555):2367–74. pmid:15556889
- 13. Komdeur J, Huffstadt A, Prast W, Castle G, Mileto R, Wattel J. Transfer experiments of Seychelles warblers to new islands: changes in dispersal and helping behaviour. Animal Behaviour. 1995;49(3):695–708.
- 14. Covas R, Griesser M. Life history and the evolution of family living in birds. Proceedings of the Royal Society B: Biological Sciences. 2007;274(1616):1349–57. pmid:17374593
- 15. Bossdorf O, Arcuri D, Richards CL, Pigliucci M. Experimental alteration of DNA methylation affects the phenotypic plasticity of ecologically relevant traits in Arabidopsis thaliana. Evolutionary Ecology. 2010;24(3):541–53.
- 16. Zhang Y-Y, Fischer M, Colot V, Bossdorf O. Epigenetic variation creates potential for evolution of plant phenotypic plasticity. New Phytologist. 2013;197:314–22. pmid:23121242
- 17. Dolinoy DC, Huang D, Jirtle RL. Maternal nutrient supplementation counteracts bisphenol A-induced DNA hypomethylation in early development. Proceedings of the National Academy of Sciences. 2007;104(32):13056–61. pmid:17670942
- 18. Kucharski R, Maleszka J, Foret S, Maleszka R. Nutritional control of reproductive status in honeybees via DNA methylation. Science. 2008;319(5871):1827–30. pmid:18339900
- 19. Angers B, Castonguay E, Massicotte R. Environmentally induced phenotypes and DNA methylation: how to deal with unpredictable conditions until the next generation and after. Molecular Ecology. 2010;19(7):1283–95. pmid:20298470
- 20. Vogt G, Huber M, Thiemann M, van den Boogaart G, Schmitz OJ, Schubart CD. Production of different phenotypes from the same genotype in the same environment by developmental variation. Journal of Experimental Biology. 2008;211(4):510–23. pmid:18245627
- 21. Gao L, Geng Y, Li B, Chen J, Yang J. Genome-wide DNA methylation alterations of Alternanthera philoxeroides in natural and manipulated habitats: implications for epigenetic regulation of rapid responses to environmental fluctuation and phenotypic variation. Plant, Cell & Environment. 2010;33(11):1820–7. pmid:20545886
- 22. Richards CL, Bossdorf O, Pigliucci M. What Role Does Heritable Epigenetic Variation Play in Phenotypic Evolution? BioScience. 2010;60(3):232–7.
- 23. Verhoeven KJ, Van Dijk PJ, Biere A. Changes in genomic methylation patterns during the formation of triploid asexual dandelion lineages. Molecular Ecology. 2010;19(2):315–24. pmid:20015141
- 24. Rubenstein DR, Skolnik H, Berrio A, Champagne FA, Phelps S, Solomon J. Sex-specific fitness effects of unpredictable early life conditions are associated with DNA methylation in the avian glucocorticoid receptor. Molecular Ecology. 2016:n/a-n/a. pmid:26588348
- 25. Provenzi L, Brambilla M, Scotto di Minico G, Montirosso R, Borgatti R. Maternal caregiving and DNA methylation in human infants and children: Systematic review. Genes, Brain and Behavior. 2020;19(3):e12616. pmid:31622002
- 26. Beery AK, McEwen LM, MacIsaac JL, Francis DD, Kobor MS. Natural variation in maternal care and cross-tissue patterns of oxytocin receptor gene methylation in rats. Hormones and Behavior. 2016;77:42–52. pmid:26122287
- 27. Weaver ICG, Cervoni N, Champagne FA, D’Alessio AC, Sharma S, Seckl JR, et al. Epigenetic programming by maternal behavior. Nature Neuroscience. 2004;7(8):847–54. pmid:15220929
- 28. Lukas D, Clutton-Brock T. Cooperative breeding and monogamy in mammalian societies. Proceedings of the Royal Society B: Biological Sciences. 2012;279(1736):2151–6. pmid:22279167
- 29. Emlen ST. An evolutionary theory of the family. Proceedings of the National Academy of Science, USA. 1995;92:8092–9. pmid:7667250
- 30. Hamilton WD. The evolution of altruistic behavior. The American Naturalist. 1963;97(896):354–6.
- 31. Smith JM. Group selection and kin selection. Nature. 1964;201(4924):1145.
- 32. Russell A. Chestnut-crowned babblers: dealing with climatic adversity and uncertainty in the Australian arid zone. In: Koenig W, Dickinson J, editors. Cooperative Breeding in Vertebrates: Studies of Ecology, Evolution and Behaviour. Cambridge, MA: Cambridge University Press MA; 2016.
- 33. Rollins LA, Browning LE, Holleley CE, Savage JL, Russell AF, Griffith SC. Building genetic networks using relatedness information: a novel approach for the estimation of dispersal and characterization of group structure in social animals. Molecular Ecology. 2012;21(7):1727–40. pmid:22335253
- 34. Russell AF, Portelli DJ, Russell DJF, Barclay H. Breeding ecology of the Chestnut-crowned Babbler: a cooperative breeder in the desert. Emu. 2010;110:324–31.
- 35. Browning LE, Patrick SC, Rollins LA, Griffith SC, Russell AF. Kin selection, not group augmentation, predicts helping in an obligate cooperatively breeding bird. Proceedings of the Royal Society B: Biological Sciences. 2012;283(1823). pmid:22787025
- 36. Mascher M, Wu S, Amand PS, Stein N, Poland J. Application of genotyping-by-sequencing on semiconductor sequencing platforms: a comparison of genetic and reference-based marker ordering in barley. PloS one. 2013;8(10):e76925. pmid:24098570
- 37. Schield DR, Walsh MR, Card DC, Andrew AL, Adams RH, Castoe TA. Epi RAD seq: scalable analysis of genomewide patterns of methylation using next‐generation sequencing. Methods in Ecology and Evolution. 2016;7(1):60–9.
- 38. Li H, Durbin R. Fast and accurate long-read alignment with Burrows–Wheeler transform. Bioinformatics. 2010;26(5):589–95. pmid:20080505
- 39. Li H, Durbin R. Fast and accurate short read alignment with Burrows–Wheeler transform. bioinformatics. 2009;25(14):1754–60. pmid:19451168
- 40. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2013;30(7):923–30. pmid:24227677
- 41. Gabry J, Simpson D, Vehtari A, Betancourt M, Gelman A. Visualization in Bayesian workflow. Journal of the Royal Statistical Society: Series A (Statistics in Society). 2019;182(2):389–402.
- 42. Bürkner P-C. brms: An R package for Bayesian multilevel models using Stan. Journal of Statistical Software. 2017;80(1):1–28.
- 43. Team SD. RStan: the R interface to Stan, version 2.19.2. 2019.
- 44. Hobbs NT, Hooten MB. Bayesian models: a statistical primer for ecologists: Princeton University Press; 2015.
- 45. Kilvitis HJ, Hanson H, Schrey AW, Martin LB. Epigenetic Potential as a Mechanism of Phenotypic Plasticity in Vertebrate Range Expansions. Integrative and Comparative Biology. 2017;57(2):385–95. pmid:28859411
- 46. Hanson HE, Wang C, Schrey AW, Liebl AL, Jiang RHY, Martin LB. Epigenetic Potential and DNA Methylation in an Ongoing House Sparrow (Passer domesticus) Range Expansion. bioRxiv. 2020:2020.03.07.981886.
- 47. Dimond JL, Gamblewood SK, Roberts SB. Genetic and epigenetic insight into morphospecies in a reef coral. Molecular Ecology. 2017;26(19):5031–42. pmid:28753237
- 48. Siller SJ, Rubenstein DR. A tissue comparison of DNA methylation of the glucocorticoid receptor gene (Nr3c1) in European starlings. Integrative and comparative biology. 2019.
- 49. Husby A. On the Use of Blood Samples for Measuring DNA Methylation in Ecological Epigenetic Studies. Integrative and Comparative Biology. 2020. pmid:32835371
- 50. Schultz MD, He Y, Whitaker JW, Hariharan M, Mukamel EA, Leung D, et al. Human body epigenome maps reveal noncanonical DNA methylation variation. Nature. 2015;523(7559):212–6. http://www.nature.com/nature/journal/v523/n7559/abs/nature14465.html#supplementary-information. pmid:26030523
- 51. Masliah E, Dumaop W, Galasko D, Desplats P. Distinctive patterns of DNA methylation associated with Parkinson disease: identification of concordant epigenetic changes in brain and peripheral blood leukocytes. Epigenetics. 2013;8(10):1030–8. pmid:23907097
- 52. Farré P, Jones MJ, Meaney MJ, Emberly E, Turecki G, Kobor MS. Concordant and discordant DNA methylation signatures of aging in human blood and brain. Epigenetics & chromatin. 2015;8(1):19. pmid:25977707
- 53. Groleau P, Joober R, Israel M, Zeramdini N, DeGuzman R, Steiger H. Methylation of the dopamine D2 receptor (DRD2) gene promoter in women with a bulimia-spectrum disorder: Associations with borderline personality disorder and exposure to childhood abuse. Journal of Psychiatric Research. 2014;48(1):121–7. pmid:24157248
- 54. Puglia MH, Lillard TS, Morris JP, Connelly JJ. Epigenetic modification of the oxytocin receptor gene influences the perception of anger and fear in the human brain. Proceedings of the National Academy of Sciences. 2015;112(11):3308–13. pmid:25675509
- 55. Ursini G, Bollati V, Fazio L, Porcelli A, Iacovelli L, Catalani A, et al. Stress-Related Methylation of the Catechol-<em>O</em>-Methyltransferase Val<sup>158</sup> Allele Predicts Human Prefrontal Cognition and Activity. The Journal of Neuroscience. 2011;31(18):6692. pmid:21543598
- 56. Provençal N, Suderman MJ, Guillemin C, Vitaro F, Côté SM, Hallett M, et al. Association of Childhood Chronic Physical Aggression with a DNA Methylation Signature in Adult Human T Cells. PLOS ONE. 2014;9(4):e89839. pmid:24691403
- 57. Meylan S, Belliure J, Clobert J, de Fraipont M. Stress and body condition as prenatal and postnatal determinants of dispersal in the common lizard (Lacerta vivipara). Hormones and Behavior. 2002;42:319–26. pmid:12460591
- 58. Dufty AM, Belthoff JR. Proximate mechanisms of natal dispersal: the role of body condition and hormones. In: Clobert J, Danchin E, Dhondt AA, Nichols JD, editors. Dispersal. Oxford: Oxford University Press; 2001. p. 217–29.
- 59. Cote J, Clobert J, Brodin T, Fogarty S, Sih A. Personality-dependent dispersal: characterization, ontogeny and consequences for spatially structured populations. Philosophical Transactions of the Royal Society B: Biological Sciences. 2010;365(1560):4065–76. pmid:21078658
- 60. Russell AF, Young AJ, Spong G, Jordan NR, Clutton-Brock TH. Helpers increase the reproductive potential of offspring in cooperative meerkats. Proceedings of the Royal Society B: Biological Sciences. 2007;274(1609):513–20. pmid:17476771
- 61. Pasinelli G, Walters JR. Social and environmental factors affect natal dispersal and philopatry of male red‐cockaded woodpeckers. Ecology. 2002;83(8):2229–39.
- 62. Ekman J, Griesser M. Siberian jays: delayed dispersal in absence of cooperative breeding. Cooperative breeding in vertebrates: studies of ecology, evolution, and behavior Cambridge University Press, Cambridge. 2016:6–18.
- 63. Mousseau TA, Fox CW. The adaptive significance of maternal effects. Trends in Ecology and Evolution. 1998;13(10):403–7. pmid:21238360
- 64. Dingemanse NJ, Both C, van Noordwijk AJ, Rutten AL, Drent PJ. Natal dispersal and personalities in great tits (Parus major). Proceedings of the Royal Society of London Series B-Biological Sciences. 2003;270(1516):741–7. WOS:000182249200012. pmid:12713749
- 65. Tschirren B, Fitze PS, Richner H. Maternal modulation of natal dispersal in a passerine bird: an adaptive strategy to cope with parasitism? The American Naturalist. 2007;169(1):87–93. pmid:17206587
- 66. de Fraipont M, Clobert J, John-Adler H, Meylan S. Increased pre-natal maternal corticosterone promotes philopatry of offspring in common lizards Lacerta vivipara. Journal of Animal Ecology. 2000;69(3):404–13.
- 67. Szyf M, Weaver I, Meaney M. Maternal care, the epigenome and phenotypic differences in behavior. Reproductive Toxicology. 2007;24(1):9–19. pmid:17561370
- 68. Balshine S, Kempenaers B, Székely T, Hasselquist D, Kempenaers B. Parental care and adaptive brood sex ratio manipulation in birds. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences. 2002;357(1419):363–72. pmid:11958704
- 69. Slagsvold T, Røskaft E, Engen S. Sex Ratio, Differential Cost of Rearing Young, and Differential Mortality between the Sexes during the Period of Parental Care: Fisher’s Theory Applied to Birds. Ornis Scandinavica (Scandinavian Journal of Ornithology). 1986;17(2):117–25.
- 70. Liebl A, Nomano F, Browning L, Russell A. Experimental evidence for fully additive care among male carers in the cooperatively breeding chestnut-crowned babbler. Animal Behaviour. 2016;115:47–53.