All the instructions required for life are encoded in the set of DNA present in a cell. It therefore seems natural to think that every bit of this genetic information should serve the organism. And yet most species carry parasitic ‘transposable’ sequences, or transposons, whose only purpose is to multiply and insert themselves at other positions in the genome.

It is possible for cells to suppress these selfish elements. Chemical marks can be deposited onto the DNA to temporarily ‘silence’ transposons and prevent them from being able to move and replicate. However, this sometimes comes at a cost: the repressive chemical modifications can spread to nearby genes that are essential for the organism and perturb their function.

Strangely, the prevalence of transposons varies widely across the tree of life. These sequences form the majority of the genome of certain species – in fact, they represent about half of the human genetic information. But their abundance is much lower in other organisms, forming a measly 6% of the genome of puffer fish for instance. Even amongst fruit fly species, the prevalence of transposable elements can range between 2% and 25%. What explains such differences?

Huang et al. set out to examine this question through the lens of transposon silencing, systematically comparing how this process impacts nearby regions in six species of fruit flies. This revealed variations in the strength of the side effects associated with transposon silencing, resulting in different levels of perturbation on neighbouring genes. A stronger impact was associated with the species having fewer transposons in its genome, suggesting that an evolutionary pressure is at work to keep the abundance of transposons at a low level in these species. Further analyses showed that the genes which determine how silencing marks are distributed may also be responsible for the variations in the impact of transposon silencing. They could therefore be the ones driving differences in the abundance of transposons between species.

Overall, this work sheds light on the complex mechanisms shaping the evolution of genomes, and it may help to better understand how transposons are linked to processes such as aging and cancer.

Transposable elements (TEs) are widespread genetic parasites that copy and insert themselves across host genomes. The presence and movement of TEs could impair host genome functions. TEs disrupt genes and functional elements (Finnegan, 1992), introduce ectopic regulatory sequences (Chuong et al., 2017), and trigger highly deleterious chromosomal rearrangements through nonhomologous recombination (Langley et al., 1988; Montgomery et al., 1991). Nevertheless, the ability to self-replicate has allowed TEs to successfully occupy nearly all eukaryotic genomes surveyed (reviewed in Wells and Feschotte, 2020). Within a eukaryotic genome, TEs are prevalent in both gene-poor, repeat-rich heterochromatic and gene-rich euchromatic regions (e.g., Kaminker et al., 2002; Bergman et al., 2006). TEs in the heterochromatic genome are oftentimes fragmented, losing their ability to replicate (e.g., Hoskins et al., 2007; Hoskins et al., 2015). On the contrary, euchromatic TEs, which could intersperse with functional elements, commonly retain the potential to replicate, making them active players for not only their own evolutionary dynamics, but also the function and evolution of the euchromatic genome. Because of that and the technical challenges associated with identifying TEs in the heterochromatic regions (Salzberg and Yorke, 2005; Treangen and Salzberg, 2011), studies have been largely focused on the evolution of TEs in the euchromatic genome. Intriguingly, the abundance of TEs in the euchromatic genome substantially varies across the phylogenetic tree (Huang et al., 2012; Elliott and Gregory, 2015; Wells and Feschotte, 2020). For instance, in the assembled vertebrate genomes, which mostly consist of euchromatic regions, the proportion occupied by TEs ranges from only 6% in pufferfish (Volff et al., 2003) to more than 65% in salamander (Nowoshilow et al., 2018). Even within the same genus, genomic TE abundance differs widely (e.g., 2.5–25% of assembled genome sequences in Drosophila; Clark et al., 2007; Rius et al., 2016). Deciphering the role of this prevalent parasite in shaping genome evolution has remained a central question in genomics (Kazazian, 2004; Feschotte and Pritham, 2007; Arkhipova and Kumar, 2018); however, the ultimate causes of such dramatic divergence in TE abundance in the euchromatic genome remain unclear.

Theoretical analyses proposed that, in panmictic host populations with unrestricted recombination, TE abundance is determined by how quickly TEs replicate and how fast they are removed from the populations by natural selection against their harmful fitness effects (Charlesworth and Charlesworth, 1983, reviewed in Lee and Langley, 2010). Under this model, divergent genome-wide TE abundance could be driven by between-species differences in the strength of selection against TEs. Currently available evolutionary models that address this possibility have focused on population genetic parameters that influence the efficacy of selection removing TEs, such as mating systems (Wright and Schoen, 1999; Dolgin and Charlesworth, 2006; Boutin et al., 2012) and effective population size (Lynch and Conery, 2003). Yet, empirical support for such a hypothesis has been mixed (Dolgin et al., 2008; Lockton and Gaut, 2010; de la Chaux et al., 2012; Arunkumar et al., 2015; Agren et al., 2014; Mérel et al., 2021; Oggenfuss et al., 2021). On the other hand, between-species differences in the magnitude of harmful effects exerted by TEs, and accordingly the strength of selection against TEs, could also determine genomic TE abundance, a plausible hypothesis that is yet to have empirical investigations.

A new avenue for exploring how these genetic parasites shape the function and evolution of eukaryotic genomes was opened by the recently discovered host-directed silencing of TEs and the associated ‘inadvertent’ deleterious epigenetic effects (reviewed in Choi and Lee, 2020). To counteract the selfish increase of TEs in host genomes, eukaryotic hosts have evolved small RNA-mediated mechanisms to transcriptionally silence TEs (reviewed in Slotkin and Martienssen, 2007; Czech et al., 2018; Deniz et al., 2019). Host protein complexes are guided by small RNAs to TEs with complementary sequences, which is followed by the recruitment of methyltransferases that modify DNA or histone tails at TE sequences (Qi et al., 2006; Aravin et al., 2008; Wang and Elgin, 2011; Sienski et al., 2012; Le Thomas et al., 2013). Such a process results in the enrichment of DNA methylation or di- and tri-methylation on lysine 9 of histone H3 (H3K9me2/3), both repressive epigenetic modifications that are typically found in heterochromatic regions and associated with repressed gene expression (reviewed in Pikaard and Mittelsten Scheid, 2014; Allis and Jenuwein, 2016). This repressed transcription of TEs results in reduced RNA intermediates (for RNA-based TEs) and proteins (e.g., transposase and reverse transcriptase) necessary for TE replication, effectively slowing the selfish propagation of TEs.

While such epigenetic silencing of TEs should benefit their hosts, studies in various model species have found that repressive marks enriched at silenced TEs ‘spread’ beyond TE boundaries, leading to local enrichment of such marks at TE-adjacent sequences across the euchromatic genomes (i.e., Mus, Drosophila, Arabidopsis, and Oryza; Rebollo et al., 2012; Sienski et al., 2012; Pezic et al., 2014; Lee, 2015; Quadrana et al., 2016; Choi and Purugganan, 2018, reviewed in Choi and Lee, 2020). Furthermore, TEs with such effects were observed to have lower population frequencies (Hollister and Gaut, 2009; Lee, 2015; Lee and Karpen, 2017), suggesting that selection acts to remove them. These discoveries highlight the potential importance of TE-triggered epigenetic effects in shaping genome evolution. Interestingly, the strength of TE-mediated local enrichment of repressive marks substantially varies between distantly related taxa (reviewed in Choi and Lee, 2020). Investigations on pairs of closely related species further revealed that this ‘epigenetic effect of TE’ differs and is stronger in the species with fewer euchromatic TEs (Arabidopsis thaliana vs. Arabidopsis lyrata and Drosophila melanogaster vs. Drosophila simulans; Hollister et al., 2011; Lee and Karpen, 2017). These observations spurred our previous hypothesis that, across species, different TE-mediated enrichment of repressive marks could result in varying functional consequences and thus differences in the strength of selection against TEs, eventually contributing to divergent TE abundance in the euchromatic genome (Lee and Karpen, 2017). We further postulated that this difference in TE-mediated epigenetic effects could have resulted from species-specific genetic modulation of the repressive chromatin landscape (Lee and Karpen, 2017), which was shown to determine the spreading of repressive epigenetic marks from constitutive heterochromatin into adjacent euchromatic sequences (reviewed in Girton and Johansen, 2008; Elgin and Reuter, 2013).

To fully examine the hypothesis that varying host chromatin landscape drives between-species differences of TEs through epigenetic mechanisms, one needs to connect species-specific regulation of chromatin landscape, TE-mediated enrichment of repressive marks, the associated functional consequence and resultant selection against TEs, and genomic TE abundance. However, former analyses that compared TE-mediated epigenetic effects between species have limited sampling (two species) and thus lack sufficient statistical power for robust inference (Hollister et al., 2011; Lee and Karpen, 2017). Also, support for key links of the hypothesis is lacking. For instance, selection against TE-mediated epigenetic effects was expected to result from the associated reducing effects on the expression of neighboring genes. Yet, investigations in multiple taxa reported weak or no associations between the epigenetic effects of TEs and neighboring gene expression (Quadrana et al., 2016; Stuart et al., 2016; Lee and Karpen, 2017; Choi and Purugganan, 2018, reviewed in Kelleher et al., 2020; Choi and Lee, 2020). These inconclusive observations cast doubt on the possibility that TE-mediated epigenetic effects impair host fitness by silencing neighboring genes and whether this particular deleterious consequence indeed shapes genome evolution. Furthermore, previous comparisons of population frequencies between TEs with and without epigenetic effects, an approach used to infer the strength of natural selection removing TEs, could not exclude the confounding influence of other harmful effects of TEs on their population frequencies (e.g., Hollister and Gaut, 2009; Lee, 2015). Accordingly, those analyses could not unequivocally support selection against TE-mediated epigenetic effects. Multi-species studies that span an appropriate evolutionary distance and connect the missing links in the proposed hypothesis would be needed to test the predicted importance of TE-mediated epigenetic effects in determining between-species differences of TEs.

In this study, we investigated the prevalence of TE-mediated local enrichment of repressive epigenetic marks, or ‘TE-mediated epigenetic effects’, in the euchromatic genome of six species in the D. melanogaster subgroup (diverged around 10 MYR; Obbard et al., 2012, Figure 1A). These species are from the two well-studied species complexes (melanogaster and yakuba complexes), providing good phylogenetic resolution to address the role of TE-mediated epigenetic effects in genome evolution. While TE insertions in all species studied result in robust local enrichment of repressive epigenetic marks, the strength of such effects varies substantially within genomes, among species, and between species complexes. Our larger sample size allowed us to re-examine the still debated question about the impacts of TE-mediated enrichment of repressive marks on neighboring gene expression, which surprisingly revealed their complex interactions. Importantly, our multi-species analysis provides the power to test the predicted associations between TE abundance in the euchromatic genome, TE-mediated epigenetic effects, selection against such effects, and host chromatin environment, while uncovering the evolutionary causes for the wide variety of TE abundance between species.

Figure 1 with 8 supplements see all

Download asset Open asset

The replicative nature of TEs has made them successful at occupying nearly all eukaryotic genomes. Yet, their ‘success’, or genomic abundance, drastically varies across the phylogenetic tree (Huang et al., 2012; Elliott and Gregory, 2015; Wells and Feschotte, 2020) and between closely related species (Hu et al., 2011; Rius et al., 2016; Legrand et al., 2019), raising important questions regarding the evolutionary causes and functional consequences of varying genomic TE abundance. It is worth noting that most studies about the evolution of TE abundance, including ours, mainly concern insertions in the euchromatic genome. Although TEs are highly abundant in the heterochromatic genome, these TEs are typically fragmented (Hoskins et al., 2007; Hoskins et al., 2015) and would no longer be part of the ‘life cycle’ of TEs. The role of euchromatic TEs in determining the abundance and evolutionary dynamics of TEs in the heterochromatic genome is still a largely unaddressed question, mainly due to the challenges associated with assembling TEs in repeat-rich heterochromatic sequences (Salzberg and Yorke, 2005; Treangen and Salzberg, 2011, but see Khost et al., 2017; Chang and Larracuente, 2019; Chang et al., 2019).

Several hypotheses have been proposed to explain the large variation in euchromatic TE abundance. Phylogenetic signals may explain some of the differences between distantly related taxa (Wells and Feschotte, 2020) and, sometimes, between species within a taxa (Szitenberg et al., 2016). Variation in TE activities was also postulated to contribute to the wide variability of TE abundance (Chen et al., 2019; Wong et al., 2019, but see Ho et al., 2021). Still another plausible cause is systematic differences in the strength of natural selection removing TEs between species. Investigations of this hypothesis have been largely focused on population genetic parameters that influence the efficacy of natural selection, such as mating systems (Wright and Schoen, 1999; Dolgin et al., 2008; Boutin et al., 2012; Arunkumar et al., 2015; Agren et al., 2014) and effective population size (Lynch and Conery, 2003; Mérel et al., 2021). Here, our multi-species study tested step-by-step yet another evolutionary mechanism by which the strength of selection removing TEs could differ—through differences in the epigenetic effects of TEs that are driven by species-specific host chromatin landscape.

In this study, we investigated the prevalence, variability, and evolutionary importance of ‘the epigenetic effects of TEs’—TE-mediated local enrichment of repressive epigenetic marks—in the euchromatic genomes of six Drosophila species. These species come from two important and well-studied complexes within the melanogaster species subgroup (melanogaster and yakuba complexes), providing good phylogenetic resolution to decipher the evolutionary role of epigenetic mechanisms in shaping species-specific TE abundance. We observed that TE-mediated local enrichment of repressive marks is prevalent in all species studied and widely varies both within and between genomes. Interestingly, in addition to the intrinsic biological properties of TEs, our analyses revealed that host genetic background plays a critical role in determining the strength of TE-mediated epigenetic effects. These TE-mediated enrichments of repressive marks alter the epigenetic states of neighboring euchromatic sequences, including actively transcribing genes. Importantly, TEs exerting such epigenetic effects are selected against across multiple species, and the strength of this TE-mediated epigenetic effect negatively correlates with genomic TE abundance within species complex. Our findings extend previous studies based on few species and provide one of the first support for the importance of the inadvertent harmful effects of TE epigenetic silencing in shaping divergent genomic TE landscapes.

Curiously, we only found a negative association between genomic TE abundance and the magnitude of TE-mediated H3K9me2 enrichment across species, but not the extent of the spreading, even though these two indexes of a TE significantly correlate within genomes (Figure 1—figure supplement 2). Similarly, the evidence for selection against TE-mediated epigenetic effects is stronger for the magnitude of TE-induced H3K9me2 enrichment than for the extent (Figure 4). Previous empirical (Belyaeva and Zhimulev, 1991; Talbert and Henikoff, 2000) and theoretical (Erdel and Greene, 2016) studies have suggested that the spreading of repressive heterochromatic marks is a nonlinear process. Accordingly, the extent of the spreading of repressive marks from TEs could be sensitive to the genomic context and subject to substantial stochasticity. This echoes our observations that the transcription of genes influences the extent of TE-mediated H3K9me2 spreading, but not the magnitude of H3K9me2 enrichment (Figure 3D and Figure 3—figure supplement 10; also see below). The magnitude of TE-mediated H3K9me2 enrichment, which is estimated in windows right next to TE boundary, may thus be a more direct measurement of the strength of TE-mediated epigenetic effects and more indicative of the associated harmful impacts.

One of our most surprising findings is perhaps the limited evidence in supporting that TE-mediated epigenetic effects reduce neighboring gene expression on a genome-wide scale. While TE-mediated epigenetic effects increase the enrichment of repressive epigenetic marks at adjacent genes, we found limited associations between such effects and gene expression within genomes (Figure 3A and B). Comparing the expression of homologous alleles with and without TEs showing epigenetic effects also reached the same conclusion (Figure 3C). These observations echo previously reported lack of genome-wide associations between TE-induced enrichment of repressive epigenetic marks and the expression of neighboring genes in several model organisms (Quadrana et al., 2016; Stuart et al., 2016; Lee and Karpen, 2017; Choi and Purugganan, 2018) and are consistent with the wider observation that TEs upstream or downstream to genes are not predominantly associated with reduced gene expression (Goubert et al., 2020; Ullastres et al., 2021; Rech et al., 2022, reviewed in Kelleher et al., 2020; Choi and Lee, 2020). The limited impacts of TE-mediated H3K9me2 enrichment on gene expression could have resulted from the complex relationship between repressive epigenetic modification and gene expression (de Wit et al., 2007; Yasuhara and Wakimoto, 2008; Riddle et al., 2011; Meng et al., 2016; Caizzi et al., 2016), the varying sensitivity of genes to the enrichment of repressive marks (Rudolph et al., 2007; Vogel et al., 2009; Riddle et al., 2011), and the presence of other types of variants that also modulate gene expression (Stranger et al., 2007). Interestingly, our findings suggested another possibility—the transcription of genes, in return, influences TE-mediated epigenetic effects across genomes. Specifically, we found a more restricted spreading of repressive marks from TEs on the side facing a gene than on the intergenic side, and this difference is mainly observed for TEs near the promoters of highly expressed genes (Figure 3—figure supplement 12). It is worth noting that this difference in the extent of H3K9me2 spreading between the two sides of a TE is unlikely driven by the presence of insulator sequences (Gaszner and Felsenfeld, 2006) because of the limited associations between the presence of insulator sequences and the reduction in TE-mediated H3K9me2 spreading on the genic side (Figure 3—figure supplement 13). On the other hand, active histone modifications enriched at transcriptionally active genes could antagonize the assembly of heterochromatin (Allshire and Madhani, 2018), potentially restraining the spreading of repressive marks from silenced TEs. Consistently, in mice, active histone modification was reported to spread from an actively transcribing candidate gene into an adjacent TE (Rebollo et al., 2012). Similarly, active gene transcription at Drosophila miranda neo-Y chromosome was found to impede the formation of heterochromatin (Wei et al., 2020). Future analysis of the distributions of active histone modifications may help reveal the mechanistic cause for our observed genome-wide dependencies of the extent of TE-mediated epigenetic effects on the transcriptional activities of adjacent euchromatic genes.

If not due to reducing the expression of adjacent genes, what functional consequences of TE-mediated epigenetic effects could have impaired host fitness and led to the observed selection against these TEs? Euchromatic TEs enriched with repressive epigenetic marks were reported to spatially interact with constitutive heterochromatin through phase separation mechanisms, a process observed to alter 3D structures of genomes and inferred to lower host fitness (Lee et al., 2020). In addition, the epigenetic effects of TEs could shift the usual DNA repair process in the gene-rich euchromatic genome, perturbing the maintenance of local genome integrity. This is because double-stranded breaks happening in constitutive heterochromatin are repaired through a distinct cellular process from those in the euchromatic genome, owing to the enrichment of repressive epigenetic modifications (Chiolo et al., 2011; Janssen et al., 2016; Janssen et al., 2019). Still, the variance of gene expression was shown to be shaped by natural selection (Metzger et al., 2015; Duveau et al., 2018). Given the variegating properties of the spreading of repressive marks (reviewed in Elgin and Reuter, 2013), TE-mediated epigenetic effects could have shifted the variance, instead of the mean, of neighboring gene expression, impacting host fitness.

Building upon our findings of the prevalence and variability of TE-mediated epigenetic effects and selection against such effects, our multi-species analysis provides strong support for the previously proposed role of TE-mediated epigenetic effects in determining genomic TE abundance (Figure 6). This interpretation is based on the assumption that TE copy number is at equilibrium and transposition rates of TEs are similar across species, leaving the strength of selection removing TEs being the major determinant of genomic TE abundance (Charlesworth and Charlesworth, 1983). By combining transcriptomic analysis and Drosophila genetics experiment, we further revealed that between-species differences in TE-mediated epigenetic effects could be driven by species-specific expression levels of host genetic factors that modulate heterochromatin. These findings connect the evolution of host genome (genomic TE abundance) with chromatin landscape through the inadvertent harmful effects of the epigenetic silencing of TEs (Figure 6). Curiously, the negative association between TE-mediated H3K9me2 enrichment and genomic TE abundance was only observed within species complex, pointing that the importance TE-mediated epigenetic effects in determining genomic TE abundance may be on a short evolutionary time scale. Specifically, even though we found stronger epigenetic effects of TEs in species of the yakuba complex than those in the melanogaster complex, TEs are more abundant in the former. This accumulation of TEs in the genomes of yakuba complex is in line with previously observed greater net gains of intron sequences (Presgraves, 2006) and duplicated genes (Hahn et al., 2007) in D. yakuba than those in the lineage leading to melanogaster complex, which could suggest a higher tolerance of D. yakuba genome with gained sequences. Such lineage-specific mechanism that shapes the gain and loss of sequences over the entire genome may also play an important role in determining genomic TE abundance and potentially override the influence of selection against varying TE-mediated epigenetic effects on a longer evolutionary time scale.

Figure 6

Download asset Open asset

It is worth noting that teasing apart the causality of the observed associations between the epigenetic effects and genomic abundance of TEs is challenging. A reduction in effective population size and accordingly weakened effectiveness of natural selection at removing TEs could drive the accumulation of TEs in some of the species (e.g., Mérel et al., 2021). Under this scenario, selection may favor epigenetic regulation that limits excessive inadvertent spreading of repressive marks from euchromatic TEs, which could also result in our observed weaker TE-mediated epigenetic effects in species with abundant TEs. Moreover, even if the difference in genomic TE abundance is indeed the ‘consequence’, varying efficacy of selection in purging deleterious TE insertions could also contribute to the observed between-species difference in TE abundance. Specifically, within species complex, estimated nucleotide polymorphism, an indicator for effective population size (Charlesworth, 2009), largely follows a similar rank order to our observed magnitude of TE-mediated H3K9me2 enrichment (nucleotide polymorphism, melanogaster complex: D. simulans > D. mauritiana > D. melanogaster (Langley et al., 2012; Meiklejohn et al., 2018) and yakuba complex: D. tessieri > D. yakuba > D. santomea (Bachtrog et al., 2006); the only difference in the rank order being the epigenetic effect of TEs is stronger in D. mauritiana than in D. simulans, Figure 5A). More effective selection at removing TEs in species with larger effective population size (but also stronger epigenetic effects of TEs) is another plausible driver for varying TE abundance that our analysis could not rule out.

Our study focuses on late-stage embryos and would capture the averaged epigenetic effects of TEs across heterogeneous cell types at this developmental stage. If TE-mediated epigenetic effects vary minimally across tissues or cell types, our estimate serves as a good proxy to study the associated fitness consequence. Yet, if TE-mediated enrichment of repressive marks substantially varies between tissues and cell types that have different importance in shaping individual functions, our estimate could fail to capture the most relevant variation in TE-mediated epigenetic effects that determine individual fitness. Even though we did find evidence supporting selection against TE-mediated enrichment of repressive marks that was estimated from whole embryos, future investigation on the variability of such effects across tissues and cell types will be important for the finer dissection of the role of TE-mediated epigenetic effects in individual fitness and, thus, genomic TE abundance.

Proper maintenance of heterochromatin ensures genome function and integrity (reviewed in Janssen et al., 2018). Intriguingly, the expression level (see above), copy number (Levine et al., 2012; Ross et al., 2013; Helleu and Levine, 2018), and amino acid sequences (Levine et al., 2012; Sasaki et al., 2019) of heterochromatin structural and enzymatic components have been observed to vary substantially within and between species. The abundance and composition of heterochromatic repeats also rapidly turnover between genomes (Larracuente, 2014; Wei et al., 2014; Wei et al., 2018). While variation in these genetic factors was postulated to contribute to the evolution of heterochromatin functions (e.g., Ferree and Barbash, 2009; Helleu and Levine, 2018; Mills et al., 2019), our discoveries further extend their evolutionary impacts from the gene-poor genomic dark matter into the gene-rich euchromatin. Heterochromatin modulators may not only determine the local epigenetic impacts of euchromatic TEs on flanking sequences, but also shape TE abundance and, accordingly, the structure and function of the euchromatic genomes. The evolution of euchromatin and heterochromatin, two genomic compartments usually presumed to function independently, may be more interconnected than previously thought by the inadvertent deleterious epigenetic effects of a widespread genetic parasite.

Short-read data for ChIP-seq, RNA-seq, and Illumina whole-genome sequencing have been deposited to NCBI BioProject under accession number PRJNA855483. PacBio long-read data and genome assembly of D. melanogaster ORw1118 strain have been deposited to NCBI BioProject under accession number PRJNA855235.

1. 1. Lee YCG

   (2022) NCBI BioProject

   ID PRJNA855235. Evolutionary causes and consequences of transposable elements' epigenetic effects.

   https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA855235
2. 1. Lee YCG

   (2022) NCBI BioProject

   ID PRJNA855483. Evolutionary causes and consequences of transponsable elements' epigenetic effects.

   https://www.ncbi.nlm.nih.gov/bioproject/PRJNA855483

1. 1. Lack JB

   (2015) NCBI

   ID SRP006733. "1000" Drosophila Genomes - DPGP3.

   https://trace.ncbi.nlm.nih.gov/Traces/index.html?view=study&acc=SRP006733
2. 1. Rogers RL

   (2014) NCBI

   ID SRP040290. Drosophila simulans strain:multiple Genome sequencing.

   https://trace.ncbi.nlm.nih.gov/Traces/index.html?view=study&acc=SRP040290
3. 1. Rogers RL

   (2014) NCBI

   ID SRP029453. Drosophila yakuba strain:multiple Genome sequencing.

   https://trace.ncbi.nlm.nih.gov/Traces/index.html?view=study&acc=SRP029453
4. 1. Garrigan D
   2. Kingan SB
   3. Geneva AJ
   4. Vedanayagam JP
   5. Presgraves DC

   (2014) NCBI

   ID SRP012053. Drosophila mauritiana genome sequencing.

   https://trace.ncbi.nlm.nih.gov/Traces/index.html?view=study&acc=SRP012053

1. 1. Agren JÅ
   2. Wang W
   3. Koenig D
   4. Neuffer B
   5. Weigel D
   6. Wright SI

   (2014) Mating system shifts and transposable element evolution in the plant genus Capsella

   BMC Genomics 15:602.

   https://doi.org/10.1186/1471-2164-15-602

   • PubMed
   • Google Scholar
2. 1. Allis CD
   2. Jenuwein T

   (2016) The molecular hallmarks of epigenetic control

   Nature Reviews. Genetics 17:487–500.

   https://doi.org/10.1038/nrg.2016.59

   • PubMed
   • Google Scholar
3. 1. Allshire RC
   2. Madhani HD

   (2018) Ten principles of heterochromatin formation and function

   Nature Reviews. Molecular Cell Biology 19:229–244.

   https://doi.org/10.1038/nrm.2017.119

   • PubMed
   • Google Scholar
4. 1. Aravin AA
   2. Sachidanandam R
   3. Bourc’his D
   4. Schaefer C
   5. Pezic D
   6. Toth KF
   7. Bestor T
   8. Hannon GJ

   (2008) A piRNA pathway primed by individual transposons is linked to de novo DNA methylation in mice

   Molecular Cell 31:785–799.

   https://doi.org/10.1016/j.molcel.2008.09.003

   • PubMed
   • Google Scholar
5. 1. Arkhipova IR
   2. Kumar S

   (2018) Neutral theory, transposable elements, and eukaryotic genome evolution

   Molecular Biology and Evolution 35:1332–1337.

   https://doi.org/10.1093/molbev/msy083

   • PubMed
   • Google Scholar
6. 1. Arunkumar R
   2. Ness RW
   3. Wright SI
   4. Barrett SCH

   (2015) The evolution of selfing is accompanied by reduced efficacy of selection and purging of deleterious mutations

   Genetics 199:817–829.

   https://doi.org/10.1534/genetics.114.172809

   • PubMed
   • Google Scholar
7. Software

   1. Babraham Bioinformatics

   (2022) Trim Galore, version GPL v3

   Babraham Bioinformatics.

   https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/
8. 1. Bachtrog D
   2. Thornton K
   3. Clark A
   4. Andolfatto P

   (2006) Extensive introgression of mitochondrial DNA relative to nuclear genes in the Drosophila yakuba species group

   Evolution; International Journal of Organic Evolution 60:292–302.

   https://doi.org/10.1111/j.0014-3820.2006.tb01107.x

   • PubMed
   • Google Scholar
9. 1. Barrón MG
   2. Fiston-Lavier AS
   3. Petrov DA
   4. González J

   (2014) Population genomics of transposable elements in Drosophila

   Annual Review of Genetics 48:561–581.

   https://doi.org/10.1146/annurev-genet-120213-092359

   • PubMed
   • Google Scholar
10. 1. Belyaeva ES
    2. Zhimulev IF

    (1991) Cytogenetic and molecular aspects of position effect variegation in Drosophila

    Chromosoma 100:453–466.

    https://doi.org/10.1007/BF00364556

    • PubMed
    • Google Scholar
11. 1. Bergman CM
    2. Quesneville H
    3. Anxolabéhère D
    4. Ashburner M

    (2006) Recurrent insertion and duplication generate networks of transposable element sequences in the Drosophila melanogaster genome

    Genome Biology 7:R112.

    https://doi.org/10.1186/gb-2006-7-11-r112

    • PubMed
    • Google Scholar
12. 1. Bolger AM
    2. Lohse M
    3. Usadel B

    (2014) Trimmomatic: a flexible trimmer for Illumina sequence data

    Bioinformatics 30:2114–2120.

    https://doi.org/10.1093/bioinformatics/btu170

    • PubMed
    • Google Scholar
13. 1. Boutin TS
    2. Le Rouzic A
    3. Capy P

    (2012) How does selfing affect the dynamics of selfish transposable elements?

    Mobile DNA 3:5.

    https://doi.org/10.1186/1759-8753-3-5

    • PubMed
    • Google Scholar
14. 1. Caizzi R
    2. Moschetti R
    3. Piacentini L
    4. Fanti L
    5. Marsano RM
    6. Dimitri P

    (2016) Comparative genomic analyses provide new insights into the evolutionary dynamics of heterochromatin in Drosophila

    PLOS Genetics 12:e1006212.

    https://doi.org/10.1371/journal.pgen.1006212

    • PubMed
    • Google Scholar
15. 1. Camacho C
    2. Coulouris G
    3. Avagyan V
    4. Ma N
    5. Papadopoulos J
    6. Bealer K
    7. Madden TL

    (2009) BLAST+: architecture and applications

    BMC Bioinformatics 10:421.

    https://doi.org/10.1186/1471-2105-10-421

    • PubMed
    • Google Scholar
16. 1. Chakraborty M
    2. Emerson JJ
    3. Macdonald SJ
    4. Long AD

    (2019) Structural variants exhibit widespread allelic heterogeneity and shape variation in complex traits

    Nature Communications 10:1–11.

    https://doi.org/10.1038/s41467-019-12884-1

    • PubMed
    • Google Scholar
17. 1. Chakraborty M
    2. Chang CH
    3. Khost DE
    4. Vedanayagam J
    5. Adrion JR
    6. Liao Y
    7. Montooth KL
    8. Meiklejohn CD
    9. Larracuente AM
    10. Emerson JJ

    (2021) Evolution of genome structure in the Drosophila simulans species complex

    Genome Research 31:380–396.

    https://doi.org/10.1101/gr.263442.120

    • PubMed
    • Google Scholar
18. 1. Chang CH
    2. Chavan A
    3. Palladino J
    4. Wei X
    5. Martins NMC
    6. Santinello B
    7. Chen CC
    8. Erceg J
    9. Beliveau BJ
    10. Wu CT
    11. Larracuente AM
    12. Mellone BG

    (2019) Islands of retroelements are major components of Drosophila centromeres

    PLOS Biology 17:e3000241.

    https://doi.org/10.1371/journal.pbio.3000241

    • PubMed
    • Google Scholar
19. 1. Chang CH
    2. Larracuente AM

    (2019) Heterochromatin-enriched assemblies reveal the sequence and organization of the Drosophila melanogaster Y chromosome

    Genetics 211:333–348.

    https://doi.org/10.1534/genetics.118.301765

    • PubMed
    • Google Scholar
20. 1. Charlesworth B
    2. Charlesworth D

    (1983) The population dynamics of transposable elements

    Genetical Research 42:1–27.

    https://doi.org/10.1017/S0016672300021455

    • Google Scholar
21. 1. Charlesworth B
    2. Langley CH

    (1989) The population genetics of Drosophila transposable elements

    Annual Review of Genetics 23:251–287.

    https://doi.org/10.1146/annurev.ge.23.120189.001343

    • PubMed
    • Google Scholar
22. 1. Charlesworth B

    (2009) Fundamental concepts in genetics: effective population size and patterns of molecular evolution and variation

    Nature Reviews. Genetics 10:195–205.

    https://doi.org/10.1038/nrg2526

    • PubMed
    • Google Scholar
23. 1. Chen J
    2. Lu L
    3. Benjamin J
    4. Diaz S
    5. Hancock CN
    6. Stajich JE
    7. Wessler SR

    (2019) Tracking the origin of two genetic components associated with transposable element bursts in domesticated rice

    Nature Communications 10:641.

    https://doi.org/10.1038/s41467-019-08451-3

    • PubMed
    • Google Scholar
24. 1. Chiang C
    2. Layer RM
    3. Faust GG
    4. Lindberg MR
    5. Rose DB
    6. Garrison EP
    7. Marth GT
    8. Quinlan AR
    9. Hall IM

    (2015) SpeedSeq: ultra-fast personal genome analysis and interpretation

    Nature Methods 12:966–968.

    https://doi.org/10.1038/nmeth.3505

    • PubMed
    • Google Scholar
25. 1. Chiolo I
    2. Minoda A
    3. Colmenares SU
    4. Polyzos A
    5. Costes SV
    6. Karpen GH

    (2011) Double-strand breaks in heterochromatin move outside of a dynamic HP1a domain to complete recombinational repair

    Cell 144:732–744.

    https://doi.org/10.1016/j.cell.2011.02.012

    • PubMed
    • Google Scholar
26. 1. Choi JY
    2. Purugganan MD

    (2018) Evolutionary epigenomics of retrotransposon-mediated methylation spreading in rice

    Molecular Biology and Evolution 35:365–382.

    https://doi.org/10.1093/molbev/msx284

    • PubMed
    • Google Scholar
27. 1. Choi JY
    2. Lee YCG

    (2020) Double-edged sword: the evolutionary consequences of the epigenetic silencing of transposable elements

    PLOS Genetics 16:e1008872.

    https://doi.org/10.1371/journal.pgen.1008872

    • PubMed
    • Google Scholar
28. 1. Chuong EB
    2. Elde NC
    3. Feschotte C

    (2017) Regulatory activities of transposable elements: from conflicts to benefits

    Nature Reviews. Genetics 18:71–86.

    https://doi.org/10.1038/nrg.2016.139

    • PubMed
    • Google Scholar
29. 1. Clark AG
    2. Eisen MB
    3. Smith DR
    4. Bergman CM
    5. Oliver B
    6. Markow TA
    7. Kaufman TC
    8. Kellis M
    9. Gelbart W
    10. Iyer VN
    11. Pollard DA
    12. Sackton TB
    13. Larracuente AM
    14. Singh ND
    15. Abad JP
    16. Abt DN
    17. Adryan B
    18. Aguade M
    19. Akashi H
    20. Anderson WW
    21. Aquadro CF
    22. Ardell DH
    23. Arguello R
    24. Artieri CG
    25. Barbash DA
    26. Barker D
    27. Barsanti P
    28. Batterham P
    29. Batzoglou S
    30. Begun D
    31. Bhutkar A
    32. Blanco E
    33. Bosak SA
    34. Bradley RK
    35. Brand AD
    36. Brent MR
    37. Brooks AN
    38. Brown RH
    39. Butlin RK
    40. Caggese C
    41. Calvi BR
    42. Bernardo de Carvalho A
    43. Caspi A
    44. Castrezana S
    45. Celniker SE
    46. Chang JL
    47. Chapple C
    48. Chatterji S
    49. Chinwalla A
    50. Civetta A
    51. Clifton SW
    52. Comeron JM
    53. Costello JC
    54. Coyne JA
    55. Daub J
    56. David RG
    57. Delcher AL
    58. Delehaunty K
    59. Do CB
    60. Ebling H
    61. Edwards K
    62. Eickbush T
    63. Evans JD
    64. Filipski A
    65. Findeiss S
    66. Freyhult E
    67. Fulton L
    68. Fulton R
    69. Garcia ACL
    70. Gardiner A
    71. Garfield DA
    72. Garvin BE
    73. Gibson G
    74. Gilbert D
    75. Gnerre S
    76. Godfrey J
    77. Good R
    78. Gotea V
    79. Gravely B
    80. Greenberg AJ
    81. Griffiths-Jones S
    82. Gross S
    83. Guigo R
    84. Gustafson EA
    85. Haerty W
    86. Hahn MW
    87. Halligan DL
    88. Halpern AL
    89. Halter GM
    90. Han MV
    91. Heger A
    92. Hillier L
    93. Hinrichs AS
    94. Holmes I
    95. Hoskins RA
    96. Hubisz MJ
    97. Hultmark D
    98. Huntley MA
    99. Jaffe DB
    100. Jagadeeshan S
    101. Jeck WR
    102. Johnson J
    103. Jones CD
    104. Jordan WC
    105. Karpen GH
    106. Kataoka E
    107. Keightley PD
    108. Kheradpour P
    109. Kirkness EF
    110. Koerich LB
    111. Kristiansen K
    112. Kudrna D
    113. Kulathinal RJ
    114. Kumar S
    115. Kwok R
    116. Lander E
    117. Langley CH
    118. Lapoint R
    119. Lazzaro BP
    120. Lee S-J
    121. Levesque L
    122. Li R
    123. Lin C-F
    124. Lin MF
    125. Lindblad-Toh K
    126. Llopart A
    127. Long M
    128. Low L
    129. Lozovsky E
    130. Lu J
    131. Luo M
    132. Machado CA
    133. Makalowski W
    134. Marzo M
    135. Matsuda M
    136. Matzkin L
    137. McAllister B
    138. McBride CS
    139. McKernan B
    140. McKernan K
    141. Mendez-Lago M
    142. Minx P
    143. Mollenhauer MU
    144. Montooth K
    145. Mount SM
    146. Mu X
    147. Myers E
    148. Negre B
    149. Newfeld S
    150. Nielsen R
    151. Noor MAF
    152. O’Grady P
    153. Pachter L
    154. Papaceit M
    155. Parisi MJ
    156. Parisi M
    157. Parts L
    158. Pedersen JS
    159. Pesole G
    160. Phillippy AM
    161. Ponting CP
    162. Pop M
    163. Porcelli D
    164. Powell JR
    165. Prohaska S
    166. Pruitt K
    167. Puig M
    168. Quesneville H
    169. Ram KR
    170. Rand D
    171. Rasmussen MD
    172. Reed LK
    173. Reenan R
    174. Reily A
    175. Remington KA
    176. Rieger TT
    177. Ritchie MG
    178. Robin C
    179. Rogers Y-H
    180. Rohde C
    181. Rozas J
    182. Rubenfield MJ
    183. Ruiz A
    184. Russo S
    185. Salzberg SL
    186. Sanchez-Gracia A
    187. Saranga DJ
    188. Sato H
    189. Schaeffer SW
    190. Schatz MC
    191. Schlenke T
    192. Schwartz R
    193. Segarra C
    194. Singh RS
    195. Sirot L
    196. Sirota M
    197. Sisneros NB
    198. Smith CD
    199. Smith TF
    200. Spieth J
    201. Stage DE
    202. Stark A
    203. Stephan W
    204. Strausberg RL
    205. Strempel S
    206. Sturgill D
    207. Sutton G
    208. Sutton GG
    209. Tao W
    210. Teichmann S
    211. Tobari YN
    212. Tomimura Y
    213. Tsolas JM
    214. Valente VLS
    215. Venter E
    216. Venter JC
    217. Vicario S
    218. Vieira FG
    219. Vilella AJ
    220. Villasante A
    221. Walenz B
    222. Wang J
    223. Wasserman M
    224. Watts T
    225. Wilson D
    226. Wilson RK
    227. Wing RA
    228. Wolfner MF
    229. Wong A
    230. Wong GK-S
    231. Wu C-I
    232. Wu G
    233. Yamamoto D
    234. Yang H-P
    235. Yang S-P
    236. Yorke JA
    237. Yoshida K
    238. Zdobnov E
    239. Zhang P
    240. Zhang Y
    241. Zimin AV
    242. Baldwin J
    243. Abdouelleil A
    244. Abdulkadir J
    245. Abebe A
    246. Abera B
    247. Abreu J
    248. Acer SC
    249. Aftuck L
    250. Alexander A
    251. An P
    252. Anderson E
    253. Anderson S
    254. Arachi H
    255. Azer M
    256. Bachantsang P
    257. Barry A
    258. Bayul T
    259. Berlin A
    260. Bessette D
    261. Bloom T
    262. Blye J
    263. Boguslavskiy L
    264. Bonnet C
    265. Boukhgalter B
    266. Bourzgui I
    267. Brown A
    268. Cahill P
    269. Channer S
    270. Cheshatsang Y
    271. Chuda L
    272. Citroen M
    273. Collymore A
    274. Cooke P
    275. Costello M
    276. D’Aco K
    277. Daza R
    278. De Haan G
    279. DeGray S
    280. DeMaso C
    281. Dhargay N
    282. Dooley K
    283. Dooley E
    284. Doricent M
    285. Dorje P
    286. Dorjee K
    287. Dupes A
    288. Elong R
    289. Falk J
    290. Farina A
    291. Faro S
    292. Ferguson D
    293. Fisher S
    294. Foley CD
    295. Franke A
    296. Friedrich D
    297. Gadbois L
    298. Gearin G
    299. Gearin CR
    300. Giannoukos G
    301. Goode T
    302. Graham J
    303. Grandbois E
    304. Grewal S
    305. Gyaltsen K
    306. Hafez N
    307. Hagos B
    308. Hall J
    309. Henson C
    310. Hollinger A
    311. Honan T
    312. Huard MD
    313. Hughes L
    314. Hurhula B
    315. Husby ME
    316. Kamat A
    317. Kanga B
    318. Kashin S
    319. Khazanovich D
    320. Kisner P
    321. Lance K
    322. Lara M
    323. Lee W
    324. Lennon N
    325. Letendre F
    326. LeVine R
    327. Lipovsky A
    328. Liu X
    329. Liu J
    330. Liu S
    331. Lokyitsang T
    332. Lokyitsang Y
    333. Lubonja R
    334. Lui A
    335. MacDonald P
    336. Magnisalis V
    337. Maru K
    338. Matthews C
    339. McCusker W
    340. McDonough S
    341. Mehta T
    342. Meldrim J
    343. Meneus L
    344. Mihai O
    345. Mihalev A
    346. Mihova T
    347. Mittelman R
    348. Mlenga V
    349. Montmayeur A
    350. Mulrain L
    351. Navidi A
    352. Naylor J
    353. Negash T
    354. Nguyen T
    355. Nguyen N
    356. Nicol R
    357. Norbu C
    358. Norbu N
    359. Novod N
    360. O’Neill B
    361. Osman S
    362. Markiewicz E
    363. Oyono OL
    364. Patti C
    365. Phunkhang P
    366. Pierre F
    367. Priest M
    368. Raghuraman S
    369. Rege F
    370. Reyes R
    371. Rise C
    372. Rogov P
    373. Ross K
    374. Ryan E
    375. Settipalli S
    376. Shea T
    377. Sherpa N
    378. Shi L
    379. Shih D
    380. Sparrow T
    381. Spaulding J
    382. Stalker J
    383. Stange-Thomann N
    384. Stavropoulos S
    385. Stone C
    386. Strader C
    387. Tesfaye S
    388. Thomson T
    389. Thoulutsang Y
    390. Thoulutsang D
    391. Topham K
    392. Topping I
    393. Tsamla T
    394. Vassiliev H
    395. Vo A
    396. Wangchuk T
    397. Wangdi T
    398. Weiand M
    399. Wilkinson J
    400. Wilson A
    401. Yadav S
    402. Young G
    403. Yu Q
    404. Zembek L
    405. Zhong D
    406. Zimmer A
    407. Zwirko Z
    408. Jaffe DB
    409. Alvarez P
    410. Brockman W
    411. Butler J
    412. Chin C
    413. Gnerre S
    414. Grabherr M
    415. Kleber M
    416. Mauceli E
    417. MacCallum I
    418. Drosophila 12 Genomes Consortium

    (2007) Evolution of genes and genomes on the Drosophila phylogeny

    Nature 450:203–218.

    https://doi.org/10.1038/nature06341

    • PubMed
    • Google Scholar
30. 1. Comeron JM
    2. Ratnappan R
    3. Bailin S

    (2012) The many landscapes of recombination in Drosophila melanogaster

    PLOS Genetics 8:e1002905.

    https://doi.org/10.1371/journal.pgen.1002905

    • PubMed
    • Google Scholar
31. 1. Cridland JM
    2. Macdonald SJ
    3. Long AD
    4. Thornton KR

    (2013) Abundance and distribution of transposable elements in two Drosophila QTL mapping resources

    Molecular Biology and Evolution 30:2311–2327.

    https://doi.org/10.1093/molbev/mst129

    • PubMed
    • Google Scholar
32. 1. Czech B
    2. Munafò M
    3. Ciabrelli F
    4. Eastwood EL
    5. Fabry MH
    6. Kneuss E
    7. Hannon GJ

    (2018) piRNA-guided genome defense: from biogenesis to silencing

    Annual Review of Genetics 52:131–157.

    https://doi.org/10.1146/annurev-genet-120417-031441

    • PubMed
    • Google Scholar
33. 1. de la Chaux N
    2. Tsuchimatsu T
    3. Shimizu KK
    4. Wagner A

    (2012) The predominantly selfing plant Arabidopsis thaliana experienced a recent reduction in transposable element abundance compared to its outcrossing relative Arabidopsis lyrata

    Mobile DNA 3:2.

    https://doi.org/10.1186/1759-8753-3-2

    • PubMed
    • Google Scholar
34. 1. de Wit E
    2. Greil F
    3. van Steensel B

    (2007) High-resolution mapping reveals links of HP1 with active and inactive chromatin components

    PLOS Genetics 3:e38.

    https://doi.org/10.1371/journal.pgen.0030038

    • PubMed
    • Google Scholar
35. 1. Deniz Ö
    2. Frost JM
    3. Branco MR

    (2019) Regulation of transposable elements by DNA modifications

    Nature Reviews. Genetics 20:417–431.

    https://doi.org/10.1038/s41576-019-0106-6

    • PubMed
    • Google Scholar
36. 1. Deshpande G
    2. Calhoun G
    3. Schedl P

    (2005) Drosophila argonaute-2 is required early in embryogenesis for the assembly of centric/centromeric heterochromatin, nuclear division, nuclear migration, and germ-cell formation

    Genes & Development 19:1680–1685.

    https://doi.org/10.1101/gad.1316805

    • PubMed
    • Google Scholar
37. 1. Dimitri P
    2. Junakovic N

    (1999) Revising the selfish DNA hypothesis: new evidence on accumulation of transposable elements in heterochromatin

    Trends in Genetics 15:123–124.

    https://doi.org/10.1016/s0168-9525(99)01711-4

    • PubMed
    • Google Scholar
38. 1. Dobin A
    2. Davis CA
    3. Schlesinger F
    4. Drenkow J
    5. Zaleski C
    6. Jha S
    7. Batut P
    8. Chaisson M
    9. Gingeras TR

    (2013) STAR: ultrafast universal RNA-seq aligner

    Bioinformatics 29:15–21.

    https://doi.org/10.1093/bioinformatics/bts635

    • PubMed
    • Google Scholar
39. 1. Dolgin ES
    2. Charlesworth B

    (2006) The fate of transposable elements in asexual populations

    Genetics 174:817–827.

    https://doi.org/10.1534/genetics.106.060434

    • PubMed
    • Google Scholar
40. 1. Dolgin ES
    2. Charlesworth B
    3. Cutter AD

    (2008) Population frequencies of transposable elements in selfing and outcrossing Caenorhabditis nematodes

    Genetics Research 90:317–329.

    https://doi.org/10.1017/S0016672308009440

    • PubMed
    • Google Scholar
41. 1. Duveau F
    2. Hodgins-Davis A
    3. Metzger BP
    4. Yang B
    5. Tryban S
    6. Walker EA
    7. Lybrook T
    8. Wittkopp PJ

    (2018) Fitness effects of altering gene expression noise in Saccharomyces cerevisiae

    eLife 7:e37272.

    https://doi.org/10.7554/eLife.37272

    • PubMed
    • Google Scholar
42. 1. Egelhofer TA
    2. Minoda A
    3. Klugman S
    4. Lee K
    5. Kolasinska-Zwierz P
    6. Alekseyenko AA
    7. Cheung MS
    8. Day DS
    9. Gadel S
    10. Gorchakov AA
    11. Gu T
    12. Kharchenko PV
    13. Kuan S
    14. Latorre I
    15. Linder-Basso D
    16. Luu Y
    17. Ngo Q
    18. Perry M
    19. Rechtsteiner A
    20. Riddle NC
    21. Schwartz YB
    22. Shanower GA
    23. Vielle A
    24. Ahringer J
    25. Elgin SCR
    26. Kuroda MI
    27. Pirrotta V
    28. Ren B
    29. Strome S
    30. Park PJ
    31. Karpen GH
    32. Hawkins RD
    33. Lieb JD

    (2011) An assessment of histone-modification antibody quality

    Nature Structural & Molecular Biology 18:91–93.

    https://doi.org/10.1038/nsmb.1972

    • PubMed
    • Google Scholar
43. 1. Elgin SCR
    2. Reuter G

    (2013) Position-effect variegation, heterochromatin formation, and gene silencing in Drosophila

    Cold Spring Harbor Perspectives in Biology 5:a017780.

    https://doi.org/10.1101/cshperspect.a017780

    • PubMed
    • Google Scholar
44. 1. Elliott TA
    2. Gregory TR

    (2015) What’s in a genome? the C-value enigma and the evolution of eukaryotic genome content

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 370:20140331.

    https://doi.org/10.1098/rstb.2014.0331

    • PubMed
    • Google Scholar
45. 1. Erdel F
    2. Greene EC

    (2016) Generalized nucleation and looping model for epigenetic memory of histone modifications

    PNAS 113:E4180–E4189.

    https://doi.org/10.1073/pnas.1605862113

    • PubMed
    • Google Scholar
46. 1. Ewing B
    2. Green P

    (1998)

    Base-calling of automated sequencer traces using phred. II. Error probabilities

    Genome Research 8:186–194.

    • PubMed
    • Google Scholar
47. 1. Felsenstein J

    (1974) The evolutionary advantage of recombination

    Genetics 78:737–756.

    https://doi.org/10.1093/genetics/78.2.737

    • PubMed
    • Google Scholar
48. 1. Ferree PM
    2. Barbash DA

    (2009) Species-specific heterochromatin prevents mitotic chromosome segregation to cause hybrid lethality in Drosophila

    PLOS Biology 7:e1000234.

    https://doi.org/10.1371/journal.pbio.1000234

    • PubMed
    • Google Scholar
49. 1. Feschotte C
    2. Pritham EJ

    (2007) DNA transposons and the evolution of eukaryotic genomes

    Annual Review of Genetics 41:331–368.

    https://doi.org/10.1146/annurev.genet.40.110405.090448

    • PubMed
    • Google Scholar
50. 1. Filion GJ
    2. van Bemmel JG
    3. Braunschweig U
    4. Talhout W
    5. Kind J
    6. Ward LD
    7. Brugman W
    8. de Castro IJ
    9. Kerkhoven RM
    10. Bussemaker HJ
    11. van Steensel B

    (2010) Systematic protein location mapping reveals five principal chromatin types in Drosophila cells

    Cell 143:212–224.

    https://doi.org/10.1016/j.cell.2010.09.009

    • PubMed
    • Google Scholar
51. 1. Finnegan DJ

    (1992) Transposable elements

    Current Opinion in Genetics & Development 2:861–867.

    https://doi.org/10.1016/s0959-437x(05)80108-x

    • PubMed
    • Google Scholar
52. 1. Flynn JM
    2. Hubley R
    3. Goubert C
    4. Rosen J
    5. Clark AG
    6. Feschotte C
    7. Smit AF

    (2020) RepeatModeler2 for automated genomic discovery of transposable element families

    PNAS 117:9451–9457.

    https://doi.org/10.1073/pnas.1921046117

    • PubMed
    • Google Scholar
53. 1. Frolov MV
    2. Birchler JA

    (1998) Mutation in P0, a dual function ribosomal protein/apurinic/apyrimidinic endonuclease, modifies gene expression and position effect variegation in Drosophila

    Genetics 150:1487–1495.

    https://doi.org/10.1093/genetics/150.4.1487

    • PubMed
    • Google Scholar
54. 1. Garrigan D
    2. Kingan SB
    3. Geneva AJ
    4. Andolfatto P
    5. Clark AG
    6. Thornton KR
    7. Presgraves DC

    (2012) Genome sequencing reveals complex speciation in the Drosophila simulans clade

    Genome Research 22:1499–1511.

    https://doi.org/10.1101/gr.130922.111

    • PubMed
    • Google Scholar
55. 1. Garrigan D
    2. Kingan SB
    3. Geneva AJ
    4. Vedanayagam JP
    5. Presgraves DC

    (2014) Genome diversity and divergence in Drosophila mauritiana: multiple signatures of faster X evolution

    Genome Biology and Evolution 6:2444–2458.

    https://doi.org/10.1093/gbe/evu198

    • PubMed
    • Google Scholar
56. 1. Gaszner M
    2. Felsenfeld G

    (2006) Insulators: exploiting transcriptional and epigenetic mechanisms

    Nature Reviews. Genetics 7:703–713.

    https://doi.org/10.1038/nrg1925

    • PubMed
    • Google Scholar
57. 1. Girton JR
    2. Johansen KM

    (2008) Chromatin structure and the regulation of gene expression: the lessons of PEV in Drosophila

    Advances in Genetics 61:1–43.

    https://doi.org/10.1016/S0065-2660(07)00001-6

    • PubMed
    • Google Scholar
58. 1. Goubert C
    2. Zevallos NA
    3. Feschotte C

    (2020) Contribution of unfixed transposable element insertions to human regulatory variation

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 375:20190331.

    https://doi.org/10.1098/rstb.2019.0331

    • PubMed
    • Google Scholar
59. 1. Gowen JW
    2. Gay EH

    (1934) Chromosome constitution and behavior in eversporting and mottling in Drosophila melanogaster

    Genetics 19:189–208.

    https://doi.org/10.1093/genetics/19.3.189

    • PubMed
    • Google Scholar
60. 1. Grabherr MG
    2. Haas BJ
    3. Yassour M
    4. Levin JZ
    5. Thompson DA
    6. Amit I
    7. Adiconis X
    8. Fan L
    9. Raychowdhury R
    10. Zeng Q
    11. Chen Z
    12. Mauceli E
    13. Hacohen N
    14. Gnirke A
    15. Rhind N
    16. di Palma F
    17. Birren BW
    18. Nusbaum C
    19. Lindblad-Toh K
    20. Friedman N
    21. Regev A

    (2011) Full-length transcriptome assembly from RNA-Seq data without a reference genome

    Nature Biotechnology 29:644–652.

    https://doi.org/10.1038/nbt.1883

    • PubMed
    • Google Scholar
61. 1. Grafen A

    (1989) The phylogenetic regression

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 326:119–157.

    https://doi.org/10.1098/rstb.1989.0106

    • PubMed
    • Google Scholar
62. 1. Greil F
    2. de Wit E
    3. Bussemaker HJ
    4. van Steensel B

    (2007) HP1 controls genomic targeting of four novel heterochromatin proteins in Drosophila

    The EMBO Journal 26:741–751.

    https://doi.org/10.1038/sj.emboj.7601527

    • PubMed
    • Google Scholar
63. 1. Hahn MW
    2. Han MV
    3. Han SG

    (2007) Gene family evolution across 12 Drosophila genomes

    PLOS Genetics 3:e197.

    https://doi.org/10.1371/journal.pgen.0030197

    • PubMed
    • Google Scholar
64. 1. Helleu Q
    2. Levine MT

    (2018) Recurrent amplification of the heterochromatin protein 1 (HP1) gene family across diptera

    Molecular Biology and Evolution 35:2375–2389.

    https://doi.org/10.1093/molbev/msy128

    • PubMed
    • Google Scholar
65. 1. Hill WG
    2. Robertson A

    (1966)

    The effect of linkage on limits to artificial selection

    Genetical Research 8:269–294.

    • PubMed
    • Google Scholar
66. 1. Ho EKH
    2. Bellis ES
    3. Calkins J
    4. Adrion JR
    5. Latta Iv LC
    6. Schaack S

    (2021) Engines of change: Transposable element mutation rates are high and variable within Daphnia magna

    PLOS Genetics 17:e1009827.

    https://doi.org/10.1371/journal.pgen.1009827

    • PubMed
    • Google Scholar
67. 1. Hollister JD
    2. Gaut BS

    (2009) Epigenetic silencing of transposable elements: A trade-off between reduced transposition and deleterious effects on neighboring gene expression

    Genome Research 19:1419–1428.

    https://doi.org/10.1101/gr.091678.109

    • PubMed
    • Google Scholar
68. 1. Hollister JD
    2. Smith LM
    3. Guo YL
    4. Ott F
    5. Weigel D
    6. Gaut BS

    (2011) Transposable elements and small RNAs contribute to gene expression divergence between Arabidopsis thaliana and Arabidopsis lyrata

    PNAS 108:2322–2327.

    https://doi.org/10.1073/pnas.1018222108

    • PubMed
    • Google Scholar
69. 1. Holt C
    2. Yandell M

    (2011) MAKER2: an annotation pipeline and genome-database management tool for second-generation genome projects

    BMC Bioinformatics 12:491.

    https://doi.org/10.1186/1471-2105-12-491

    • PubMed
    • Google Scholar
70. 1. Hoskins RA
    2. Carlson JW
    3. Kennedy C
    4. Acevedo D
    5. Evans-Holm M
    6. Frise E
    7. Wan KH
    8. Park S
    9. Mendez-Lago M
    10. Rossi F
    11. Villasante A
    12. Dimitri P
    13. Karpen GH
    14. Celniker SE

    (2007) Sequence finishing and mapping of Drosophila melanogaster heterochromatin

    Science 316:1625–1628.

    https://doi.org/10.1126/science.1139816

    • PubMed
    • Google Scholar
71. 1. Hoskins RA
    2. Carlson JW
    3. Wan KH
    4. Park S
    5. Mendez I
    6. Galle SE
    7. Booth BW
    8. Pfeiffer BD
    9. George RA
    10. Svirskas R
    11. Krzywinski M
    12. Schein J
    13. Accardo MC
    14. Damia E
    15. Messina G
    16. Méndez-Lago M
    17. de Pablos B
    18. Demakova OV
    19. Andreyeva EN
    20. Boldyreva LV
    21. Marra M
    22. Carvalho AB
    23. Dimitri P
    24. Villasante A
    25. Zhimulev IF
    26. Rubin GM
    27. Karpen GH
    28. Celniker SE

    (2015) The Release 6 reference sequence of the Drosophila melanogaster genome

    Genome Research 25:445–458.

    https://doi.org/10.1101/gr.185579.114

    • PubMed
    • Google Scholar
72. 1. Hu TT
    2. Pattyn P
    3. Bakker EG
    4. Cao J
    5. Cheng JF
    6. Clark RM
    7. Fahlgren N
    8. Fawcett JA
    9. Grimwood J
    10. Gundlach H
    11. Haberer G
    12. Hollister JD
    13. Ossowski S
    14. Ottilar RP
    15. Salamov AA
    16. Schneeberger K
    17. Spannagl M
    18. Wang X
    19. Yang L
    20. Nasrallah ME
    21. Bergelson J
    22. Carrington JC
    23. Gaut BS
    24. Schmutz J
    25. Mayer KFX
    26. Van de Peer Y
    27. Grigoriev IV
    28. Nordborg M
    29. Weigel D
    30. Guo YL

    (2011) The Arabidopsis lyrata genome sequence and the basis of rapid genome size change

    Nature Genetics 43:476–481.

    https://doi.org/10.1038/ng.807

    • PubMed
    • Google Scholar
73. 1. Huang CRL
    2. Burns KH
    3. Boeke JD

    (2012) Active transposition in genomes

    Annual Review of Genetics 46:651–675.

    https://doi.org/10.1146/annurev-genet-110711-155616

    • PubMed
    • Google Scholar
74. Software

    1. Huang Y

    (2022) TE population frequency analysis, version swh:1:rev:24218fab83996f657e489402c5ff1c4cc06bfe9c

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:e8a29a082fb85376996ba58ccad333ab02738278;origin=https://github.com/harsh-shukla/TE_freq_analysis;visit=swh:1:snp:81910efac0d8d326402f111c4bd2fecd5c7696b3;anchor=swh:1:rev:24218fab83996f657e489402c5ff1c4cc06bfe9c
75. 1. Janssen A
    2. Breuer GA
    3. Brinkman EK
    4. van der Meulen AI
    5. Borden SV
    6. van Steensel B
    7. Bindra RS
    8. LaRocque JR
    9. Karpen GH

    (2016) A single double-strand break system reveals repair dynamics and mechanisms in heterochromatin and euchromatin

    Genes & Development 30:1645–1657.

    https://doi.org/10.1101/gad.283028.116

    • PubMed
    • Google Scholar
76. 1. Janssen A
    2. Colmenares SU
    3. Karpen GH

    (2018) Heterochromatin: guardian of the genheterochromatin: guardian of the genome

    Annual Review of Cell and Developmental Biology 34:265–288.

    https://doi.org/10.1146/annurev-cellbio-100617-062653

    • PubMed
    • Google Scholar
77. 1. Janssen A
    2. Colmenares SU
    3. Lee T
    4. Karpen GH

    (2019) Timely double-strand break repair and pathway choice in pericentromeric heterochromatin depend on the histone demethylase dkdm4a

    Genes & Development 33:103–115.

    https://doi.org/10.1101/gad.317537.118

    • PubMed
    • Google Scholar
78. 1. Kaminker JS
    2. Bergman CM
    3. Kronmiller B
    4. Carlson J
    5. Svirskas R
    6. Patel S
    7. Frise E
    8. Wheeler DA
    9. Lewis SE
    10. Rubin GM
    11. Ashburner M
    12. Celniker SE

    (2002) The transposable elements of the Drosophila melanogaster euchromatin: a genomics perspective

    Genome Biology 3:RESEARCH0084.

    https://doi.org/10.1186/gb-2002-3-12-research0084

    • PubMed
    • Google Scholar
79. 1. Kapitonov VV
    2. Jurka J

    (2003) Molecular paleontology of transposable elements in the Drosophila melanogaster genome

    PNAS 100:6569–6574.

    https://doi.org/10.1073/pnas.0732024100

    • PubMed
    • Google Scholar
80. 1. Kazazian HH

    (2004) Mobile elements: drivers of genome evolution

    Science 303:1626–1632.

    https://doi.org/10.1126/science.1089670

    • PubMed
    • Google Scholar
81. 1. Kelleher ES
    2. Barbash DA
    3. Blumenstiel JP

    (2020) Taming the turmoil within: New insights on the containment of transposable elements

    Trends in Genetics 36:474–489.

    https://doi.org/10.1016/j.tig.2020.04.007

    • PubMed
    • Google Scholar
82. 1. Kent TV
    2. Uzunović J
    3. Wright SI

    (2017) Coevolution between transposable elements and recombination

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 372:20160458.

    https://doi.org/10.1098/rstb.2016.0458

    • PubMed
    • Google Scholar
83. 1. Kharchenko PV
    2. Alekseyenko AA
    3. Schwartz YB
    4. Minoda A
    5. Riddle NC
    6. Ernst J
    7. Sabo PJ
    8. Larschan E
    9. Gorchakov AA
    10. Gu T
    11. Linder-Basso D
    12. Plachetka A
    13. Shanower G
    14. Tolstorukov MY
    15. Luquette LJ
    16. Xi R
    17. Jung YL
    18. Park RW
    19. Bishop EP
    20. Canfield TK
    21. Sandstrom R
    22. Thurman RE
    23. MacAlpine DM
    24. Stamatoyannopoulos JA
    25. Kellis M
    26. Elgin SCR
    27. Kuroda MI
    28. Pirrotta V
    29. Karpen GH
    30. Park PJ

    (2011) Comprehensive analysis of the chromatin landscape in Drosophila melanogaster

    Nature 471:480–485.

    https://doi.org/10.1038/nature09725

    • PubMed
    • Google Scholar
84. 1. Khost DE
    2. Eickbush DG
    3. Larracuente AM

    (2017) Single-molecule sequencing resolves the detailed structure of complex satellite DNA loci in Drosophila melanogaster

    Genome Research 27:709–721.

    https://doi.org/10.1101/gr.213512.116

    • PubMed
    • Google Scholar
85. 1. Kokot M
    2. Długosz M
    3. Deorowicz S
    4. Berger B

    (2017) KMC 3: counting and manipulating k-mer statistics

    Bioinformatics 33:2759–2761.

    https://doi.org/10.1093/bioinformatics/btx304

    • Google Scholar
86. 1. Koren S
    2. Walenz BP
    3. Berlin K
    4. Miller JR
    5. Bergman NH
    6. Phillippy AM

    (2017) Canu: scalable and accurate long-read assembly via adaptive k-mer weighting and repeat separation

    Genome Research 27:722–736.

    https://doi.org/10.1101/gr.215087.116

    • PubMed
    • Google Scholar
87. 1. Kouzarides T

    (2007) Chromatin modifications and their function

    Cell 128:693–705.

    https://doi.org/10.1016/j.cell.2007.02.005

    • PubMed
    • Google Scholar
88. 1. Lack JB
    2. Cardeno CM
    3. Crepeau MW
    4. Taylor W
    5. Corbett-Detig RB
    6. Stevens KA
    7. Langley CH
    8. Pool JE

    (2015) The Drosophila genome nexus: A population genomic resource of 623 Drosophila melanogaster genomes, including 197 from A single ancestral range population

    Genetics 199:1229–1241.

    https://doi.org/10.1534/genetics.115.174664

    • PubMed
    • Google Scholar
89. 1. Lam KWG
    2. Brick K
    3. Cheng G
    4. Pratto F
    5. Camerini-Otero RD

    (2019) Cell-type-specific genomics reveals histone modification dynamics in mammalian meiosis

    Nature Communications 10:3821.

    https://doi.org/10.1038/s41467-019-11820-7

    • PubMed
    • Google Scholar
90. 1. Langley C H
    2. Montgomery E
    3. Hudson R
    4. Kaplan N
    5. Charlesworth B

    (1988) On the role of unequal exchange in the containment of transposable element copy number

    Genetical Research 52:223–235.

    https://doi.org/10.1017/s0016672300027695

    • PubMed
    • Google Scholar
91. 1. Langley CH
    2. Stevens K
    3. Cardeno C
    4. Lee YCG
    5. Schrider DR
    6. Pool JE
    7. Langley SA
    8. Suarez C
    9. Corbett-Detig RB
    10. Kolaczkowski B
    11. Fang S
    12. Nista PM
    13. Holloway AK
    14. Kern AD
    15. Dewey CN
    16. Song YS
    17. Hahn MW
    18. Begun DJ

    (2012) Genomic variation in natural populations of Drosophila melanogaster

    Genetics 192:533–598.

    https://doi.org/10.1534/genetics.112.142018

    • PubMed
    • Google Scholar
92. 1. Larracuente AM

    (2014) The organization and evolution of the Responder satellite in species of the Drosophila melanogaster group: dynamic evolution of a target of meiotic drive

    BMC Evolutionary Biology 14:233.

    https://doi.org/10.1186/s12862-014-0233-9

    • PubMed
    • Google Scholar
93. 1. Larson DE
    2. Abel HJ
    3. Chiang C
    4. Badve A
    5. Das I
    6. Eldred JM
    7. Layer RM
    8. Hall IM

    (2019) svtools: population-scale analysis of structural variation

    Bioinformatics 35:4782–4787.

    https://doi.org/10.1093/bioinformatics/btz492

    • PubMed
    • Google Scholar
94. 1. Layer RM
    2. Chiang C
    3. Quinlan AR
    4. Hall IM

    (2014) LUMPY: a probabilistic framework for structural variant discovery

    Genome Biology 15:R84.

    https://doi.org/10.1186/gb-2014-15-6-r84

    • PubMed
    • Google Scholar
95. 1. Le Thomas A
    2. Rogers AK
    3. Webster A
    4. Marinov GK
    5. Liao SE
    6. Perkins EM
    7. Hur JK
    8. Aravin AA
    9. Tóth KF

    (2013) Piwi induces piRNA-guided transcriptional silencing and establishment of a repressive chromatin state

    Genes & Development 27:390–399.

    https://doi.org/10.1101/gad.209841.112

    • PubMed
    • Google Scholar
96. 1. Lee YCG
    2. Langley CH

    (2010) Transposable elements in natural populations of Drosophila melanogaster

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 365:1219–1228.

    https://doi.org/10.1098/rstb.2009.0318

    • PubMed
    • Google Scholar
97. 1. Lee YCG

    (2015) The role of piRNA-mediated epigenetic silencing in the population dynamics of transposable elements in Drosophila melanogaster

    PLOS Genetics 11:e1005269.

    https://doi.org/10.1371/journal.pgen.1005269

    • PubMed
    • Google Scholar
98. 1. Lee YCG
    2. Karpen GH

    (2017) Pervasive epigenetic effects of Drosophila euchromatic transposable elements impact their evolution

    eLife 6:e25762.

    https://doi.org/10.7554/eLife.25762

    • PubMed
    • Google Scholar
99. 1. Lee YCG
    2. Ogiyama Y
    3. Martins NMC
    4. Beliveau BJ
    5. Acevedo D
    6. Wu C
    7. Cavalli G
    8. Karpen GH

    (2020) Pericentromeric heterochromatin is hierarchically organized and spatially contacts h3k9me2 islands in euchromatin

    PLOS Genetics 16:e1008673.

    https://doi.org/10.1371/journal.pgen.1008673

    • PubMed
    • Google Scholar
100. 1. Legrand S
     2. Caron T
     3. Maumus F
     4. Schvartzman S
     5. Quadrana L
     6. Durand E
     7. Gallina S
     8. Pauwels M
     9. Mazoyer C
     10. Huyghe L
     11. Colot V
     12. Hanikenne M
     13. Castric V

     (2019) Differential retention of transposable element-derived sequences in outcrossing Arabidopsis genomes

     Mobile DNA 10:30.

     https://doi.org/10.1186/s13100-019-0171-6

     • PubMed
     • Google Scholar
101. 1. Levine MT
     2. McCoy C
     3. Vermaak D
     4. Lee YCG
     5. Hiatt MA
     6. Matsen FA
     7. Malik HS

     (2012) Phylogenomic analysis reveals dynamic evolutionary history of the Drosophila Heterochromatin Protein 1 (HP1) gene family

     PLOS Genetics 8:e1002729.

     https://doi.org/10.1371/journal.pgen.1002729

     • PubMed
     • Google Scholar
102. 1. Li H
     2. Durbin R

     (2009) Fast and accurate short read alignment with burrows-wheeler transform

     Bioinformatics 25:1754–1760.

     https://doi.org/10.1093/bioinformatics/btp324

     • PubMed
     • Google Scholar
103. 1. Li H

     (2011) A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data

     Bioinformatics 27:2987–2993.

     https://doi.org/10.1093/bioinformatics/btr509

     • PubMed
     • Google Scholar
104. 1. Li Q
     2. Brown JB
     3. Huang H
     4. Bickel PJ

     (2011) Measuring reproducibility of high-throughput experiments

     The Annals of Applied Statistics 5:1752–1779.

     https://doi.org/10.1214/11-AOAS466

     • Google Scholar
105. 1. Li H

     (2018) Minimap2: pairwise alignment for nucleotide sequences

     Bioinformatics 34:3094–3100.

     https://doi.org/10.1093/bioinformatics/bty191

     • PubMed
     • Google Scholar
106. Software

     1. Li H

     (2022) Seqtk, version v1.3-r107-dirty

     GitHub.

     https://github.com/lh3/seqtk
107. 1. Locke J
     2. Kotarski MA
     3. Tartof KD

     (1988) Dosage-dependent modifiers of position effect variegation in Drosophila and a mass action model that explains their effect

     Genetics 120:181–198.

     https://doi.org/10.1093/genetics/120.1.181

     • PubMed
     • Google Scholar
108. 1. Lockton S
     2. Gaut BS

     (2010) The evolution of transposable elements in natural populations of self-fertilizing Arabidopsis thaliana and its outcrossing relative Arabidopsis lyrata

     BMC Evolutionary Biology 10:10.

     https://doi.org/10.1186/1471-2148-10-10

     • PubMed
     • Google Scholar
109. 1. Lynch M
     2. Conery JS

     (2003) The origins of genome complexity

     Science 302:1401–1404.

     https://doi.org/10.1126/science.1089370

     • PubMed
     • Google Scholar
110. 1. Martins EP
     2. Hansen TF

     (1997) Phylogenies and the comparative method: A general approach to incorporating phylogenetic information into the analysis of interspecific data

     The American Naturalist 149:646–667.

     https://doi.org/10.1086/286013

     • Google Scholar
111. 1. Meiklejohn CD
     2. Landeen EL
     3. Gordon KE
     4. Rzatkiewicz T
     5. Kingan SB
     6. Geneva AJ
     7. Vedanayagam JP
     8. Muirhead CA
     9. Garrigan D
     10. Stern DL
     11. Presgraves DC

     (2018) Gene flow mediates the role of sex chromosome meiotic drive during complex speciation

     eLife 7:e35468.

     https://doi.org/10.7554/eLife.35468

     • PubMed
     • Google Scholar
112. 1. Meng D
     2. Dubin M
     3. Zhang P
     4. Osborne EJ
     5. Stegle O
     6. Clark RM
     7. Nordborg M

     (2016) Limited contribution of DNA methylation variation to expression regulation in Arabidopsis thaliana

     PLOS Genetics 12:e1006141.

     https://doi.org/10.1371/journal.pgen.1006141

     • PubMed
     • Google Scholar
113. 1. Mérel V
     2. Gibert P
     3. Buch I
     4. Rodriguez Rada V
     5. Estoup A
     6. Gautier M
     7. Fablet M
     8. Boulesteix M
     9. Vieira C

     (2021) The worldwide invasion of Drosophila suzukii is accompanied by a large increase of transposable element load and a small number of putatively adaptive insertions

     Molecular Biology and Evolution 38:4252–4267.

     https://doi.org/10.1093/molbev/msab155

     • PubMed
     • Google Scholar
114. 1. Metzger BPH
     2. Yuan DC
     3. Gruber JD
     4. Duveau F
     5. Wittkopp PJ

     (2015) Selection on noise constrains variation in a eukaryotic promoter

     Nature 521:344–347.

     https://doi.org/10.1038/nature14244

     • PubMed
     • Google Scholar
115. 1. Mills WK
     2. Lee YCG
     3. Kochendoerfer AM
     4. Dunleavy EM
     5. Karpen GH

     (2019) RNA from a simple-tandem repeat is required for sperm maturation and male fertility in Drosophila melanogaster

     eLife 8:e48940.

     https://doi.org/10.7554/eLife.48940

     • PubMed
     • Google Scholar
116. 1. Montgomery EA
     2. Huang SM
     3. Langley CH
     4. Judd BH

     (1991) Chromosome rearrangement by ectopic recombination in Drosophila melanogaster: genome structure and evolution

     Genetics 129:1085–1098.

     https://doi.org/10.1093/genetics/129.4.1085

     • PubMed
     • Google Scholar
117. 1. Nakato R
     2. Shirahige K

     (2017) Recent advances in ChIP-seq analysis: from quality management to whole-genome annotation

     Briefings in Bioinformatics 18:279–290.

     https://doi.org/10.1093/bib/bbw023

     • PubMed
     • Google Scholar
118. 1. Nègre N
     2. Brown CD
     3. Ma L
     4. Bristow CA
     5. Miller SW
     6. Wagner U
     7. Kheradpour P
     8. Eaton ML
     9. Loriaux P
     10. Sealfon R
     11. Li Z
     12. Ishii H
     13. Spokony RF
     14. Chen J
     15. Hwang L
     16. Cheng C
     17. Auburn RP
     18. Davis MB
     19. Domanus M
     20. Shah PK
     21. Morrison CA
     22. Zieba J
     23. Suchy S
     24. Senderowicz L
     25. Victorsen A
     26. Bild NA
     27. Grundstad AJ
     28. Hanley D
     29. MacAlpine DM
     30. Mannervik M
     31. Venken K
     32. Bellen H
     33. White R
     34. Gerstein M
     35. Russell S
     36. Grossman RL
     37. Ren B
     38. Posakony JW
     39. Kellis M
     40. White KP

     (2011) A cis-regulatory map of the Drosophila genome

     Nature 471:527–531.

     https://doi.org/10.1038/nature09990

     • PubMed
     • Google Scholar
119. 1. Ninova M
     2. Godneeva B
     3. Chen YCA
     4. Luo Y
     5. Prakash SJ
     6. Jankovics F
     7. Erdélyi M
     8. Aravin AA
     9. Fejes Tóth K

     (2020) The SUMO Ligase Su(var)2-10 controls hetero- and euchromatic gene expression via establishing H3K9 trimethylation and negative feedback regulation

     Molecular Cell 77:571–585.

     https://doi.org/10.1016/j.molcel.2019.09.033

     • PubMed
     • Google Scholar
120. 1. Nowoshilow S
     2. Schloissnig S
     3. Fei JF
     4. Dahl A
     5. Pang AWC
     6. Pippel M
     7. Winkler S
     8. Hastie AR
     9. Young G
     10. Roscito JG
     11. Falcon F
     12. Knapp D
     13. Powell S
     14. Cruz A
     15. Cao H
     16. Habermann B
     17. Hiller M
     18. Tanaka EM
     19. Myers EW

     (2018) The axolotl genome and the evolution of key tissue formation regulators

     Nature 554:50–55.

     https://doi.org/10.1038/nature25458

     • PubMed
     • Google Scholar
121. 1. Obbard DJ
     2. Maclennan J
     3. Kim KW
     4. Rambaut A
     5. O’Grady PM
     6. Jiggins FM

     (2012) Estimating divergence dates and substitution rates in the Drosophila phylogeny

     Molecular Biology and Evolution 29:3459–3473.

     https://doi.org/10.1093/molbev/mss150

     • PubMed
     • Google Scholar
122. 1. Oggenfuss U
     2. Badet T
     3. Wicker T
     4. Hartmann FE
     5. Singh NK
     6. Abraham L
     7. Karisto P
     8. Vonlanthen T
     9. Mundt C
     10. McDonald BA
     11. Croll D

     (2021) A population-level invasion by transposable elements triggers genome expansion in A fungal pathogen

     eLife 10:e69249.

     https://doi.org/10.7554/eLife.69249

     • PubMed
     • Google Scholar
123. 1. Park PJ

     (2009) ChIP-seq: advantages and challenges of a maturing technology

     Nature Reviews. Genetics 10:669–680.

     https://doi.org/10.1038/nrg2641

     • PubMed
     • Google Scholar
124. 1. Petrov DA
     2. Aminetzach YT
     3. Davis JC
     4. Bensasson D
     5. Hirsh AE

     (2003) Size matters: non-LTR retrotransposable elements and ectopic recombination in Drosophila

     Molecular Biology and Evolution 20:880–892.

     https://doi.org/10.1093/molbev/msg102

     • PubMed
     • Google Scholar
125. 1. Petrov DA
     2. Fiston-Lavier AS
     3. Lipatov M
     4. Lenkov K
     5. González J

     (2011) Population genomics of transposable elements in Drosophila melanogaster

     Molecular Biology and Evolution 28:1633–1644.

     https://doi.org/10.1093/molbev/msq337

     • PubMed
     • Google Scholar
126. 1. Pezic D
     2. Manakov SA
     3. Sachidanandam R
     4. Aravin AA

     (2014) piRNA pathway targets active LINE1 elements to establish the repressive H3K9me3 mark in germ cells

     Genes & Development 28:1410–1428.

     https://doi.org/10.1101/gad.240895.114

     • PubMed
     • Google Scholar
127. 1. Pikaard CS
     2. Mittelsten Scheid O

     (2014) Epigenetic regulation in plants

     Cold Spring Harbor Perspectives in Biology 6:a019315.

     https://doi.org/10.1101/cshperspect.a019315

     • PubMed
     • Google Scholar
128. 1. Presgraves DC

     (2006) Intron length evolution in Drosophila

     Molecular Biology and Evolution 23:2203–2213.

     https://doi.org/10.1093/molbev/msl094

     • PubMed
     • Google Scholar
129. 1. Qi Y
     2. He X
     3. Wang XJ
     4. Kohany O
     5. Jurka J
     6. Hannon GJ

     (2006) Distinct catalytic and non-catalytic roles of ARGONAUTE4 in RNA-directed DNA methylation

     Nature 443:1008–1012.

     https://doi.org/10.1038/nature05198

     • PubMed
     • Google Scholar
130. 1. Quadrana L
     2. Bortolini Silveira A
     3. Mayhew GF
     4. LeBlanc C
     5. Martienssen RA
     6. Jeddeloh JA
     7. Colot V

     (2016) The Arabidopsis thaliana mobilome and its impact at the species level

     eLife 5:e15716.

     https://doi.org/10.7554/eLife.15716

     • PubMed
     • Google Scholar
131. 1. Quinlan AR
     2. Hall IM

     (2010) BEDTools: a flexible suite of utilities for comparing genomic features

     Bioinformatics 26:841–842.

     https://doi.org/10.1093/bioinformatics/btq033

     • PubMed
     • Google Scholar
132. 1. Rebollo R
     2. Miceli-Royer K
     3. Zhang Y
     4. Farivar S
     5. Gagnier L
     6. Mager DL

     (2012) Epigenetic interplay between mouse endogenous retroviruses and host genes

     Genome Biology 13:R89.

     https://doi.org/10.1186/gb-2012-13-10-r89

     • PubMed
     • Google Scholar
133. 1. Rech GE
     2. Radío S
     3. Guirao-Rico S
     4. Aguilera L
     5. Horvath V
     6. Green L
     7. Lindstadt H
     8. Jamilloux V
     9. Quesneville H
     10. González J

     (2022) Population-scale long-read sequencing uncovers transposable elements associated with gene expression variation and adaptive signatures in Drosophila

     Nature Communications 13:1948.

     https://doi.org/10.1038/s41467-022-29518-8

     • PubMed
     • Google Scholar
134. 1. Riddle NC
     2. Minoda A
     3. Kharchenko PV
     4. Alekseyenko AA
     5. Schwartz YB
     6. Tolstorukov MY
     7. Gorchakov AA
     8. Jaffe JD
     9. Kennedy C
     10. Linder-Basso D
     11. Peach SE
     12. Shanower G
     13. Zheng H
     14. Kuroda MI
     15. Pirrotta V
     16. Park PJ
     17. Elgin SCR
     18. Karpen GH

     (2011) Plasticity in patterns of histone modifications and chromosomal proteins in Drosophila heterochromatin

     Genome Research 21:147–163.

     https://doi.org/10.1101/gr.110098.110

     • PubMed
     • Google Scholar
135. 1. Rius N
     2. Guillén Y
     3. Delprat A
     4. Kapusta A
     5. Feschotte C
     6. Ruiz A

     (2016) Exploration of the Drosophila buzzatii transposable element content suggests underestimation of repeats in Drosophila genomes

     BMC Genomics 17:344.

     https://doi.org/10.1186/s12864-016-2648-8

     • PubMed
     • Google Scholar
136. 1. Rogers RL
     2. Cridland JM
     3. Shao L
     4. Hu TT
     5. Andolfatto P
     6. Thornton KR

     (2014) Landscape of standing variation for tandem duplications in Drosophila yakuba and Drosophila simulans

     Molecular Biology and Evolution 31:1750–1766.

     https://doi.org/10.1093/molbev/msu124

     • PubMed
     • Google Scholar
137. 1. Ross BD
     2. Rosin L
     3. Thomae AW
     4. Hiatt MA
     5. Vermaak D
     6. de la Cruz AFA
     7. Imhof A
     8. Mellone BG
     9. Malik HS

     (2013) Stepwise evolution of essential centromere funcstepwise evolution of essential centromere function in a Drosophila neogene

     Science 340:1211–1214.

     https://doi.org/10.1126/science.1234393

     • PubMed
     • Google Scholar
138. 1. Rudolph T
     2. Yonezawa M
     3. Lein S
     4. Heidrich K
     5. Kubicek S
     6. Schäfer C
     7. Phalke S
     8. Walther M
     9. Schmidt A
     10. Jenuwein T
     11. Reuter G

     (2007) Heterochromatin formation in Drosophila is initiated through active removal of H3K4 methylation by the LSD1 homolog SU(VAR)3-3

     Molecular Cell 26:103–115.

     https://doi.org/10.1016/j.molcel.2007.02.025

     • PubMed
     • Google Scholar
139. 1. Sabl JF
     2. Henikoff S

     (1996) Copy number and orientation determine the susceptibility of a gene to silencing by nearby heterochromatin in Drosophila

     Genetics 142:447–458.

     https://doi.org/10.1093/genetics/142.2.447

     • PubMed
     • Google Scholar
140. 1. Salzberg SL
     2. Yorke JA

     (2005) Beware of mis-assembled genomes

     Bioinformatics 21:4320–4321.

     https://doi.org/10.1093/bioinformatics/bti769

     • PubMed
     • Google Scholar
141. 1. Sasaki E
     2. Kawakatsu T
     3. Ecker JR
     4. Nordborg M

     (2019) Common alleles of CMT2 and NRPE1 are major determinants of CHH methylation variation in Arabidopsis thaliana

     PLOS Genetics 15:e1008492.

     https://doi.org/10.1371/journal.pgen.1008492

     • PubMed
     • Google Scholar
142. 1. Sentmanat MF
     2. Elgin SCR

     (2012) Ectopic assembly of heterochromatin in Drosophila melanogaster triggered by transposable elements

     PNAS 109:14104–14109.

     https://doi.org/10.1073/pnas.1207036109

     • PubMed
     • Google Scholar
143. 1. Sexton T
     2. Yaffe E
     3. Kenigsberg E
     4. Bantignies F
     5. Leblanc B
     6. Hoichman M
     7. Parrinello H
     8. Tanay A
     9. Cavalli G

     (2012) Three-dimensional folding and functional organization principles of the Drosophila genome

     Cell 148:458–472.

     https://doi.org/10.1016/j.cell.2012.01.010

     • PubMed
     • Google Scholar
144. 1. Sienski G
     2. Dönertas D
     3. Brennecke J

     (2012) Transcriptional silencing of transposons by Piwi and maelstrom and its impact on chromatin state and gene expression

     Cell 151:964–980.

     https://doi.org/10.1016/j.cell.2012.10.040

     • PubMed
     • Google Scholar
145. 1. Singh N
     2. Morlock H
     3. Hanes SD

     (2011) The Bin3 RNA methyltransferase is required for repression of caudal translation in the Drosophila embryo

     Developmental Biology 352:104–115.

     https://doi.org/10.1016/j.ydbio.2011.01.017

     • PubMed
     • Google Scholar
146. 1. Slotkin RK
     2. Martienssen R

     (2007) Transposable elements and the epigenetic regulation of the genome

     Nature Reviews. Genetics 8:272–285.

     https://doi.org/10.1038/nrg2072

     • PubMed
     • Google Scholar
147. 1. Stranger BE
     2. Forrest MS
     3. Dunning M
     4. Ingle CE
     5. Beazley C
     6. Thorne N
     7. Redon R
     8. Bird CP
     9. de Grassi A
     10. Lee C
     11. Tyler-Smith C
     12. Carter N
     13. Scherer SW
     14. Tavaré S
     15. Deloukas P
     16. Hurles ME
     17. Dermitzakis ET

     (2007) Relative impact of nucleotide and copy number variation on gene expression phenotypes

     Science 315:848–853.

     https://doi.org/10.1126/science.1136678

     • PubMed
     • Google Scholar
148. 1. Stuart T
     2. Eichten SR
     3. Cahn J
     4. Karpievitch YV
     5. Borevitz JO
     6. Lister R

     (2016) Population scale mapping of transposable element diversity reveals links to gene regulation and epigenomic variation

     eLife 5:e20777.

     https://doi.org/10.7554/eLife.20777

     • PubMed
     • Google Scholar
149. 1. Swenson JM
     2. Colmenares SU
     3. Strom AR
     4. Costes SV
     5. Karpen GH

     (2016) The composition and organization of Drosophila heterochromatin are heterogeneous and dynamic

     eLife 5:e16096.

     https://doi.org/10.7554/eLife.16096

     • PubMed
     • Google Scholar
150. 1. Szitenberg A
     2. Cha S
     3. Opperman CH
     4. Bird DM
     5. Blaxter ML
     6. Lunt DH

     (2016) Genetic drift, not life history or RNAi, determine long-term evolution of transposable elements

     Genome Biology and Evolution 8:2964–2978.

     https://doi.org/10.1093/gbe/evw208

     • PubMed
     • Google Scholar
151. 1. Talbert PB
     2. Henikoff S

     (2000) A reexamination of spreading of position-effect variegation in the white-roughest region of Drosophila melanogaster

     Genetics 154:259–272.

     https://doi.org/10.1093/genetics/154.1.259

     • PubMed
     • Google Scholar
152. 1. Thorvaldsdóttir H
     2. Robinson JT
     3. Mesirov JP

     (2013) Integrative Genomics Viewer (IGV): high-performance genomics data visualization and exploration

     Briefings in Bioinformatics 14:178–192.

     https://doi.org/10.1093/bib/bbs017

     • PubMed
     • Google Scholar
153. 1. Treangen TJ
     2. Salzberg SL

     (2011) Repetitive DNA and next-generation sequencing: computational challenges and solutions

     Nature Reviews. Genetics 13:36–46.

     https://doi.org/10.1038/nrg3117

     • PubMed
     • Google Scholar
154. 1. Turissini DA
     2. Matute DR

     (2017) Fine scale mapping of genomic introgressions within the Drosophila yakuba clade

     PLOS Genetics 13:e1006971.

     https://doi.org/10.1371/journal.pgen.1006971

     • PubMed
     • Google Scholar
155. 1. Ullastres A
     2. Merenciano M
     3. González J

     (2021) Regulatory regions in natural transposable element insertions drive interindividual differences in response to immune challenges in Drosophila

     Genome Biology 22:265.

     https://doi.org/10.1186/s13059-021-02471-3

     • PubMed
     • Google Scholar
156. 1. Vogel MJ
     2. Pagie L
     3. Talhout W
     4. Nieuwland M
     5. Kerkhoven RM
     6. van Steensel B

     (2009) High-resolution mapping of heterochromatin redistribution in a Drosophila position-effect variegation model

     Epigenetics & Chromatin 2:1.

     https://doi.org/10.1186/1756-8935-2-1

     • PubMed
     • Google Scholar
157. 1. Volff JN
     2. Bouneau L
     3. Ozouf-Costaz C
     4. Fischer C

     (2003) Diversity of retrotransposable elements in compact pufferfish genomes

     Trends in Genetics 19:674–678.

     https://doi.org/10.1016/j.tig.2003.10.006

     • PubMed
     • Google Scholar
158. 1. Wang SH
     2. Elgin SCR

     (2011) Drosophila Piwi functions downstream of piRNA production mediating a chromatin-based transposon silencing mechanism in female germ line

     PNAS 108:21164–21169.

     https://doi.org/10.1073/pnas.1107892109

     • PubMed
     • Google Scholar
159. 1. Wei KHC
     2. Grenier JK
     3. Barbash DA
     4. Clark AG

     (2014) Correlated variation and population differentiation in satellite DNA abundance among lines of Drosophila melanogaster

     PNAS 111:18793–18798.

     https://doi.org/10.1073/pnas.1421951112

     • PubMed
     • Google Scholar
160. 1. Wei KHC
     2. Lower SE
     3. Caldas IV
     4. Sless TJS
     5. Barbash DA
     6. Clark AG

     (2018) Variable rates of simple satellite gains across the Drosophila phylogeny

     Molecular Biology and Evolution 35:925–941.

     https://doi.org/10.1093/molbev/msy005

     • PubMed
     • Google Scholar
161. 1. Wei KHC
     2. Gibilisco L
     3. Bachtrog D

     (2020) Epigenetic conflict on a degenerating Y chromosome increases mutational burden in Drosophila males

     Nature Communications 11:5537.

     https://doi.org/10.1038/s41467-020-19134-9

     • PubMed
     • Google Scholar
162. 1. Wells JN
     2. Feschotte C

     (2020) A field guide to eukaryotic transposable elements

     Annual Review of Genetics 54:539–561.

     https://doi.org/10.1146/annurev-genet-040620-022145

     • PubMed
     • Google Scholar
163. 1. Wicker T
     2. Sabot F
     3. Hua-Van A
     4. Bennetzen JL
     5. Capy P
     6. Chalhoub B
     7. Flavell A
     8. Leroy P
     9. Morgante M
     10. Panaud O
     11. Paux E
     12. SanMiguel P
     13. Schulman AH

     (2007) A unified classification system for eukaryotic transposable elements

     Nature Reviews. Genetics 8:973–982.

     https://doi.org/10.1038/nrg2165

     • PubMed
     • Google Scholar
164. 1. Wong WY
     2. Simakov O
     3. Bridge DM
     4. Cartwright P
     5. Bellantuono AJ
     6. Kuhn A
     7. Holstein TW
     8. David CN
     9. Steele RE
     10. Martínez DE

     (2019) Expansion of a single transposable element family is associated with genome-size increase and radiation in the genus Hydra

     PNAS 116:22915–22917.

     https://doi.org/10.1073/pnas.1910106116

     • PubMed
     • Google Scholar
165. 1. Wright SI
     2. Schoen DJ

     (1999)

     Transposon dynamics and the breeding system

     Genetica 107:139–148.

     • PubMed
     • Google Scholar
166. 1. Yang HP
     2. Barbash DA

     (2008) Abundant and species-specific DINE-1 transposable elements in 12 Drosophila genomes

     Genome Biology 9:R39.

     https://doi.org/10.1186/gb-2008-9-2-r39

     • PubMed
     • Google Scholar
167. 1. Yasuhara JC
     2. Wakimoto BT

     (2008) Molecular landscape of modified histones in Drosophila heterochromatic genes and euchromatin-heterochromatin transition zones

     PLOS Genetics 4:e16.

     https://doi.org/10.1371/journal.pgen.0040016

     • PubMed
     • Google Scholar
168. 1. Zhang Y
     2. Liu T
     3. Meyer CA
     4. Eeckhoute J
     5. Johnson DS
     6. Bernstein BE
     7. Nusbaum C
     8. Myers RM
     9. Brown M
     10. Li W
     11. Liu XS

     (2008) Model-based analysis of chIP-Seq (MACS)

     Genome Biology 9:R137.

     https://doi.org/10.1186/gb-2008-9-9-r137

     • PubMed
     • Google Scholar

[Editors' note: this paper was reviewed by Review Commons.]

1. General Statements

This study builds upon a previous study published in eLife (Lee and Karpen 2017), which established the importance of transposable element (TE)-mediated local enrichment of repressive marks in TE evolution in the model species, D. melanogaster. In the current study, we investigated this TE-mediated phenomenon in six closely related Drosophila species to have a good phylogenetic resolution, aiming to address step-by-step how such TE-mediated effects may shape the evolution of genomic TE content, a long-standing question in evolutionary genomics. Our analyses successfully connect species-specific host chromatin regulation, TE-mediated epigenetic effects, the strength of natural selection against TEs, and genomic TE abundance unique to individual species, while uncovering the evolutionary causes for the wide variety of TE profiles between species.

We appreciate the reviewers for their overall positive response and helpful comments on the previous versions of the manuscript. We appreciate the opportunity to correct our previous oversights, revise analysis to corroborate findings, and improve interpretations of our observations. Thanks for these suggestions, we believe the revised manuscript is much improved and hope it is now acceptable for publication in eLife. Please see below for our pointby-point responses to reviewers’ comments.

In addition to analyses suggested by the reviewers, we also updated one of our analyses to avoid non-independence of samples. Specifically, in Figures 3A and B and Supplementary Figures S1-6, we investigated the associations between TE-mediated epigenetic effects and genic epigenetic states/expression. In our previous analysis, a TE could be associated with more than one gene, if it is the nearest TE to more than one gene. Yet, this could lead to nonindependence of observations. In this revised manuscript, we restricted the analysis to TE and its nearest gene, excluding this potential issue. The results are qualitatively the same and quantitatively similar, and our conclusions from these results remain unchanged.

2. Point-by-point description of the revisions

Reviewer #1 (Evidence, reproducibility and clarity (Required)):

Summary:

In this work titled "Species-specific chromatin landscape determines how transposable elements shape genome evolution", Huang, Shukla and Lee examined the epigenetic effect of euchromatic TE insertions in 6 Drosophila species (3 in melanogaster species complex and 3 in yakuba species complex). They profiled the H3K9me2 mark and transcriptome from 16-18hr embryos and, using genomics, population genetics and statistical tools, they extended claims made previously in D. mel to 5 more species: (1) H3K9me2 spreads to nearby regions, exerting an epigenetic effect; (2) TEs with epigenetic effects are selected against. More importantly, this work aims to identify rules that explain the differences in TE-mediated epigenetic effects across the 6 species examined and how their distinct chromatin landscape might affect the way in which TE-mediated epigenetic effects shape genome evolution. The analysis seems to be well done, rather rigorous, and considers important counterarguments in several places throughout the manuscript. The figure is clear, and the method is well detailed. I have some comments that should be addressed before publication, involving additional bioinformatic analysis and text edits to either tone down or re-frame the conclusions.

Major comments:

1. Exclusive usage of Spearman rank correlation for key conclusions requires further support. I worry that using Spearman rank correlation loses too much of the quantitative information in e.g., gene expression. Have authors ever looked at Pearson's correlation (perhaps upon log-transformed levels when necessary) and does it lead to similar or different results? While this concern holds for all conclusions based on Spearman rank correlation, it is particularly a concern for analysis involving gene expression, especially for the surprising claim that TE-mediated K9 spreading does not affect adjacent gene expression.

We agree with the reviewer that, compared to parametric tests (e.g., Pearson tests), nonparametric tests have more restricted statistical power. An insignificant Spearman correlation test could either be due to true biological reasons or simply a lack of statistical power. Yet, we found that the magnitude and extent of TE-mediated H3K9me2 enrichment and the genic HMD do not follow normal distribution even after log transformation (Anderson-Darling normality test, all p-value < 10^-16). Accordingly, we are unable to perform the suggested Pearson correlation tests. Another reason why we would like to stick with the Spearman tests is because it is less sensitive to outliers than the Pearson correlation tests. This is especially important because our goal is to identify genome-wide average patterns.

It is worth noting that, in addition to Spearman rank correlation analysis, we also compared the expression level between homologous alleles with and without TEs showing H3K9me2 spreading in D. simulans and D. yakuba (Figure 3C and Figure 3 – Supplementary Figure S7-9) and found consistent results. We included additional discussion to explain why we chose to use nonparametric tests (p50, line 12-18; p51, line 27-31) and emphasize that, in addition to the correlation analysis, other tests also support limited impacts of TE-mediated H3K9me2 enrichment on gene expression (p39, line 13-15).

We agree with the reviewer that the observation of the limited associations between TE-mediated H3K9me2 spreading and gene expression is surprising. We hope our addressing the reviewer’s comment 3 and additional discussions and citations of other relevant/suggestive studies that reached similar conclusions (p39, line 15-22) help address this issue.

2. Choice of binning procedure over correlation analysis seems unjustified. I understand the binary binning in e.g., Figure 1C, but for quantitative variables e.g., distance to TE/gene, allele frequency in Figures3 and 4, I worry that the binning procedure based on somewhat arbitrary cutoffs could mask the real relationship. Are the conclusions robust against the cutoff chosen?

Is there actual correlation between the two variables examined?

For previous Figure 3A and 3B (now Figure 3 – Supplementary Figures 1-4), following the reviewer’s suggestion, we performed regression analysis that uses the distance between genes and TEs as continuous variables to test whether the associations between TE-mediated epigenetic effects and genic H3K9me2 enrichment/expression depends on TE-gene distance. Such analysis reached the same conclusion and is now included in p15, line31 – p16, line 4; p16, line 16-19. We chose to present the analyses that bin genes based on their distance to TEs because it is challenging to interpret them as three-dimensional graphs; current Figure 3 – Supplementary Figures 1-4 have TE-mediated epigenetic effects as X-axis, genic attribute as Y-axis, and bins of gene-TE distance as a binary color. It is worth noting that, in this revised version of the manuscript, we chose to present the correlation coefficients between TE-mediated H3K9me2 enrichment and genic epigenetic states/expression (new Figures 3A and 3B). This is because these figures provide a good visual summary for all the comparisons (magnitude/extent, genic H3K9me2 enrichment/expression, and TEs 5’ and 3’ to genes) and are easier to be interpreted than original Figures 3A and 3B (now Figure 3 – Supplementary Figures 1-4).

In Figure 3D, we estimated the z-score to compare the extent of TE-mediated spreading of H3K9me2 on the genic and intergenic sides between TEs close and distant from genes (the binning is based on 50% percentile of the distance between TEs and genes). Following the reviewer’s suggestion, we performed correlation tests between the z-score and TE-gene distance and found consistent results – we identified significant negative associations between the z-score of a TE and its distance from the nearest gene. These results are now presented in p19, line18-23.

For the population frequencies analysis (Figure 4), we were unclear about how we categorized low-frequency and high-frequency TEs. A TE is “low-frequency” if it is only identified in the focused strain, but not in the population (in other words, a private TE insertion). This information is now included in p23, line 19-21. Surveys on TE population frequencies in Drosophila found highly skewed frequency spectra and that the majority of the TEs appear as singletons (i.e., present in only one individual in the population sample); analysis typically categorizes TEs according to whether they are singleton (low-frequency in our analysis) or not (high-frequency in our analysis). Accordingly, we stick with this binning in our analysis.

3.1. TE position relative to genes seems oversimplified. The manuscript compares TEs that are close to or far from genes, but what happens to TEs in the genes (either in UTRs, CDSs or introns)? Do these different TE positions inside and outside genes exert different effects on K9 magnitude, K9 spreading extent, gene expression and/or allele frequency in the population?

(We split the reviewer’s comment #3 into two and answered them individually).

Following the reviewer’s suggestion, we compared the magnitude and extent of TE-mediated H3K9me2 enrichment between intergenic, intronic, and exonic TEs. While we don’t find any consistent significant differences for the magnitude of TE-mediated epigenetic effects, intergenic TEs exert a larger extent of H3K9me2 enrichment than other TEs in multiple genomes. These results are now included in p13, line 4-15 and Figure 2 – Supplementary Figure S4.

Our goal is to identify how TE-mediated epigenetic effects associate with genic epigenetic status, gene expression, and TE population frequencies. The difference in TE-mediated epigenetic effects between intergenic and genic TEs would have been captured by the estimated magnitude and extent of H3K9me2 enrichment. Accordingly, for the downstream analysis, we did not further distinguish between genic and intergenic TEs with one exception. Because TEs inside genes could influence gene expression through mechanisms other than altering the epigenetic states, a major confounding factor of previous studies (reviewed in (Choi and Lee 2020 PLoS Genetics, Kelleher et al. 2020 Trends in Genetics)), we excluded TEs inside genes when testing the associations between TE-mediated epigenetic effects and gene expression. We further emphasized this point in p15, line 15-18.

3.2. Are there any differences to TEs being 5' or 3' to the genes, since TEs 5' to the genes are more likely to affect the promoter? I worry that simple comparison of close vs. far might skew the actual relationships.

We appreciate the reviewer for pointing this out. Our additional analysis following the reviewer’s suggestions led to some new observations that further corroborate our conclusions, mainly whether neighboring gene expression could influence the extent of TE-mediated epigenetic effects.

We agree with the reviewer that TE-mediated epigenetic effects on the epigenetic states and expression of genes could differ when TEs are 5’ or 3’ to genes, especially given that TEs 5’ to genes should be more likely to influence the promoter. We performed correlation analyses between TE’s epigenetic effects and genic H3K9me2 enrichment or expression separately for TEs 5’ and 3’ to genes and reached similar conclusions to those of analyzing all TEs together (Figure 3B). Specifically, we found significant associations between TE’s epigenetic effects and genic H3K9me2 enrichment for both TEs 5’ and 3’ to genes, such associations depend on TE-gene distance, and we did not notice any major difference between TEs 5’ and 3’ to genes. On the other hand, we observed limited associations between TE-mediated epigenetic effects and gene expression, except for one incidence. These results are included in p16, line 23-p15, line 16 and Figure 3B.

We also tested whether the impact of adjacent gene expression on the extent of TE-mediated enrichment of H3K9me2 differs between TEs 5’ or 3’ to genes. We found that the difference in the extent of TE-mediated H3K9me2 enrichment is greater for TEs 5’ than 3’ to genes, likely due to the effects mentioned by the reviewer and providing stronger support for the hypothesized effects of gene expression on the extent of TE-mediated local H3K9me2 enrichment. We included these results in p20, line 5-10, Figure 3 – Supplementary Figure S12.

4. Transcriptional activity of TEs was not considered. Many TEs are (5'-) truncated and thus transcriptionally inactive. There is literature that suggests K9 installation and spreading only happen on transcriptionally active TEs. Have authors considered separating active/inactive or intact/truncated TEs? Combining TEs with different transcriptional activity (or K9 biology) despite the same length, family or position could confound the analysis and lead to different conclusions.

We agree with the reviewer that this is an important axis of the analysis that we did not consider. We categorized TEs whose length is shorter than 70% of the annotated, canonical sequence in D. melanogaster as truncated, and otherwise as full length. Consistent with what is suggested by the reviewer, full-length TEs tend to show a larger magnitude and extent of H3K9me2 enrichment. These results are now included in p11, line 18-26 and Figure 1 – Supplementary Figure S4. Because our following analysis focuses on how TE-mediated epigenetic effects, instead of individual TE attributes, influence genic H3K9me2

enrichment/expression and selection, we did not further consider full-length/truncated TEs for the downstream analysis.

5. Usage of 16-18hr whole-embryo averaged transcriptome and H3K9me2 epigenome require justification. The 16-18hr embryo is presumably consisted of a very heterogenous population of cells, how does such a population average reflect bona fide biology? Authors should either justify this or comment explicitly in text on potential problems such datasets could have on major conclusions.

We agree with the reviewer that 16-18hr embryos consist of heterogeneous cell types, which are precursors of different tissues and organs of a whole animal. By examining the epigenetic effects of TEs in whole embryos, our analysis would capture the averaged effects across cell types. If the relative importance of tissues/organs in determining individual fitness is proportional to the abundance of their specific cell precursors, our estimate could provide a reasonable estimate of the epigenetic effects of TEs that may be relevant to individual performance and thus fitness. Yet, if the fitness of an individual is predominantly determined by a tissue/organ that originated from a rare cell type and TE-mediated epigenetic effects vary substantially between cell types, our analysis of whole embryos may not be a good indication of the associated fitness consequence. Even though we did find evidence supporting selection against TE-mediated enrichment of repressive marks estimated from whole embryos, future investigation on the variability of such effects across tissues and cell types, which we aim to address in the future, will be important to further dissect the role of TE-mediated epigenetic effects in individual fitness and, thus, genomic TE abundance.

Following the reviewer’s suggestion, we emphasized that our dataset should be viewed as an average of all cell types and stated the potential caveats in p42, line 22-p43, lin2.

6. Euchromatic TE and genome-wide TE abundance should not be used interchangeably. This work is done solely on euchromatic TEs, but throughout the text, authors comment on how a factor examined might shape genome-wide TE abundance or even genome evolution. I wonder how much euchromatic TE copy number reflects genome-wide TE abundance? I'd expect heterochromatin harbors most TE copies and to be the major place that determines the overall TE abundance. Authors should either discuss how euchromatic TEs do reflect genome-wide TE abundance (and genome evolution) or change all related claims to better describe analysis done on euchromatic TEs.

We agree with the reviewer that this is a very important point that we did not specify previously. The genome-wide TE abundance we discussed in the text and also in the field mainly concerns euchromatic TEs. This is because of two major reasons. One is the technical difficulties associated with assembling and annotating TEs in the highly repetitive heterochromatic regions of the genome. The other reason is that, in the few species where heterochromatin has been assembled, most TEs in the heterochromatic regions are fragmented and have lost of transposition ability, making them no longer important players in the evolutionary dynamics of TEs (i.e., TE number increase through transposition and decrease through selection and excision). Following the reviewer's suggestions, we clearly defined that our study focuses on euchromatic TEs in the opening paragraph (p2, line 9-30), in the first paragraph of result (p7, line 21-22), and in the opening paragraph of discussion (p37, line 7-16).

We also emphasized that our study focused on euchromatic TEs and revised our claim at places throughout the manuscript (e.g., p6, line 8; p29, line 8, p37, line18).

7. Potential contradiction between key conclusion and the "species complex effect". In Figures5A and 5C, the species complex effect is obvious and, when we compare the two species complexes, one would conclude the opposite of what the authors attempted to argue. For instance, K9 and TE copy number correlate between mel and yak species complexes. Does this actually mean that there is different biology (or forces) at play at different evolutionary timescales? Authors' major claims seem valid within a shorter period of time (<3 million years) within each of the two species complexes, but they do not appear to hold when one looks at longer timescales. If so, the authors should modify key conclusions.

We appreciate the reviewer for pointing this out and agree with the reviewer’s interpretation. Indeed, according to our data, the importance of TE-mediated epigenetic effects in shaping genomic TE abundance mainly acts on a short evolutionary time scale.

Previous studies have found greater net gains of intron sequence and duplicated genes in D. yakuba than in D. melanogaster and/or D. simulans. These observations suggest that the D. yakuba genome may be more tolerant of gained sequence. If similar mechanisms apply to the other two species in the yakuba species complex, lineage-specific mechanisms that shape the evolution of the whole-genome may also play a vital role in determining genomic TE abundance, and might potentially override the influence of selection against TE-mediated epigenetic effects. We revised our conclusion (by emphasizing it supports association within species complex, p30, line 8-12 (in Results)) and included these discussions in and p41, line 16-29 (in Discussions).

8. Biological difference between H3K9me2 magnitude and spreading extent seems unclear. Roughly speaking, the beginning of the manuscript made the two seem similar, the middle part found an effect on the extent only, and the last part found an effect on the magnitude only.

While I appreciate the authors examining two aspects of H3K9me2 spreading, I am unsure what either aspect truly reflects. I worry that some readers would find the authors cherry-picking either aspect to support their claims – perhaps authors should've commented a priori the difference between the two aspects such that conclusions made on one, but not the other, aspect of TE-mediated epigenetic effects are more convincing.

We agree with the reviewer that we did not make clear distinctions between the magnitude and extent of H3K9me2 enrichment, which could confuse readers about why we would need both indexes.

Previous studies on the spreading of repressive marks from constitutive heterochromatin suggested that the magnitude and extent of H3K9me2 enrichment may not be perfectly associated, mainly because the local genomic context could influence the extent of the spread. If similar molecular mechanisms are also applicable to epigenetically silenced TEs in euchromatin, the magnitude and extent should provide different information about TE-mediated local enrichment of repressive marks. Before conducting this study, we had no a priori expectation for which index would be more relevant to the questions we aim to address. Accordingly, we feel it is most appropriate to present both of them. Especially, in several analyses where the results differ between the two indexes, they provide important information about the underlying mechanisms (see below).

According to our observations, the major difference between these two indexes is that the magnitude of TE-mediated H3K9me2 enrichment is more significantly associated with lower TE population frequencies (i.e., strong evidence supporting being selected against) and is correlated with genomic TE abundance. On the other hand, the impact of neighboring gene expression only influences the extent, but not the magnitude, of TE-mediated H3K9me2 enrichment. In the discussion, we interpreted these two findings together as the result of the extent being more sensitive to the local genomic context (which is similar to previous studies) than the magnitude, making the latter a more direct estimate for the functional and thus fitness impact (which we still don’t know, see p40, line 18-p41, line 2 for discussion) of TE-mediated H3K9me2 enrichment.

To make better distinctions between these two indexes, we (1) included definitions for what these two indexes are at where they were first mentioned in Results (instead of just in Materials and methods; p10, line 3-9), (2) discussed previous observations to explain why we used both indexes and provided our a priori guess that the extent may be more sensitive to local genomic context (p10, line 19-p11, line 8), (3) moved discussions regarding the difference and relative importance of these two indexes from the end of Discussion to the front of Discussion (p38, line 23-p39, line 7), and (4) emphasize their difference throughout the text when necessary. We hope these changes addressed the reviewer’s concerns.

9.1. Chromatin landscape was examined by Su(var) gene expression and satellite repeat abundance, which seems insufficient and incomplete. Chromatin landscape is too broad of a term when authors only examined Su(var) gene expression and simple repeat abundance. Have authors look at other heterochromatin genes (e.g., ZAD-ZNF, piRNA pathway) besides Su(var)?

(We split the reviewer’s comment #9 into two and answered them individually). Most of the piRNA pathway genes are expressed in the female germline and at the early embryonic stage. Accordingly, nearly none of them are expressed in our dataset, which was generated using 16-18hr late-stage embryos. In Drosophila, ZAD-ZNF proteins have a wide range of functions. For a ZAD-ZNF gene that has a demonstrated role in heterochromatin function (Oddjob, a Su(var)), we included it in our analysis.

It is worth noting that, instead of looking at the initiation of the epigenetic silencing of TEs, we mainly focus on the inadvertent side effect associated with the silencing – the enrichment of repressive marks at flanking non-TE sequences. This phenomenon is similar to the well-studied Position Effect Variegation (PEV). Accordingly, our analysis focuses on Su(var), whose role in determining the strength of PEV has been well characterized. To explain this point, we included more motivation for looking at Su(var) at p31, line 8-15. We agree with the reviewer and tone down several places that our study mainly concerns repressive chromatin landscape (e.g., p38, line 1) in addition to other existing places.

9.2 What is the relationship between simple repeat abundance, TE abundance, and genomewide repeat content? This latter point is related to the point #6 above about euchromatic TE vs. genome-wide TE. I kept wondering, in the context of this work, whether there is biological difference between satDNA and TE – are they similar because they cause similar epigenetic effects or do they differ?

As described in our response to reviewer’s comment #6, our study mainly concerns the evolutionary dynamics of TEs in the euchromatic genome. We hope changes made while addressing that comment clarify this point.

We are unaware of studies that report H3K9me2/3 enrichment mediated by satellites in the Drosophila euchromatic genomes, except for those extreme cases through experimental manipulation (e.g., a large tandem array of constructs). The replication/expansion mechanisms and potential fitness impacts also differ between TEs and satellite repeats. Accordingly, we would argue the evolutionary dynamics would differ between these two types of repetitive sequences.

Minor comments:

– Figure 2 beautifully shows the epigenetic effect depends on the TE family and other factors. I wonder if the authors would want to expand Figures3-5 to see if any TE family exerts a particularly strong (or weak) epigenetic effect in different species?

Because there are only a few TE families that have at least five copies included in the analysis for all genomes, our power to perform the suggested analysis is limited. Also, in Figures 3-5, our main goal is to identify the associations between the strength of TE-mediated epigenetic effects and various factors (e.g., gene expression, genomic TE abundance), irrespective of TE identity.

Due to these two reasons, we did not perform the analysis suggested by the reviewer.

Nevertheless, we hope the data presented in Figure 2C provides a glimpse into the reviewer’s question. We added some more discussions in the text regarding this p12, line 21-26, pointing to TE families that are of particularly strong/weak epigenetic effects across species.

– Figure 6 cartoon probably should have K9 spread less towards the gene side.

We agree with the reviewer and revised the figure accordingly.

Reviewer #1 (Significance (Required)):

The senior author has previously found that euchromatic TE exerts an epigenetic effect via H3K9me2 spreading, which confers a fitness cost. Now, this work builds on the previous finding to attempt a conceptual advance that such a TE-mediated epigenetic effect differs across six species examined and depends on species-specific chromatin landscape. The examination of six species and the connection between molecular biology, population genetics and potentially genome evolution are rare among existing literature and thus present novelty. It should interest people working on TE, heterochromatin, epigenetics and population genetics. My expertise is in chromatin, epigenetics and RNAi. I do not have sufficient expertise to evaluate the regression analysis and TE population frequency estimation done in this work.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

Summary: In this article, Huang and colleagues investigate the degree to which TEs can induce heterochromatin formation within the euchromatin across species. They hypothesize that different TE abundances across species might be explained by variation in the degree to which TEs impact a selective effects. In this article, they show that there does seem to be a big difference in the magnitude of H3K9me2 induction between the mel clade and the yak clade. This is extremely interesting and very worthy of publication. The authors do a great job exploring a variety of alternate hypotheses.

Major Comments:

I have no major comments that may significantly alter the conclusions. I found their approach robust. However, I will admit there is one thing I find difficult to understand regarding their model and conclusion. In the introduction they state:

"These observations spurred our previous hypothesis that varying epigenetic effects of TEs could result in between-species differences in the strength of selection against TEs, eventually contributing to divergent genomic TE abundance (Lee and Karpen 2017)."

And in the final figure (figure 6) they provide a schematic of this hypothesis (note typo in the worded "mediated" in the figure). In particular, they show (indicating the * where they state the paper found) that (1) The magnitude of Heterochromatin induction can lead to (2) lower TE population frequencies and this can lead to (3) lower genomic TE abundance. But doesn't 5A show that the yak clade has HIGHER magnitude of heterochromatin induction (as shown earlier) but also higher estimated TE copy number? So, 5A shows the opposite conclusion with respect to genomic copy number. I feel the disconnect might arise from mixing up the notions of low frequency of TE insertions from the notion of genomic copy number. It is possible to have a VERY large number of TEs that are segregating at very low numbers. So, selection against TEs (as evident by low allele frequency) doesn't easily translate to total copy number. I hope this makes sense.

We appreciate the reviewer for pointing this out and this is indeed something we should address. Part of the issue raised by the reviewer is hopefully addressed in our response to reviewer #1’s comment #7 (please see above). In short, the strong species complex effect suggests that the role of TE-mediated epigenetic effects in determining genomic TE abundance may be acting on a short evolutionary time scale and could be overridden by other general processes that determine gain/loss of sequences (irrespective of types). This may explain why TEs in yakuba complex species show stronger epigenetic effects but are also more abundant.

We agree with the reviewer that our model (presented in Figure 6) has unstated underlying assumptions. According to the population genetic theory of TE copy number, the equilibrium copy number is determined by the strength of selection removing TEs and TE increase through transposition rate. Our model assumes that the transposition rates are similar, and TE copy numbers are at equilibrium between species. Here, the population frequencies of TEs serve as an indicator for the strength of selection removing TEs. It is possible to have a large number of TEs segregating at low population frequencies, but the average copy number across individuals of a species (i.e., the number of TEs in each genome) would still be small. We hope this interpretation makes sense to the reviewer and are open to suggestions to further revise the model. We included additional discussions regarding the assumptions of our model, which can be found at p41, line 7-10 and in Figure 6 legend.

Minor Comments: I have some minor comments for clarity.

1) Since the paper is motivated to explain differences between species in genomic copy number, this information should be provided in the first figure in 1A (alongside the species name, with the two sim and yak samples also both provided numerically)

We revised Figure 1A according to the reviewer’s suggestion.

2) Something should be made of the fact that some of these arguments can be circular and the arrow of causality is hard to tease apart. Genomic TE abundance may be explained by species level differences in the epigenetic impact of TEs. But, it may also be the case that as TE abundance varies, there may be evolution of the epigenetic impact. Causality is tricky here and worth being up front about.

We agree with the reviewer, and we included this possibility in Discussion. In this revised manuscript, we rewrote the discussion to further emphasize this point (p42, line 1-8).

3) Some clarity in the text (not just in the methods) should be provided for the reader about how the "extent of spreading" was calculated.

We agree with the reviewer and added these details in the Results section when this index was first mentioned (p10, line 3-9).

4) In figure S1, what does it mean for points to have zero magnitude of enrichment but a great extent of enrichment? Shouldn't all points for zero magnitude be at the origin? I am looking at the scattered points that run along the Y axis in S1 of just the mel group (not seen in yak group)

We greatly appreciate the reviewer for pointing this out.

We identified an error in the coding for generating this plot. In this script, a few “NA”s in the data file were treated as 0 when calculating the average magnitude of the two replicates. This error affected the results of Figure 1- Supplementary Figure S1, Figure 2B and 2C, Figure 2 – Supplementary Figures S1 & S2. We updated these figures and revised the correlation coefficients in Figure 1 – Supplementary Figure S1. After removing these TEs with erroneous estimates, the correlation between the magnitude and extent of TE-mediated epigenetic effects becomes stronger. Other updated results reached the same conclusions as previous results.

We thoroughly checked our coding, and all the other results throughout the study are unaffected. For other analyses, we used a different script in R, which removes the “NA”s.

While addressing this comment, we noticed that we forgot to provide details about why a TE may have NA estimates for TE-mediated epigenetic effects (low local sequencing coverage) and how we treated them in the analysis (remove them from the analysis). We included these details in p48, line 20-21; p48, line 31.

5. In S2, how they compare the correlations across two genomes is confusing. Is that for shared TEs? Also, do estimates from two strains (Y axis) include values from X axis? It seems like this is saying X is correlated with the average of X and Y, which will always be true if Y is a random variable. Overall, I was a fairly confused about what S2 was showing.

We agreed with the reviewer that neither the main text nor the legend for Figure1 –

Supplementary Figure 1 (previous Figure1 – Supplementary Figure 2) contains sufficient details for how these two estimates were generated.

The concepts for estimating the magnitude and extent of TE-mediated epigenetic effects are similar for using one genome and two genomes. The major difference is the baseline used. For one genome, we used H3K9me2 enrichment > 1 as a criterion to estimate the extent. In contrast, for two genomes, the comparison is made between homologous sequences of the focal genome with the focused TE insertion to the alternative genome without the TE insertion (i.e., whether H3K9me2 enrichment in strain 1 > H3K9me2 enrichment in strain 2). For the magnitude of TE-mediated H3K9me2 enrichment, the two-genome estimates were further normalized by the enrichment level in the homologous sequence without the TE insertion. We agreed with the reviewer that these two estimates are related due to shared variables, but the relationship between them is not a simple arithmetic calculation. Accordingly, we feel that presenting this correlation result is still important.

We included some more details in the main text (p10, line 9-13) and Figure 1 – Supplementary Figure S1 legend that hopefully could help readers interpret the result. We also noted that detailed methods can be found in Materials and methods.

When revising according to this comment, we noticed that we did not include method details about how we identify homologous regions in the two PacBio genome assemblies. We added these details (using minimap) in p50, line 22-25.

6) In figure 2A. Do not includes lines connecting values between species. Just used colored dots This is not a series. I think this issue is elsewhere as well.

We appreciate the reviewer for pointing out our oversight and revised Figures 2A and 2C accordingly.

Reviewer #2 (Significance (Required)):

This article provides some extremely important insight into the manner in which heterochromatin induction can shape TE dynamics across species. What explains variation in genome size and TE content is one of the big questions in evolutionary genomics. The results are somewhat complex, but that is because biological systems are complex.

My expertise is in evolutionary genomics of TEs. I am not an expert in chromatin biology and regulation of gene expression.

Reviewer #3 (Evidence, reproducibility and clarity (Required)):

In this manuscript, Huang et al. expand upon results presented in two previous influential publications led by the corresponding author of this manuscript: Lee (2015) and Lee and Karpen (2017). Lee (2015) reported that H3K9me3 spreads from euchromatic TE insertions, these insertions lead to reduced expression of nearby genes, and this phenomenon depends on the piRNA pathway. Lee and Karpen (2017) reported a significant negative correlation between both H3K9me2 magnitude and extent and TE population frequencies in D. melanogaster and demonstrated strong spreading of H3K9me2 from TE insertions, an average of 4.5 kb into flanking regions. They also report stronger epigenetic effects of TEs in D. simulans compared to D. melanogaster which they propose could be due to higher expression of Su(var) 3-9.

Here, Huang et al. examine H3K9me2 magnitude and extent (i.e. spreading) across six species of Drosophila, including two different strains of D. simulans and D. yakuba. Across all species and strains, the authors observe a strong enrichment of H3K9me2 at euchromatic TE insertions. They find a large amount of variation in H3K9me2 magnitude and extent across TE classes and families, and among species and strains, but this variation does not appear to show any meaningful and/or consistent pattern. While they find that genic H3K9me2 is positively associated with the magnitude and extent of TE-derived H3K9me2, only the magnitude effect depends on gene distance. They find no relationship between TE epigenetic effects and expression of adjacent genes and some evidence of asymmetric H3K9me2 spreading due to neighboring highly expressed genes. Across all species, the magnitude of H3K9me2, but not the extent, was negatively correlated with TE population frequency. Similarly, the magnitude of H3K9me2, but not the extent, was correlated with TE abundance across species. Finally, they find that the magnitude of H3K9me2 (extent not tested) is positively correlated with expression of Su(var) genes across species.

The analyses performed here are rigorous and the appropriate replicates have been included. The authors also clearly go to great lengths to try to address confounding factors that could explain the correlations they are seeing.

1. My main concern related to evidence and reproducibility is that the results shown for H3K9me2 magnitude and spreading are highly variable between species and strains. It is not clear to me how much of this variation is due to biology versus technical issues related to ChIP, however I am surprised at how low the correlations among replicates are for these measures (Figure 1, Supplementary Figure S4). I would encourage the authors to perform additional QC measures for their ChIP experiments and replace any low quality replicates so as to minimize the contribution of technical variation. Some QC suggestions: the K-MetStat Panel spike-in should allow for an assessment of antibody specificity for the K9me2 modification and the Irreproducibility Discovery Rate (IDR) framework can quantify reproducibility among replicates.

We appreciate the reviewer’s helpful comment and performed analyses following the reviewer’s suggestion. We used ChIP-seq reads mapping to SNAP-ChIP K-MetStat Panel to assay the specificity of the used antibody (Abcam 1220). Consistent with previous investigations by modEncode (Egelhofer et al. 2011) and others (Sentmanat and Elgin 2012), our analysis found that this antibody shows high specificity against H3K9me2. We included this result in p45, line 23-25 and Figure 1 – Supplementary Figure S5.

We performed the suggested IDR analysis and found no direct associations between IDR and the correlations between replicates for the index of TE-mediated epigenetic effects (the diagnostic IDR plots are now in Figure 1 – Supplementary Figure S7). For instance, D. melanogaster shows the lowest correlation for the magnitude of TE-mediated H3K9me2 enrichment between replicates, but has decent consistency between replicates by IDR analysis.

A plausible technical reason for this discrepancy is that IDR analysis relies on called peaks. Custom pipelines, which were designed to identify “sharp peaks,” are likely to miss the enrichment of repressive marks, which typically have broad and lower level of enrichment (reviewed in Park 2009 Nature Review Genetics). Accordingly, TE-mediated H3K9me2 enrichment in euchromatic genome could be missed in the IDR analysis. Indeed, a previous study found that only 12% of TE-induced H3K9me2 enrichment was detected by peak calling (Lee et al. 2020 PLoS Genetics). Another more important biological reason is that the spreading of repressive marks from constitutive heterochromatin has been observed to be a variegating phenotype and could differ between cells and individuals. Accordingly, the variability for the magnitude and extent of TE-mediated local enrichment of heterochromatic marks could have contributed to the low correlation between replicates for some samples. We included these discussions in p49, line 10-29.

Thanks to the reviewer’s comment, we noticed that D. simulans strain 2 has much fewer peaks showing low IDR between replicates. To be conservative, we thus tried excluding this sample for the key analysis for our study and still found negative associations between the magnitude of TE-mediated H3K9me2 and genomic TE abundance. We included this analysis in p29, line 2224; p29, line 28-29; p49, line 29-p50, line 3.

2.1. Related to the comment above: does the "extent" measurement used here differ from the "spreading" measurement used in Lee (2015) and Lee and Karpen (2017). If so, why was a new metric used?

(We split the reviewer’s comment #2 into two and answered them individually).

We are a bit unsure about the “spreading” measurement the reviewer is referring to. In Lee (2015), we did not estimate TE-mediated local enrichment of H3K9me3 and instead investigated the associations between genic H3K9me3 enrichment and the presence of adjacent TEs. In Lee and Karpen (2017), we used the same measurement as the current study – the extent and magnitude of TE-mediated local enrichment of H3K9me2.

We agreed with the reviewer that we did not include enough descriptions of the two indexes for the epigenetic effects of individual TE and this can cause significant confusion. To address this, we added text in p10, line 3-9.

2.2. Could that explain why spreading no longer seems to be an important contributor to TE epigenetic effects?

We agree with the reviewer that we did not clearly define “epigenetic effects of TEs” and, without that, the term may implicitly imply functional consequences (e.g., gene expression). We used the term “epigenetic effects” to describe TE-mediated local enrichment of H3K9me2. We added definition for this term when first mentioning this term in Introduction (p5, line 28) and Results (p10, line 3-5).

The spreading of H3K9me2 beyond TE boundaries could lead to local enrichment of H3K9me2. The strength of such an event could be described by the enrichment level (magnitude) and extent (extent) of the local enrichment of H3K9me2 outside TE sequence. We see that the word “spreading” might also be interpreted as the “extent” of the H3K9me2 enrichment. Accordingly, we replaced this word throughout the manuscript with a more accurate description (“local enrichment of H3K9me2”) to avoid confusion and misinterpretation (e.g., p1, line 23; p4, line 15; p5, line 1; p38, line 10).

According to Figure 1B and C, we observed that the presence of TEs not only lead to H3K9me2 enrichment at immediate TE-flanking regions, but also could extend for several kbs, suggesting that the “spreading” (used in reviewer’s context) of repressive marks from TEs is still an important “epigenetic effects.” We also observed that the extent of TE-mediated enrichment of H3K9me2 is positively correlated with genic H3K9me2 enrichment (Figure 3A). It is when we investigated the role of TE-mediated local enrichment of H3K9me2 that we did not find the predicted associations between the extent of such effect and genomic TE abundance, but did find significant associations between the magnitude of such effect and TE abundance. It is worth noting that Lee and Karpen 2017 made similar observations with two species (i.e., TE copy number is associated with the magnitude of epigenetic effects, but not with the extent), which we noted in p30, line 1-3. In addressing the reviewer #1’s comment 8, we also made more clear distinctions between the two estimates (magnitude and extent of TE-mediated H3K9me2 enrichment; see above).

3. In terms of clarity: if the authors demonstrate that their ChIP replicates are highly reproducible and the "extent" metric is not meaningfully different from the previously used "spreading" metric, then it is hard for me to think of technical or analytical reasons why the findings reported here differ substantially from those reported previously. The previous model was intuitive: an underappreciated aspect of the deleterious nature of TE insertions is that euchromatic TEs can initiate heterochromatin formation which spreads into nearby genes leading to harmful reductions in their expression. This study suggests that magnitude of H3K9me2 is the major contributor to the "epigenetic effect" of TEs and that there is no relationship between TE-induced H3K9me2 and neighboring gene expression. I would encourage the authors to use the discussion to acknowledge these contradictions and try to reconcile these results with those from the previous two papers. Is there reason to believe the previous studies were incorrect? Otherwise, it is very hard to know what to take away from this study.

We agreed with the reviewer that without clearly defining “epigenetic effects,” readers could get the impression that the current study contradicts previous findings. Hopefully, our revision according to the reviewer’s comment #2 above would clarify the meaning of “epigenetic effects of TEs.” Also, as noted above, similar to Lee and Karpen (2017), we used two indexes to describe the epigenetic effects of TEs, or TE-mediated local enrichment of H3K9me2 in the euchromatic genome. Both the previous study with a limited number of species and our study found that (1) TEs lead to a significant increase in the magnitude and extent of local enrichment of H3K9me2, and (2) TE copy number is negatively associated with the magnitude, but not extent, of the effect. Our findings do not contradict previous studies.

The generally limited associations between TE-mediated epigenetic effects and lowered gene expression, as mentioned by the reviewer, was also reported in other model species, including in D. melanogaster (Lee and Karpen 2017 eLife), Arabidopsis thaliana (Quadrana et al. 2016 eLife; Stuart et al. 2016 eLife), rice (Choi and Purugganan 2018 MBE), and mouse (Pezic et al. 2014 Genes & Development). This has been a perplexing finding in the field of TE evolution, which was extensively discussed in two recent reviews (Kelleher et al. 2020 Trends in Genetics; Choi & Lee 2020, PLoS Genetics; discussed in our Introduction p5, line 9-16). Our current study investigated this question in multiple species and used both within-genome and across-genome (i.e. between homologous alleles) comparisons, further revealing the complex relationship between TE-mediated epigenetic effects and neighboring gene expression. We hope that our new analyses addressing reviewers' comments (reviewer #1’s comment 3 and reviewer #3’s comment 4) further help corroborate our findings. We also included additional discussions regarding the limited associations between TE-mediated epigenetic effects and gene expression (p39, line 9-22) and provided possible reasons why we did not find the expected negative associations between them (p39, line 22 – p40, line 16).

It is worth noting that our analysis focuses on identifying the genome-wide average pattern; it is still plausible that the local enrichment of H3K9me2 induced by individual TE occasionally lowers nearby gene expression (e.g., genes with positive z-scores in Figure 3C). In fact, many early examples of TE-mediated enrichment of repressive marks were discovered by the phenotypic consequences of reduced neighboring gene expression (reviewed in (Choi and Lee 2020 PLoS Genetics)). We included this discussion in p18, line 14-21.

In this study, we tested the model proposed by Lee and Karpen (2017) (Figure 6). We were able to find support for most of the models except for the generally expected, but unsupported prediction that TEs’ epigenetic effect would lower neighboring gene expression. We proposed several possibilities for this missing link, which can be found in p40, line 18 – p41, line 2.

Other major comments:

4. Figure 3C: For the gene expression analysis, it would be helpful to see a scatterplot where each point is a gene and the axes are the log2 expression level of the gene in the strain with an adjacent TE versus the log2 expression of the same gene in the strain without an adjacent TE. The slope of the regression line would show whether there is any tendency for genes with adjacent TEs to have lower expression.

We performed the suggested analysis following the reviewer’s suggestion (p18, line 9-11 and Figure 3 – Supplementary Figure 9). In these plots, we have the log2 expression level of genes in the strain with a TE on the X-axis and in the strain without TE on the Y-axis. We plotted genes whose nearest TEs with and without epigenetic effects with different colors. The slope is close to 1 (i.e., no difference in gene expression with and without TE presence), and there is no difference in slope for genes whose nearest TEs with and without epigenetic effects. These observations are consistent with our z-score analysis, finding a limited influence of TE-mediated epigenetic effects on gene expression.

5. It is interesting that there is clear evidence supporting enrichment of H3K9me2 at TEs leading to genic H3K9me2 enrichment but not expression downregulation. It would be very interesting if the authors could extend this analysis to include all genes from the euchromatic chromosome arms showing H3K9me2 enrichment in at least one strain or species, irrespective of TEs. Overall, are differences in genic H3K9me2 enrichment between strains/species associated with differences in gene expression?

This is a very interesting question raised by the reviewer. Yet, addressing this question is beyond the scope of this study. The suggested analysis would address the general role of the enrichment of H3K9me2 in gene expression, which is a broad topic.

It may be worth noting that, according to a previous analysis (Lee et al. 2020 PLoS Genetics), most, if not all, of the polymorphic H3K9me2 enrichment in the euchromatic genome is due to the presence/absence of TEs within species. Few H3K9me2 enriched regions in the euchromatic genome will thus be TE-free and could be suitable for performing the suggested analysis. On the other hand, the between-species comparison of gene expression is likely to be confounded by factors other than the enrichment of H3K9me2 (e.g., the evolution of regulatory sequences) and thus be challenging to interpret. We thus did not proceed with the suggested analysis.

6. I'm confused by the difference between Figure 3D and Figure 3 – Supplementary Figure S7. If I understand correctly, for D. melanogaster, there is no difference in the extent of spreading between the two sides of a TE when the highly expressed gene is near the TE, nor is there a difference when the highly expressed gene is far from the TE (Figure 3D). However, when expression measurements are used from a different strain (Figure 3 – S7), there is a clear difference in the extent of spreading on the gene side, specifically when the highly expressed gene is close to the TE. What explains these contradictory results? Are a large proportion of the genes identified as highly expressed when using the other strain's data no longer highly expressed in the strain of interest? (presumably due to the effect of the nearby TE?). If so, that seems notable as it lends support to an association between spreading and gene downregulation.

We appreciate the reviewer for pointing out this incongruence. Following the reviewer’s suggestion, we counted the number of genes that have a TE nearby and are categorized as highly expressed-

In both our data and modEncode: 179

Only in the modEncode data: 93

Only in our data: 83

There is no significant difference in the number of genes that are categorized as highly expressed in only modEncode data or only in our data. Accordingly, the suggested possibility (the difference between Figure 3D and previous Figure 3 – Supplementary Figure S7 is due to TE’s epigenetic effects lowering gene expression) may not be applicable to our observations.

The main reason why the result in Figure 3D differs from those of previous Figure 3 –

Supplementary Figure S7 is probably due to the fact that we used different statistics. In Figure 3D, we estimated the z-score for the difference in the extent of TE-mediated H3K9me2 enrichment between the genic and intergenic sides, normalized by the standard deviation of the estimates. In previous Figure 3 – Supplementary Figure S7, we simply compared the extent of TE-mediated H3Kme2 enrichment on the two sides of a TE. The z-score should be a more appropriate statistic given that it is normalized, so comparable across TE insertions. By using the z-score to perform the analysis in previous Figure 3 – Supplementary Figure S7, we found the same results like those in Figure 3D, and there is no statistically significant difference in the extent of TE-mediated H3K9me2 enrichment between the two sides of a TE in D. melanogaster.

Author response image 1

Download asset Open asset

Given that there is no major difference between Figure 3D and updated Figure 3 – Supplementary Figure S7, and the latter does not add additional information, we decided to exclude analysis based on categorizing genes according to modEncode data (i.e., previous Figure 3 – Supplementary Figure S7).

7. The authors report that H3K9me2 spreading from TEs is counteracted by the presence of a highly expressed gene nearby (either upstream or downstream), leading to asymmetric spreading. However, the spreading extent index is calculated by averaging the extent of spreading between the two flanking regions of the TE. Is this the best choice given the presence of asymmetric spreading? Would using the longer of the two distances be better?

We think the reviewer is suggesting we use the “intergenic side,” which is less likely influenced by the expression of adjacent genes. Our response is based on this interpretation.

According to our data presented in Figure 3D, the extent of TE-mediated H3K9me2 enrichment is not symmetrical even for TEs far from genes. This is consistent with findings from constitutive heterochromatin that the spreading of repressive marks is a stochastic process. Also, our goal is to understand the potential functional and thus fitness consequences of TE-mediated H3K9me2 enrichment; such effect on the genic side may thus be of higher importance. Accordingly, it is challenging to predict a priori which side of the extent of H3K9m2 enrichment is more important, and we still chose to use the average of the two sides for our analysis.

8. An important prediction related to the proposed relationship between differences among species in expression of Su(Var)s and differences in the magnitude of H3K9me2 at TEs is that the same pattern should be present for other heterochromatic loci. Do other (non-TE) heterochromatic loci show stronger H3K9me2 enrichment in species with higher expression of Su(Var)s?

We agree with the reviewer that this is an interesting question. Yet, the suggested analysis and interpretation of the results are outside the scope of the current study. In addition, the effects of reduced dosage effects of Su(var)s were mainly observed at the boundaries of heterochromatin (e.g., at translocated genes located at the heterochromatin/euchromatin boundary and associated with PEV phenotype, reviewed in Elgin and Reuter 2013). In order to perform such analysis, one needs to first define the heterochromatin (HC)/euchromatin (EU) boundaries epigenetically using H3K9me2 enrichment. A potential issue is that the enrichment level of H3K9me2 at genes at HC/EU boundaries would highly depend on how the boundaries were first defined by H3K9me2, making the interpretation of the analysis challenging. Because of these reasons, we did not proceed with the suggested analysis.

9. It is not clear to me how the 1360 reporter experiment is applicable to the Su(Var)s expression results. The authors only consider H3K9me2 magnitude in their comparison to Su(Var)s expression rank yet, if I understand correctly, the 1360 reporter experiment measures spreading/extent of H3K9me2 which leads to downregulation of the adjacent mini-white gene. The reporter experiment described here suggests that reducing Su(Var) gene dosage leads to a reduction of H3K9me2 spreading along with a concomitant reduction in reporter gene silencing, however the other results reported here suggest that spreading of H3K9me2 is not an important contributor to the epigenetic effects of TE nor does spreading lead to gene downregulation.

We agree with the reviewer that our motivation for performing the 1360 reporter experiment was unclear and needs to be further clarified. As discussed in response to the reviewer’s comments 2 & 3, we did find that euchromatic TEs lead to local enrichment of H3K9me2, which is characterized by higher H3K9me2 enrichment level (magnitude) and an extent beyond the TE sequence. Even though on a genome-wide average, we did not find predominantly negative impacts of TE’s epigenetic effects on neighboring gene expression, our analysis does not preclude individual cases where a gene’s expression is reduced by TE-mediated enrichment of H3K9me2 (see discussions in comment 3 above and p18, line 14-21).

The 1360 reporter consists of a TE right next to the reporter gene. Accordingly, for this reporter system, the magnitude, instead of the extent, of TE-mediated H3K9me2 enrichment may be more important. In this specific reporter system, there are associations between the presence of 1360, the enrichment level of H3K9me2 at the mini-white reporter gene, and the reduced amounts of red-eye pigmentation (likely due to reduced gene expression). Accordingly, the eye pigmentation level in this system serves as a convenient readout for the magnitude of TE-mediated H3K9me2 enrichment. We clarified these points in p33, line 23-29.

Minor Comments

1. Lee (2015) examined H3K9me3 while Lee and Karpen (2017) and this manuscript study H3K9me2 levels. I think it is important to explain why one modification was used versus the other. Do you expect them to be redundant or might they play different roles in spreading versus magnitude of heterochromatin?

We agree with the reviewer that we did not have enough context for why we chose to study H3K9me2. Both H3K9me2 and H3K9me3 are found to be highly enriched in the heterochromatic regions of the genome in D. melanogaster (Riddle et al. 2011 Genome Research; Kharchenko et al. 2011 Nature). According to Lee and Karpen (2017), both H3K9me2 and H3K9me3 captured TE-mediated local enrichment of repressive epigenetic marks. The major reason that our study focused on H3K9me2 is due to the availability of suitable antibodies. Antibody against H3K9me2 is monoclonal and has been validated by modEncode to show high consistency (Egelhofer et al. 2011), while commonly used H3K9me3 ChIP antibody available at the time when this study started (Abcam ab8898) is polyclonal and likely cross-react with other histone modifications, in particular H3K27me3 (personal communications with modEncode team). Because H3K9me2 and H3K9me3 would both provide the needed information about TE-mediated local enrichment of heterochromatic marks, we proceeded with only H3K9me2. To provide more context, we added text at p7, line 6-12.

2. For the TE population frequency analysis, I was not able to find the cutoffs for what were considered high frequency and low frequency TEs

This is an error at our end; we forgot to include details about the categorization of TEs based on population frequencies. High-frequency TEs are those that are found in both the studied strain and the population sample, while low-frequency TEs are those that are unique to the studied strain and absent in the population sample. We included this information p23, line 19-21.

3. The Su(Var) section is a bit confusing to interpret because gene rank should correlate negatively with H3K9me2 but expression level should correlate positively and thus, both positive and negative correlations are referenced in this section when describing the same pattern.

Could you switch the gene rank order (i.e. lowest to highest) so that the expected correlation is in the same direction regardless of whether you are referring to expression level or expression rank?

In order to be consistent, we used expression rank for all expression analyses presented in Figure 3 (TE’s impact on gene expression) and Figure 5 (Su(var) expression). To help readers interpret the results, we added additional texts to either define expression rank (“rank from the highest expressed genes”) or interpret the result (“lower expression rank (i.e., higher expression)”, e.g., p16, line 7; p17, line 26). We hope these could help readers and avoid confusion.

4. Is there a relationship between local recombination rate and H3K9me2 extent and/or magnitude? If so, that could also be a potential confounder with respect to TE population frequency. I believe this was addressed in Lee and Karpen (2017) but it would be useful to reiterate here.

We repeated a similar analysis as Lee and Karpen (2017) in D. melanogaster, the only species in which a high-resolution recombination map is publicly available. Consistently, we found no associations between TE-mediated epigenetic effects and local recombination rate, further corroborating the previous analysis. We included these results in p24, line 17-29.

Reviewer #3 (Significance (Required)):

This study builds upon two previous studies that have made a large impact on the field. This work is significant because it suggests that the epigenetic effects of euchromatic TE insertions are more complicated than previously suggested by work that was only conducted in D.

melanogaster. It also suggests that the mechanisms by which TEs exert these effects and the mechanisms that make these effects harmful need to be revised and studied further.

This work will be of general interest to those whose work involved evolutionary biology, population genetics, epigenetics and chromatin, and transposable elements.

My expertise is in the evolutionary and functional genomics of Drosophila.lex and 3 in yakuba species comple

Author details

1. Yuheng Huang

   Department of Ecology and Evolutionary Biology, University of California, Irvine, Irvine, United States

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Methodology, Writing - original draft, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1586-953X

2. Harsh Shukla

   Department of Ecology and Evolutionary Biology, University of California, Irvine, Irvine, United States

   Contribution

   Software, Writing - review and editing

   Competing interests

   No competing interests declared

3. Yuh Chwen G Lee

   Department of Ecology and Evolutionary Biology, University of California, Irvine, Irvine, United States

   Contribution

   Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing

   For correspondence

   grylee@uci.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0081-7892

• 3,974

  Page views

• 641

  Downloads

• 6

  Citations

Article citation count generated by polling the highest count across the following sources: PubMed Central, Crossref, Scopus.