The role of centromeric transposable elements in shaping chromosome evolution

The role of centromeric transposable elements in shaping chromosome evolution | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article The role of centromeric transposable elements in shaping chromosome evolution Manuel Poretti, Terezie Mandáková, Rimjhim Choudhury, Martin Lysak, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5461468/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Transposable elements (TEs) play pivotal roles in genome evolution, yet their impact on pericentromeric regions of chromosomes, characterized by high sequence turnover and TE abundance, remains largely unclear. This gap in knowledge limits our understanding of TEs biology and their role within host genomes. In this study, we address this gap by analysing chromosome-scale assemblies to explore the content and dynamics of pericentromeric regions in four closely related Biscutella species. Although they share substantial synteny, we observe significant variability in the non-coding genome, especially within pericentromeric regions of the species affected by strongest genetic drift due to smallest population size. By comparing TEs from the CRM clade, which specifically target centromeric regions, with those from the Athila clade, we identify specialized CRMs that follow centromeres after recent repositioning, alongside an invasion by Athila copies that exhibit less insertion bias. Additionally, we find that TEs migration from pericentromeric towards distal nucleolus organizer regions correlates with increased DNA methylation and decreased gene expression. These results highlight how rapid pericentromeric evolution driven by bursts of TE activity can significantly impact genome functionality and stability. Our findings offer new insights into the evolutionary mechanisms shaping genome organization and underscore the broader implications for understanding genome dynamics and adaptation. Biological sciences/Genetics/Genomics Biological sciences/Plant sciences/Plant genetics Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Transposable elements (TEs) are major drivers of genome evolution, promoting high sequence turnover 1 , 2 . The ability of these interspersed repeats to mobilize throughout the genome is an important source of genetic diversity and has profound direct and indirect consequences on genome size and chromosome number 3 , 4 . Recent evidence in Arabidopsis thaliana 5 and rice 6 demonstrates a substantial impact of TEs on chromosomal rearrangements, genome divergence, and speciation. The interactions between TEs and their host genomes however remain poorly understood, and their influence on the evolution of functional genes is yet to be fully quantified 7 . TEs can generate high-impact mutations when inserting into coding sequences 8 , significantly affecting functional genes. Additionally, TEs are thought to regulate the expression of nearby genes through the spread of DNA methylation 9 , 10 , although the impact of TE-induced methylation on gene expression in TE-rich non-model genomes is still debated 11 , 12 . Purifying selection at the genomic level is expected to efficiently remove TE insertions from gene-rich chromosome arms with high recombination rates. In contrast, TEs and arrays of tandem repeats tend to accumulate in regions of low recombination, such as centromeres and flanking pericentromeres, though empirical evidence supporting this model is limited. Furthermore, certain TEs have been reported to target centromeric chromatin through a specific chromodomain (CHDCR) 13 . In crucifer species (Brassicaceae), CRM elements, which present this chromodomain, show a biased insertion toward centromeric regions, whereas Athila elements lack this domain 14 . Similarly, in Brassica rapa , Ale elements were also observed to be enriched in the centromeres 15 . Comparative analysis of these TEs is expected to shed light on the drivers of their genome-wide distribution. Centromeres are crucial for chromosome segregation, yet their evolution remain poorly understood. Despite their conserved role, centromeric sequences evolve rapidly, a phenomenon known as the centromere paradox 16 . Recent technological advances have enabled the assembly of telomere-to-telomere genomes in non-model species, providing new opportunities to investigate enigmatic chromosomal regions such as pericentromeres 17 , 18 . In this study, we present chromosome-scale assemblies of four closely related Biscutella species (Buckler mustards; Brassicaceae), which diverged within the last 2.5 million years following a reduction in chromosome number and genome size after an 11.5-million-year-old whole-genome duplication event 19 , 20 . We characterize the gene space and repetitive fraction of these Biscutella genomes and assess the impact of centromeric and pericentromeric TEs on chromosome structure and genome content. Results Chromosome-scale genome assembly and annotation of four Biscutella species We combined Oxford Nanopore (ONT; 60x coverage), Illumina (40x coverage), and Hi-C (80x coverage) sequencing technologies to generate high-quality chromosome-scale genome assemblies of four Biscutella species (n = 9). These included the early diverging B. frutescens and three closely related taxa within the B. laevigata species complex (thereafter, B. austriaca , B. prealpina and B. varia 21 ). These diploid-like genomes underwent the alpha whole-genome duplication (α-WGD) event, which is shared by all Brassicaceae around 32 million years myr ago 22 , followed by a subsequent mesopolyploid WGD (Bl-m-WGD) around 11.5 myr ago 19 , 20 . In contrast, the related Megadenia pygmaea ( n = 6) did not experience the Biscutella -specific Bl-m-WGD 23 , 24 . All genomes were assembled into nine main scaffolds, spanning 710 to 1,001 Mb, consistent with genome size estimates, and presented BUSCO completeness scores ranging from 92.0–99.1% (Supplementary Tables S1-3). Synteny analysis against A. thaliana through the mapping of 22 Brassicaceae-specific genomic blocks 25 (GBs) onto our genomes showed high collinearity with the Ancestral Crucifer Karyotype 26 (Fig. 1 a; Supplementary Fig. S1 ). Given that GBs are restricted to chromosome arms, we utilized the extremities of the syntenic GBs flanking the centres of scaffolds to delineate the boundaries of pericentromeric regions (Fig. 1 a, Supplementary Table S4). For clarity and consistency, “pericentromere” is hence defined as the entire chromosomal region extending from the centromere to the beginning of the chromosome arm. Comparison with the comparative cytogenetic map of B. varia validates the structure of our assemblies and identifies six previously missing GBs 19 , along with characterizing the assembly of highly repetitive regions, such as pericentromeres and nucleolus organizer regions (NORs)-bearing short arms of chromosomes 2 and 3. The identification of telomeric repeats in the distal regions of most scaffolds further confirmed the near-complete assemblies, with large scaffolds considered equivalent to chromosomes. Despite small-scale inversions and translocations within GBs (Supplementary Fig. S2), high chromosomal collinearity was observed among taxa within the B. laevigata species complex, which diverged approximately 0.1 myr ago 21 . In contrast, large-scale structural rearrangements between chromosomes 1, 3 and 4 occurred during the 2.5 myr of divergence between B. frutescens and the most recent common ancestor of the B. laevigata complex 20 (Fig. 1 b). The consistency of Hi-C data across chromosomes confirmed the accuracy of the assemblies and further revealed anti-diagonal interactions between chromosome arms and interchromosomal interactions involving telomeres and centromeres (Fig. 1 c; Supplementary Fig. S3). Similar interactions have been documented in several large genomes, such as in Triticeae, where the Rabl configuration has been proposed 27 , 28 . Rabl involves the alignment of chromosome arms, with pericentromeres and telomeres clustering on opposite sides of the nucleus 29 . Because the interchromosomal interactions between Biscutella centromeres are less distinct than in Triticeae, we cannot exclude that Biscutella chromosomes may adopt a dispersed chromatin state more similar to rosette configuration 30 . Further analyses are necessary to clarify this matter. Additionally, Hi-C data reveal that NOR-bearing short arms of chromosomes 2 and 3 19 are densely compacted around the pericentromeric regions, indicating that in these chromosomes, centromeres might be located near telomeres and the nucleolus. Annotation of protein-coding genes and repeats (tandem repeats and TEs) characterized the genomes of Biscutella species as highly repetitive. Protein-based BUSCO analysis supported accurate gene models with completeness scores between 89.3% and 95.2% (Supplementary Tables S2-3). Only B. prealpina had a higher rate of missing BUSCO (genome-based: 7.1%; protein-based 8.5%), suggesting unassembled chromosomal regions. With 41,655 to 57,692 predicted gene models ( B. frutescens and B. austriaca , respectively; Supplementary Table S2), the annotations of Biscutella genomes, with almost twice the genes in the diploid M. pygmaea (25,607) 24 , are consistent with the estimates in other mesopolyploid genomes, such as B. rapa (47,531 gene models) 15 , indicating advanced post-WGD diploidization. In contrast to M. pygmaea , that shows a low proportion of duplicated BUSCO genes and TEs (2.2% and 42.6%, respectively) 24 , Biscutella had substantial amounts of duplicates (i.e. between 14.3% and 24.9% BUSCO genes) and TEs (i.e. between 69.88% and 71.89% of chromosomes; Supplementary Table S5). Although the small genome of B. frutescens (710 Mb) has the lowest percentage of duplicated BUSCO genes and TE content (14.3% and 69.88%, respectively), the extensive TE content in Biscutella genomes is unlikely due to duplication of pre-existing repeats and is hence likely the result of TE proliferation following WGD. Indeed, the dynamics of TEs estimated from sequence divergence (Fig. 1 d; Supplementary Fig. S4) were consistent with different TE classes, particularly retrotransposons, showing a peak of activation following the Bl-m-WGD event that is not observed in M. pygmaea . Analysis of LTR Gypsy retrotransposons along chromosomes LTR retrotransposons were characterized in terms of their abundance and distribution along the Biscutella chromosomes, with a particular focus on their dynamics in the pericentromeres. As expected, the distribution of their superfamilies showed distinct patterns, with Copia elements making up ca. 8% of the genomes and exhibiting a preference for distal, gene-rich chromosomal regions, whereas Gypsy elements comprising ca. 30% of the genomes were mostly localized in pericentromeric regions (Supplementary Table S5, Supplementary Fig. S5). Unlike the compact A. thaliana genome, where TEs are primarily confined to the pericentromeres 31 , Biscutella displayed a chromosome-wide distribution of TEs, similar to that observed in large, repeat-rich genomes 32 . Sequence homology within LTR retrotransposon protein domains (REXdb) 13 identified distinct TE clades that showed varying abundances and distribution patterns. Among the most abundant TE clades, Athila and CRM were the only LTR retrotransposons that showed a distribution closely matching pericentromeric regions (Fig. 2 a; Supplementary Fig. S6). Notably, B. varia and B. prealpina displayed a higher abundance of pericentromeric Athila and CRM clades (15.85% and 18.56% of their genomes, respectively) compared to B. frutescens and B. austriaca (11.65% and 12.15%, respectively; Supplementary Table S6). Consistent with their insertion bias 14 , CRM elements were mostly restricted to pericentromeres, whereas Athila elements, although primarily pericentromeric, also showed sparse occurrence in the gene-rich flanking GBs (Fig. 2 a-c). Only the NOR-bearing chromosomes 2 and 3 in Biscutella species exhibited a contrasting pattern, with both Athila and CRM clades extending beyond the pericentromeric boundaries towards NORs in the distal part of the short arms (Fig. 2 a; Supplementary Fig. S7). These short arms were characterized by reduced gene density and elevated CpG DNA methylation (Supplementary Fig. S7). Unlike most chromosomes, where DNA methylation gradually increases from telomeres (~ 30%) to centromeres (~ 85%), the short arms of chromosomes 2 and 3 maintained high DNA methylation level similar to those in the pericentromeres (Fig. 2 d; Supplementary Fig. S7), reflecting the enhanced chromatin interactions detected by Hi-C data (Fig. 1 C; Supplementary Fig. S3). NORs, which are known to co-localize with telomeres, centromeres and nucleolus 30 and to contain arrays of ribosomal DNA sequences (35S rDNA) similar to those in pericentromeres (5S rDNA) 19 , may promote the spread of pericentromeric CRM and Athila elements to the short arms. This phenomenon may lead to the heterochromatization of the short arms of NOR-bearing chromosomes. Supporting our findings, Chandrasekhara et al. 33 proposed that silencing of NOR2 in A. thaliana results from a centromere-proximal TE-rich region initiating heterochromatin formation towards the short-arm telomere. To further characterize the content of pericentromeres, we identified satellite DNA (satDNA; i.e, clusters of small tandem repeats) in these regions, as active centromeres are often associated with centromeric satDNA that colocalizes with centromere-specific (CENH3) histones 17 , 18 . We identified three such satDNA sequences (hereafter referred to as Cent213, Cent234, and Cent405) that are present in all four Biscutella species (Supplementary Fig. S7). Cent213 arrays, likely marking the position of active centromeres, were predominantly found at the centre of pericentromeric regions across most chromosomes. In contrast, Cent234 arrays were predominantly located in the centromeres of chromosomes 6 and 7. In situ hybridization of these tandem repeats confirmed the localization of Cent213 and Cent234 within primary constrictions corresponding to centromeres in B. frutescens and B. varia , respectively (Supplementary Fig. S8), validating the centromeric distribution of these satellites and their colocalizing TEs. Notably, perfect sequence homology was found between Cent405 and the highly abundant CRM family TE_00002954, with 586 TE copies colocalizing with Cent405 within the putative active centromere of chromosome 7 in B. varia . Similar relationships between TEs and satDNA have been reported in other plants 34 , 35 , suggesting a crucial role of the CRM clade in providing substrate for the formation and maintenance of tandem repeats in active centromeres. Further analysis of the distribution of Athila and CRM elements within the pericentromeres revealed that Athila copies generally outnumber CRM elements across all pericentromeres. However, an unusually high number of CRM copies overlap with centromeric satDNAs (Fig. 2 b-c). This pattern was even more pronounced in younger TE insertions (< 11 myr; Fig. 2 c), suggesting that CRMs initially targeted the centromeres, followed by the subsequent colonization of these regions by Athila elements. After aligning the pericentromeric sequences across species to detect possible structural rearrangements, we observed substantial synteny that reflects the phylogenetic relationships 21 among those species. However, we also identified considerable variation in both the size of pericentromeres and location of satDNA. Notably, chromosome 7 of B. prealpina and B. austriaca shared a large region (~ 10 Mb) rich in CRM elements and satDNA sequences that was absent in in B. varia (Fig. 2 b). Further supporting the association of CRM elements with the active centromere, our results suggest that species-specific bursts of pericentromeric TEs, particularly CRMs, may lead to significant divergence within 100 000 years and contribute to variation in the pericentromere size as well as repositioning of active centromeres within the pericentromeres. The evolution of pericentromeric TE families In contrast to ancient TE clades, which are conserved across all plants, TE families consist of small groups of similar TE copies that descend from transpositionally-active mother copies - evolutionary young TEs encoding all the proteins necessary for genome mobilization 36 . Within the CRM and Athila clades, TE families exhibited species-specific differences in pericentromeres of closely related Biscutella genomes. Compared to B. austriaca and the early diverging B. frutescens , the recently diverged B. varia and B. prealpina displayed a substantially higher number of CRM and Athila families, which is reflected not only in the total number of TE copies but also in the prevalence of intact copies (Table 1 ). Specifically, B. frutescens and B. austriaca had only 1.7% (358) and 2.2% (449) intact copies within the CRM clade, respectively, whereas B. varia and B. prealpina had 5.3% (1,360) and 8.5% (1,915) intact copies, respectively. This suggests that pericentromeres of the latter species have been evolutionarily more dynamic, featuring more diverse and younger TE communities. As estimated by coalescent modelling in Grünig et al. 21 , B. varia and B. prealpina have effective population sizes two to three times smaller than those of other species, suggesting that reduced efficiency of selection under stronger genetic drift has contributed to the significant changes observed in the distribution and accumulation of pericentromeric TEs 37 . Table 1 Diversity of Athila and CRM retrotransposons within Biscutella pericentromeres, presented as the number of TE families, along with the total number of annotated copies and the number of intact copies. Species CRM families Athila families CRM copies Athila copies CRM intact copies # Athila intact copies # B. varia 213 229 25,894 52,738 1,360 1,659 B. prealpina 100 277 22,609 70,640 1,915 1,118 B. frutescens 78 214 20,788 42,234 358 821 B. austriaca 75 162 20,483 52,574 449 734 # Structurally intact TE copies represent young TE insertions that contain all proteins required for transpositional activity. Both CRM and Athila families exhibited variable abundances both within and between genomes (Supplementary Tables S7-S9). While each CRM family averaged between 122 copies in B. varia and 273 copies in B. austriaca , we identified six families with exceptionally high copy numbers, exceeding 3,000 copies and reaching up to 8,360 copies (family TE_00005306) in B. prealpina . A broad array of large CRM families indeed diversified in B. varia , with the ten most abundant families accounting for only ~ 45% of all CRM copies, as compared to ~ 70% in the other genomes. TE families also displayed distinct distribution patterns across pericentromeric regions and chromosome arms. Copies were either confined to pericentromeric or distal regions or evenly distributed across chromosomes (Supplementary Tables S7-8). By focusing on the largest TE families (the top 20% by abundance per genome) to ensure robust predictions, we found that most CRM families were restricted to pericentromeric regions, with very few showing distal or dispersed distribution (Fig. 2 b; Supplementary Table S9). In situ hybridization of probes targeting four high-copy number CRM families, each with distinct in silico distributions, confirmed that families TE_00005474 and TE_00003221 are restricted to pericentromeres. In contrast, TE_00005833 and TE_00003812 exhibited a widespread distribution along the chromosomes of B. varia and B. frutescens (Supplementary Fig. S9). These results validate our in silico analyses and confirm that related CRM retrotransposon families have distinct distribution patterns (Supplementary Tables S7-8). In particular, B. varia presented an increased proportion of distal and dispersed CRM copies, with 10 families and approximately 26% of copies distributed across chromosomes, that likely reflect recent transpositional activity. In contrast, few Athila families were restricted to pericentromeres and generally show a more dispersed distribution than CRMs (Supplementary Table S9), with multiple copies overlapping with gene-rich regions (Supplementary Fig. S7). The mechanism by which CRM copies spread from pericentromeric regions and diversified into multiple families after having invaded new genomic contexts remains elusive. To assess the evolution of pericentromeres among Biscutella genomes, we further analysed the dynamics of Athila and CRM families (Fig. 3 ). We identified two main waves of transpositional activity in the genomes of B. austriaca and B. frutescens . An older peak (~ 30 myr ago) reflects a substantial TE burst (mode: 1,075 TEs), predominantly driven by CRM activity. This is contrasted by a more recent peak (> 11 myr ago), characterized by an equal contribution of Athila and CRM elements (modes: 467 and 353 TEs, respectively). These finding suggest that pericentromeres in these species are relatively stable, with older CRM copies being subsequently colonized by Athila elements. In contrast, B. prealpina and B. varia showed a higher turnover of CRM sequences, with many old CRM copies being replaced by more recent ones. Similar patterns were observed for Athila families with dispersed distribution. Finally, we extended our analysis to the last 3 myr of pericentromeric evolution by examining the dynamics of full-length Athila and CRM copies and estimating the insertion age based on sequence divergence between LTR pairs (Fig. 3 ). Notably, full-length copies in the pericentromeres are younger than those dispersed throughout the chromosomes (~ 0.7 vs 1.3 my in B. varia ), and are predominantly CRM rather than Athila. This supports the hypothesis that centromeres are specifically targeted by CRM elements and that Athila retrotransposons establish themselves on top of CRM copies (Fig. 2 c). Influence of pericentromeric TEs on gene expression Given that a substantial fraction of pericentromeric TEs are found in gene-rich regions, with up to ~ 49% of Athila and ~ 32% of CRM copies being found in chromosome arms of B. austriaca and B. varia , respectively (Supplementary Tables S7 and S8), we aimed to investigate their impact on CpG DNA methylation and the expression of neighboring genes. As expected, we observed very high levels of methylation in Athila and CRM elements, with approximately 90% and 88% of methylated cytosines, respectively (Fig. 4 a). This pattern matches the CpG methylation state typically associated with pericentromeric regions (~ 85%). In contrast, genes exhibited lower methylation levels than chromosome arms (~ 70%), with a median CpG methylation of ~ 12%. Generally, only genes with CRM or Athila insertions within 200 bp showed increased DNA methylation, while this influence diminishes at distances ranging from 200 bp to 2000 bp (Fig. 4 b; Supplementary Fig. S10). These results are in line with previous findings in Brachypodium distachion , where methylation extends only a few hundred base pairs from TEs 12 . As an exception, we found evidence that CRM-induced methylation can spread over longer distances and have a greater impact on neighboring genes in B. prealpina , with DNA methylation levels (up to 89%) generally matching those observed in pericentromeric regions (Fig. 4 c; Supplementary Fig. S10). Finally, our analysis reveals a negatively correlation between gene methylation and gene expression (Supplementary Fig. S11), suggesting that Athila and CRM elements dispersed within gene-rich regions may indeed contribute to reduced gene expression. Discussion Chromosome-scale assemblies of closely related Biscutella mesopolyploid genomes highlight their highly repetitive and redundant nature, whereas Hi-C analysis reveal that such large chromosomes (> 100 Mb) adopt a dispersed chromatin organization. Although synteny analyses matching cytogenetic maps confirm highly collinear genomes among species having split 0.1 myr ago within the B. laevigata complex, comparison with the early-diverging (2.5 myr ago) B. frutescens highlights large-scale restructuring including inter-chromosomal rearrangements. The identification of different chromosomal compartments based on TE and gene distribution supports similar spatial organization among species, with shared satDNA sequences across TE-rich pericentromeric regions delineating active centromeres. Interestingly, as expected under the hypothesis that TEs form communities within genomes 38 , abundant TEs exhibit both clade- and family-specific distributions along chromosomes in Biscutella (i.e. pericentromeric, dispersed or distal patterns). While Athila and CRM clades have long been recognized as abundant centromeric TEs, they have always been investigated separately 17 , 18 . To our knowledge, this is the first time that these clades are shown to co-localize to such an extent and to potentially interact, suggesting an intricate interplay. Comparative analysis indeed reveals contrasting genomic distribution and activity, indicating distinct roles in genome and centromere evolution. More diverse and younger communities of Athila and CRM sequences are found in species with lower effective population sizes such as B. varia and B. prealpina. Furthermore, alignment of pericentromeric regions revealed substantial sequence diversity compared to chromosome arms, with size variation and relocalization of centromeric satellite within the B. laevigata complex, coinciding with the recent origin of these closely related species in spatial isolation 21 . This is in line with recent findings in A. thaliana , where large rearrangements were identified in and near centromeres of 69 accessions 39 . Notably, the enrichment of young CRM insertions in regions containing centromere-specific satellites (i.e. putative active centromeres) as well as the dynamics of full-length copies within the last 3 myr, support CRMs as main drivers of the evolution of pericentromeres. We suggest that CRM elements, benefiting from their chromodomain, track active centromeres to establish new pericentromeres, which are later colonized by Athila elements that preferentially occupy these gene-poor regions primarily composed of CRM copies. Consistent with these findings, the rapidly diverging centromeres of Brassica rapa also contain only a few Athila copies and are predominantly invaded by CRM elements 15 . However, the interactions between co-occurring TEs that potentially compete for landing sites across pericentromeres, and how these constraints shape their diversification, remain to be investigated in more detail. Highlighting the impact of pericentromeric TEs on DNA methylation and its subsequent influence on the expression of neighboring gene, our study is consistent with intricate relationships between TEs and their host genomes 10 , 15 , 40 . In particular, NOR-bearing chromosomes 2 and 3 are distinct from other chromosomes due to the higher abundance of pericentromeric TEs along their chromosome arms, which correlates with increased DNA methylation levels and reduced gene density. Interestingly, A. thaliana shows similar levels of high sequence diversity in the NOR-bearing short arms of chromosomes 2 and 4 39 . Implying a role for NOR in influencing the distribution of specialized TEs, the organization of mesopolyploid Biscutella genomes shows that pericentromeric TEs can shape the structure and function of chromosomes beyond the vicinity of centromeres, and potentially affect chromosome restructuring 20 . These findings lay a foundation for future investigations into the functional consequences of pericentromeric TEs, as well as other TEs, and open new avenues for understanding the interplay between transposition-related mutations and drivers of TE distribution across heterogeneous chromosomes. Material and Methods Plant material, DNA and RNA extraction, sequencing The following accessions were used for this study: B. austriaca (A2Schnee 3B; Austria, Schneealpe Altenberg; 47.6968°, 15.6100°) 20 , B. prealpina (RCBO_NC17; Italy, Recoaro Terme; 45.696929°, 11.150395°) 21 , B. varia (V12-4; Germany, Beuron; 48.0516139°, 008.9835583°) 21 , and B. frutescens (PI 650129; Spain; no precise geolocation available) 41 . High molecular weight (HMW) DNA extraction was adapted by combining the A. thaliana leaf DNA protocol from ONT (community.nanoporetech.com/extraction_methods/arabidopsis-leaf-dna) and the HMW DNA protocol from Driguez et al 42 . Leaf tissues (approximately 1 g) were ground into a fine powder, transferred to a falcon tube and pre-cooled at -20°C for 10–30 minutes to facilitate lysis. According to the QIAGEN Genomic DNA Handbook, lysis buffer (19 ml Buffer G2 + 38 µl RNase A 100 mg/ml) was added to the ground powder, followed by gentle homogenization through inversion. Proteinase K (1000 µl) was added, and the mixture was incubated for over 3.5 hours at 50°C with periodic inversion to ensure homogeneity. The lysate was then centrifuged, and the supernatant was carefully transferred to pre-calibrated Genomic-tip 500/G columns (QIAGEN). The supernatant was washed twice with 15 mL buffer QC. Elution was performed with 15 mL pre-heated elution buffer QF, and the DNA was precipitated by adding 10.5 mL isopropanol. After centrifugation, DNA pellet was washed twice with 4 mL of fresh 70% ethanol. After air-drying, the DNA was resuspended in > 100 µl TE buffer and incubated for complete dissolution. Long-read whole-genome sequencing was performed on Oxford Nanopore Technologies (ONT) system by Novogene Company (Beijing, China). Sequencing libraries were prepared using the ONT ligation sequencing kit V14 and later sequenced on Nanopore PromethION (estimated sequencing depth of 60X). B. prealpina sequences were complemented with sequencing data produced on-site at the University of Bern (Switzerland) using Nanopore MinION. Short-read whole-genome sequencing (used for polishing ONT contigs) was performed at the Next Generation Sequencing (NGS) facility of Bern, Switzerland ( https://www.ngs.unibe.ch/ ), using the Illumina TruSeq DNA PCR-free kit for library preparation and 150bp paired-end reads (insert size of 550 bp) for sequencing on Illumina NovaSeq 6000 (estimated sequencing depth of 40X). Hi-C sequencing of three species ( B. varia, B. prealpina and B. frutescens ) was carried out by Phase Genomics company (Seattle, USA). During cross-linking and DNA digestion steps, a four restriction enzymes cocktail (DPNII, DDE1, HINF, MSEI) was used. After library preparation, 150bp Illumina paired-end reads were generated. Hi-C sequences of B. austriaca were previously published in Beringer et al 20 . For B. varia, B. prealpina and B. frutescens , RNA samples from root, leaf and flower tissues were extracted using the miRNeasy kit (Qiagen), according to the manufacturer’s instructions. Quality of the extracted RNA were assessed with the NanoDrop ND1000 spectrophotometer based on the 260:280 ratios. The SMRTbell prep kit 3.0 from PacBio was used for generating Iso-Seq libraries, and sequencing was carried out on PacBio Sequel II at the NGS facility of Bern. For B. austriaca , the comprehensive atlas of Illumina RNA-Seq data from seven different tissues (bud, leaf, senescent leaf, meristem, flower, roots, stem) from Beringer et al 20 was used. Sequences generated in this study are available under the NCBI BioProject PRJNA1124645. Genome assembly Sequencing adapters of ONT reads that passed basecalling quality check (q-score ≥ 7) were trimmed with Porechop ( https://github.com/rrwick/Porechop ) using standard parameters. Similarly, Illumina PE data were treated with Trim Galore 43 using standard parameters. ONT reads were initially assembled using NextDenovo 44 adapting following parameters in the run.cfg option file: “read_cutoff = 1k” (minimum read length), “genome_size = 1G” (estimated genome size), “seed_depth = 30” (estimated average sequencing depth), and “sort_options = -k 30” (estimated average sequencing depth). NextPolish 45 was then used for polishing the assembly using both Illumina and ONT reads. The standard run.cfg option file for short and long reads ( https://nextpolish.readthedocs.io/en/latest/TUTORIAL.html ) was modified using minimap2 46 as mapping tool for Illumina reads and ONT reads, with 5 Kb as minimum read length cut-off. Redundans 47 was used for reducing the heterozygosity of the polished genome assembly with following parameters: “--identity 0.90 --noscaffolding --nogapclosing”. Merqury copy number spectrum plots 48 with k-mer size of 19 were used for assessing the genome assembly quality before and after heterozygosity reduction (Supplementary Fig. S12). Paired HiC reads were mapped separately to the genome using BWA-MEM 49 with standard parameters. Scripts from the E.S. Rice HiC pipeline ( https://github.com/esrice/hic-pipeline ) were used for filtering HiC mapped reads ( filter-chimeras.py removes experimental artifacts from the alignments and keeps uniquely mapped reads) and for combining BAM files from paired reads ( combine_ends.py ). Then, samtools v1.13 was used for fixing mates (samtools fixmate -m), removing PCR duplicates (samtools markdup -r) and sorting mapped reads by name (samtools sort -n). Finally, the clean and sorted BAM file was used for scaffolding the draft genome assembly (redundans output) using the HiC scaffolding tool YaHS 50 . For B. austriaca , YaHS was run using the parameters “-e GATC -l 10000 -q 30”, while for B. frutescens, B. varia, and B. prealpina the following parameters were used “-e GATC, CTNAG, GANTC, TTAA -l 10000 -q 30”. As a last step, to manually curate the scaffolded genome assemblies, HiC contact map were generated with “ juicer pre” and “ juicer_tools pre” 51 as shown on the YaHS github page ( https://github.com/c-zhou/yahs ). The HiC contact map was loaded on Juicebox 52 and technical misassemblies were visually corrected. The final corrected FASTA file was generated with “ juicer post” . TEs annotation The Extensive de novo TE Annotator (EDTA v1.9.6) 53 was used for automatically annotating TEs in each Biscutella species with the following parameters: “--species others --step all --anno 1”. In addition, the non-redundant coding sequences of Arabidopsis thaliana (TAIR10_cds_20110103) 54 were used to filter out protein-coding gene related sequences with the option “--cds”. The final non-redundant curated TE libraries ( $ genome.mod.EDTA.TElib.fa) were further integrated in the annotation of gene models (see below). Compared with the previously available reference genome of B. austriaca 20 , all our assemblies, including the highly reduced B. frutescens , presented a higher TE content, chiefly of long terminal repeat (LTR) retrotransposons from the Gypsy superfamily that raised from ca. 20% to more than 30%. Gypsy elements are known to be mainly localized within centromeres and their pericentromeres 13 , therefore suggesting that our assembly pipeline better resolved such highly repetitive regions. TE families were defined as groups of TEs sharing 80% sequence homology over at least 80 bp and 80% of their length 36 , and represent evolutionary young TE clades being species specific or shared by only closely related species. The EDTA pipeline automatically classifies TE copies into TE families and select the sequences of the most representative copy for each family in the final non-redundant TE libraries. To evaluate the distribution of TE families, we quantified the number of TE copies within both pericentromeric regions and chromosome arms. This allowed us to determine whether a particular TE family presented a biased distribution. If more than three-quarters of TE copies were located within either the pericentromere or the chromosome arms, the family was classified as pericentromeric or distal, respectively. Conversely, if there was no observable preference, we categorized the TE family as evenly and randomly “dispersed”. To ensure robust predictions, we restricted our analysis to the top 20% most abundant TE families per species. The relative age of TEs, indicative of their time of insertion, was assessed by dating the divergence of each TE copy to its consensus sequence (i.e., the most representative sequence per TE family). Absolute estimates were based on a synonymous substitution rate of 8.22 × 10 − 9 substitutions per synonymous site per year, as documented for Brassicaceae species 55 . Divergence values, expressed as percentages, were extracted from the RepeatMasker 56 output files "mod.out," which were generated through the EDTA pipeline during TE annotation. To explain the past dynamics, or activity, of various TE superfamilies, we employed the script parseRM.pl (available at https://github.com/4ureliek/Parsing-RepeatMasker-Outputs ) with options “-l 50,1 -v.” This script allowed the calculation of the cumulative amount of DNA (in base pairs) diverging by 1% from its consensus up to 50%. Subsequently, the parseRM.pl output was used to generate TE landscape plots for each genome, where the 1% divergence bins were translated into million-year windows using the aforementioned synonymous substitution rate. RNA-Seq analysis Available Illumina RNA-Seq data from seven different tissues of B. austriaca 20 were de novo assembled with Trinity 57 using standard parameters. In addition, each paired-end dataset was mapped separately against the B. austriaca genome with STAR 58 using the parameters “--outFilterMultimapNmax 10 --outFilterMismatchNoverLmax 0.05 --alignIntronMax 10000”. For the other three Biscutella species ( B. frutescens, B. varia, and B. prealpina ), the PacBio secondary analysis tools ( https://github.com/PacificBiosciences/IsoSeq ) were used for retrieving non-redundant transcripts from PacBio Iso-Seq data. First, high-quality transcripts (supported by ≥ 99% accuracy and ≥ 2 full-length non-concatemer reads) from different plant tissues were mapped separately to the respective genomes using pbmm2 (v1.13.1) with “align --preset ISOSEQ --sort” parameters. Then, for each species, non-redundant PacBio isoforms were selected by merging bam files from different tissues with “samtools merge” and collapsing redundant transcripts with “isoseq3 collapse” (v4.0.0) using standard parameters. Finally, PacBio full-length non-concatemer FLNC reads ( B. frutescens, B. varia, and B. prealpina ) or Illumina RNA-Seq reads ( B. austriaca ) from different plant tissues were used for measuring the expression of final MAKER gene models. Salmon 59 was used for estimating the expression these gene models. It was run in mapping-based mode with “-l A --validateMappings” parameters. Transcripts per million (TPM) were considered as proxy for expression. Gene annotation Gene models were predicted using a combination of two different gene prediction pipelines, namely MAKER 60 and BRAKER 61 , 62 . Given that different RNA sequencing technologies were used for B. austriaca compared to B. frutescens, B. varia, and B. prealpina , the workflow was adapted accordingly. For B. austriaca , the latest BRAKER3 pipeline 62 was used, combining bam files of the previously mapped Illumina RNA-Seq reads and the curated protein sequences from the Viridiplantae UniProtKB/Swiss-Prot database (release 2023) 63 . On the other side, for B. frutescens, B. varia, and B. prealpina , the long-read BRAKER protocol (github.com/Gaius-Augustus/BRAKER/blob/master/docs/long_reads/long_read_protocol.md) was used to integrate PacBio Iso-Seq data and protein reference sequences into a single prediction. Here, the workflow consists of three parts. First, BRAKER2 61 was run integrating the Viridiplantae UniProtKB/Swiss-Prot database as only extrinsic evidence. Second, the script stringtie2fa.py was used to extract genomic sequences of non-redundant PacBio isoforms and GeneMarkS-T 64 was later used to predict the protein-coding regions in these transcripts with the scripts gmst.pl and gmst2globalCoords.py, as shown in the github protocol. Finally, the long-read version of TSEBRA 65 was used to combine the two gene sets predicted by GeneMarkS-T and BRAKER2. The MAKER pipeline (v2.31.9) 60 was used in three steps, consisting of 1) homology-based gene prediction using transcript sequences and protein sequences as extrinsic evidence, 2) training of the ab-initio gene prediction software SNAP, and 3) ab-initio gene prediction and final integration of all gene models from MAKER and BRAKER pipelines. Step 1: Protein sequences from the Viridiplantae UniProtKB/Swiss-Prot database and from high-quality Brassicaceae genomes, namely Arabidopsis thaliana (TAIR V10), Arabidopsis lyrata (cv. MN47 V2.1; phytozome V12), Megadenia pygmaea 24 , Brassica rapa (cv. Chiifu V3.5) 66 , and Eutrema salsugineum (v1.0) 67 , were used as protein evidence (MAKER option “protein=”). In addition, transcripts from previously assembled Illumina RNA-Seq reads ( B. austriaca ) or non-redundant PacBio isoform sequences ( B. frutescens, B. varia, and B. prealpina ) were used as transcriptomic evidence (MAKER option “est=). TE nucleotide and protein sequences from the TREP database ( https://www.botinst.uzh.ch/en/research/genetics/thomasWicker/trep-db.html ) and additional TE nucleotide sequences from the Brassicaceae repbase database 68 , and from the previously generated EDTA TE libraries were used as TE evidence for repeat masking (MAKER options “rmlib=” and “repeat_protein=”). Additional options were specified in the MAKER option file (maker_opts.ctl): “model_org = all, softmask = 1, est2genome = 1, protein2genome = 1”. When MAKER is finished, a GFF3 is generated using the MAKER scripts gff3_merge . Step 2: SNAP 69 is trained using gene models with an AED of 0.5 or better and a length of 50 or more amino acids. First, high-confidence gene models are and converted from GFF3 to ZFF format using “maker2zff -l 50 -x 0.5 -d $ {base}_master_datastore_index.log”. Then, training sequences and flanking 1000bp are collected using “fathom -categorize 1000 genome.ann genome.dna” and “fathom -export 1000 -plus uni.ann uni.dna”. Finally, training parameters are created using “forge export.ann export.dna” and the script hmm-assembler.pl as shown on https://github.com/KorfLab/SNAP . Step3: Second round of MAKER gene prediction including homology-based gene models (step 1), final gene models from TSEBRA (BRAKER pipeline), and training parameters for SNAP and AUGUSTUS (generated during BRAKER prediction). The MAKER option file from step 1 (maker_opts.ctl) was modified by removing FASTA sequences (protein, EST, and TEs), adding the GFF3 derived from step1 (maker_gff=) with options “est_pass = 1, protein_pass = 1, rm_pass = 1, model_pass = 1”, and setting-up gene prediction options as follow: “snaphmm=” (SNAP training parameter), “augustus_species=” (species folder with AUGUSTUS parameter), “pred_gff=” (final TSEBRA GFF3), “est2genome = 0”, “protein2genome = 0”. Finally, GFF3 and FASTA files of final gene models were retrieved using MAKER scripts fasta_merge and gff3_merge . Synteny analyses In Brassicaceae the presence of 22 conserved genomic blocks (GBs) 25 offers valuable insights to compare the genomic structure of related species. Because A. thaliana was the first chromosome-scale plant genome assembly, GB boundaries are defined by A. thaliana gene loci and colored based on the eight linkage groups of the ancestral crucifer karyotype 25 . A. thaliana coding sequences were mapped on each Biscutella genome with GMAP 70 using “--cross-species -t 30 -f 2” parameters. Coding sequences were then extracted from GFF3 files with GffRead 71 and translated to proteins. Protein sequences were aligned (Blastp, -evalue 1e-10 -outfmt 6) against themselves and against A. thaliana proteins. Finally, Dupgenfinder 72 was used to measure pairwise syntenic relationships between A. thaliana and each Biscutella species using standard parameters. The collinearity output file was filtered to only consider synteny blocks containing at least 20 collinear genes. Based on homology to A. thaliana , these collinear genes were assigned to the respective GBs (corresponding to eight linkage groups of the ancestral crucifer karyotype) and their genomic positions were finally used to color Biscutella assembled chromosomes. To investigate synteny and structural rearrangements within Biscutella genomes, we employed two computational tools: GENESPACE 73 and the SyRI/plotsr pipeline 74 , 75 . Protein sequences and gene coordinates in BED format (chr, start, end, name) were used as input for GENESPACE, which identifies orthologous genes within genomic regions. On the other hand, Biscutella genomes were aligned pairwise using minimap2 46 with the "−ax asm5" parameters. Next, SyRI, with default settings, was run to detect syntenic regions and structural rearrangements. Finally, plotsr was used to visualize such relationships among pericentromeric regions. DNA methylation As initial procedures, ONT sequencing files were converted from FAST5 to BLOW5 format using “slow5tools f2s” 76 . Then, ONT reads were aligned to the respective genomes using minimap2 46 with the “-a -x map-ont” options. Following alignment, the tool f5c 77 was employed to quantify DNA methylation frequency through several steps. After indexing the BLOW5 files with “f5c index”, “f5c call-methylation” was used to call methylation in each genome with the parameters “-B 7.0M -K 800 -t 20”. Then, methylation frequency (calculated as percent of methylated Cytosines) was measured with “f5c meth-freq”. Finally, Methylartist 78 was used to estimate the average CpG methylation frequency within specific genomic features, including genes and TEs, as well as across 200bp genomic windows (corresponding for example to chromosome arms or pericentromeres). Initially, the methylation calling output was refined to contain only CpG motifs. Subsequently, Methylartist was used with the following parameters “db-nanopolish -t 2.0 -s” to create a database of the CpG-filtered methylation calling output. Finally, the “methylartist segmeth” command was used to calculate the average methylation frequency values within the regions of interest in the genome. Fluorescent in situ hybridization (FISH) FISH analyses were performed to validate the putative localization of centromeric satellite repeats and CRM families showing contrasting distribution patterns in B. varia and B. frutescens. Mitotic chromosome spreads from fixed root tips were prepared as described previously 79 . As FISH probes, 60nt sequences were designed to target Cent213 and Cent234 satellite repeats. For B. varia , the highly abundant CRM families TE_00005474 (pericentromeric) and TE_00005833 (distal) were selected, while for B. frutescens , the families TE_00003221 (pericentromeric) and TE_00003812 (dispersed) were chosen (Supplementary Table S7). To identify the most repetitive sequence of each family, LTR sequences were clustered using cd-hit 80 with the parameters “-sc 1 -sf 1 -d 0 -c 0.8”. The most representative sequences were then aligned with ClustalW 81 using standard settings to pinpoint the most conserved regions, which were subsequently used for primer design. DNA probe preparation and labelling followed the published 82 . For satellites with longer monomers, PCR primers were designed to face outward from the monomer; therefore, PCR amplification was performed only between monomers tandemly arrayed. For retrotransposons, PCR primers were designed to the GAG domain which is generally the most variable domain among different retrotransposon families. PCR products were purified using NucleoSpin Gel and PCR Clean-up kit (Macherey-Nagel) and labeled by nick translation. FISH was performed as described previously 82 . The preparations were photographed using a Zeiss Axioimager Z2 epifluorescence microscope with a CoolCube camera (MetaSystems). Images were acquired separately for all individual fluorochromes using appropriate excitation and emission filters (AHF Analysentechnik). The monochromatic images were pseudocolored, merged and cropped using Photoshop CS (Adobe Systems). Declarations Acknowledgments We thank Yile Huang, Hussein Anani, Marc Beringer, Matthias Heuberger and Thomas Wicker for insightful discussions during this work. This research was funded by the Czech Science Foundation (project no. 21-07748L to MAL), the Masaryk University Grant Agency (project no. MUNI/R/1268/2022 to TM) and the Swiss National Science Foundation (Grants 31003A_178938 and 310030L_197839 to CP). References Tenaillon MI, Hollister JD, Gaut B (2010) A triptych of the evolution of plant transposable elements. Trends Plant Sci 15:471–478 Kumar A, Bennetzen JL (1999) Plant retrotransposons. Annu Rev Genet 33:479–532 Larrimore KE, Rancati G (2019) The conditional nature of gene essentiality. Curr Opin Genet Dev 58–59:55–61 Schubert I, Lysak MA (2011) Interpretation of karyotype evolution should consider chromosome structural constraints. Trends Genet 27:207–216 Jiao WB, Schneeberger K (2020) Chromosome-level assemblies of multiple Arabidopsis genomes reveal hotspots of rearrangements with altered evolutionary dynamics. Nat Commun 11:1–10 Stein JC et al (2018) Genomes of 13 domesticated and wild rice relatives highlight genetic conservation, turnover and innovation across the genus Oryza. Nat. Genet. 50, 285–296 (2018) Fedoroff NV (2012) Transposable elements, epigenetics, and genome evolution. Science 338:758–767 Joly-Lopez Z, Bureau TE (2018) Exaptation of transposable element coding sequences. Curr Opin Genet Dev 49:34–42 Hollister JD et al (2011) Transposable elements and small RNAs contribute to gene expression divergence between Arabidopsis thaliana and Arabidopsis lyrata. Proc. Natl. Acad. Sci. U. S. A. 108, 2322–2327 Zhang H, Lang Z, Zhu JK (2018) Dynamics and function of DNA methylation in plants. Nat Rev Mol Cell Biol 19:489–506 Choi JY, Purugganan MD (2018) Evolutionary epigenomics of retrotransposon-mediated methylation spreading in Rice. Mol Biol Evol 35:365 Wyler M, Stritt C, Walser JC, Baroux C, Roulin AC (2020) Impact of transposable elements on methylation and gene expression across natural accessions of Brachypodium distachyon. Genome Biol Evol 12:1994–2001 Neumann P, Novák P, Hoštáková N, Macas J (2019) Systematic survey of plant LTR-retrotransposons elucidates phylogenetic relationships of their polyprotein domains and provides a reference for element classification. Mob DNA 10:1–17 Neumann P et al (2011) Plant centromeric retrotransposons: A structural and cytogenetic perspective. Mob DNA 2:1–16 Zhang L et al (2023) A near-complete genome assembly of Brassica rapa provides new insights into the evolution of centromeres. Plant Biotechnol J 21:1022–1032 Henikoff S, Ahmad K, Malik HS (2001) The centromere paradox: stable inheritance with rapidly evolving DNA. Science 293:1098–1102 Wlodzimierz P et al (2023) Cycles of satellite and transposon evolution in Arabidopsis centromeres. Nat 618:557–565 Chen J et al (2023) A complete telomere-to-telomere assembly of the maize genome. Nat Genet 55:1221–1231 Geiser C, Mandáková T, Arrigo N, Lysak MA, Parisod C (2016) Repeated whole-genome duplication, karyotype reshuffling, and biased retention of stress-responding genes in Buckler Mustard. Plant Cell 28:17–27 Beringer M et al (2024) Biased retention of environment-responsive genes following genome fractionation. Mol Biol Evol 41:msae155 Grünig S, Patsiou T, Parisod C (2024) Ice age-driven range shifts of diploids and expanding autotetraploids of Biscutella laevigata within a conserved niche. New Phytol 244:1616–1628 Hohmann N, Wolf EM, Lysak MA, Koch MA (2015) A time-calibrated road map of Brassicaceae species radiation and evolutionary history. Plant Cell 27:2770–2784 Guo X et al (2021) Linked by ancestral bonds: Multiple whole-genome duplications and reticulate evolution in a Brassicaceae tribe. Mol Biol Evol 38:1695–1714 Yang W et al (2021) The chromosome-level genome sequence and karyotypic evolution of Megadenia pygmaea (Brassicaceae). Mol Ecol Resour 21:871–879 Lysak MA, Mandáková T, Schranz ME (2016) Comparative paleogenomics of crucifers: ancestral genomic blocks revisited. Curr Opin Plant Biol 30:108–115 Schranz ME, Lysak MA, Mitchell-Olds T (2006) The ABC’s of comparative genomics in the Brassicaceae: building blocks of crucifer genomes. Trends Plant Sci 11:535–542 Jia J et al (2021) Homology-mediated inter-chromosomal interactions in hexaploid wheat lead to specific subgenome territories following polyploidization and introgression. Genome Biol 22:1–21 Mascher M et al (2017) A chromosome conformation capture ordered sequence of the barley genome. Nature 544:427–433 Tiang CL, He Y, Pawlowski WP (2012) Chromosome organization and dynamics during interphase, mitosis, and meiosis in plants. Plant Physiol 158:26–34 Shan W, Kubová M, Mandáková T, Lysak MA (2021) Nuclear organization in crucifer genomes: nucleolus-associated telomere clustering is not a universal interphase configuration in Brassicaceae. Plant J 108:528–540 Quesneville H (2020) Twenty years of transposable element analysis in the Arabidopsis thaliana genome. Mob DNA 11:1–13 Hloušková P, Mandáková T, Pouch M, Trávníček P, Lysak MA (2019) The large genome size variation in the Hesperis clade was shaped by the prevalent proliferation of DNA repeats and rarer genome downsizing. Ann Bot 124:103–120 Chandrasekhara C, Mohannath G, Blevins T, Pontvianne F, Pikaard CS (2016) Chromosome-specific NOR inactivation explains selective rRNA gene silencing and dosage control in Arabidopsis. Genes Dev 30:177–190 Sharma A, Wolfgruber TK, Presting GG (2013) Tandem repeats derived from centromeric retrotransposons. BMC Genomics 14:1–11 Belyayev A et al (2020) Transposons and satellite DNA: On the origin of the major satellite DNA family in the Chenopodium genome. Mob DNA 11:1–10 Wicker T et al (2007) A unified classification system for eukaryotic transposable elements. Nat Rev Genet 8:973–982 Lynch M, Walsh B (2007) The origins of genome architecture Brookfield JF (2005) Y. The ecology of the genome — mobile DNA elements and their hosts. Nat Rev Genet 6:128–136 Lian Q et al (2024) A pan-genome of 69 Arabidopsis thaliana accessions reveals a conserved genome structure throughout the global species range. Nat Genet 56:982–991 Sigman MJ, Slotkin RK (2015) The first rule of plant transposable element silencing: Location, location, location. Plant Cell 28:304–313 Kiefer C et al (2019) Interspecies association mapping links reduced CG to TG substitution rates to the loss of gene-body methylation. Nat Plants 5:846–855 Driguez P, Carty K, Putra A, Ermini L (2021) Workflow for generating HMW plant DNA for third generation sequencing with high N50 and high accuracy. protocols.io doi:dx.doi.org/10.17504/protocols.io.bafmibk6 Krueger F, James F, Ewels P, Afyounian E (2021) & Schuster-Boeckler, B. FelixKrueger/TrimGalore: v0.6.7. Zenodo 10.5281/ZENODO.5127899 Hu J et al (2024) NextDenovo: an efficient error correction and accurate assembly tool for noisy long reads. Genome Biol 25:1–19 Hu J, Fan J, Sun Z, Liu S (2020) NextPolish: a fast and efficient genome polishing tool for long-read assembly. Bioinformatics 36:2253–2255 Li H (2018) Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics 34:3094–3100 Pryszcz LP, Gabaldón T (2016) Redundans: an assembly pipeline for highly heterozygous genomes. Nucleic Acids Res 44:e113–e113 Rhie A, Walenz BP, Koren S, Phillippy AM, Merqury (2020) Reference-free quality, completeness, and phasing assessment for genome assemblies. Genome Biol 21:1–27 Li H (2013) Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM Zhou C, McCarthy SA, Durbin R (2023) YaHS: yet another Hi-C scaffolding tool. Bioinformatics 39 Durand NC et al (2016) Juicer provides a one-click system for analyzing loop-resolution Hi-C experiments. Cell Syst 3:95–98 Durand NC et al (2016) Juicebox provides a visualization system for Hi-C contact maps with unlimited zoom. Cell Syst 3:99–101 Ou S et al (2019) Benchmarking transposable element annotation methods for creation of a streamlined, comprehensive pipeline. Genome Biol 20:1–18 Berardini TZ et al (2015) The Arabidopsis information resource: Making and mining the ‘gold standard’ annotated reference plant genome. Genesis 53:474–485 Kagale S et al (2014) Polyploid evolution of the Brassicaceae during the Cenozoic era. Plant Cell 26:2777–2791 Smit A, Hubley R, Green P (2015) RepeatMasker Open-4.0 Grabherr MG et al (2011) Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol 29:644–652 Dobin A et al (2013) STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29:15–21 Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C (2017) Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods 14:417–419 Cantarel BL et al (2008) An easy-to-use annotation pipeline designed for emerging model organism genomes. Genome Res 18:188–196 Brůna T, Hoff KJ, Lomsadze A, Stanke M, Borodovsky M (2021) BRAKER2: automatic eukaryotic genome annotation with GeneMark-EP + and AUGUSTUS supported by a protein database. NAR Genomics Bioinforma 3:1–11 Gabriel L et al (2024) BRAKER3: Fully automated genome annotation using RNA-seq and protein evidence with GeneMark-ETP, AUGUSTUS, and TSEBRA. Genome Res 34:769–777 Bateman A et al (2023) UniProt: the Universal Protein Knowledgebase in 2023. Nucleic Acids Res 51:D523–D531 Tang S, Lomsadze A, Borodovsky M (2015) Identification of protein coding regions in RNA transcripts. Nucleic Acids Res 43:e78–e78 Gabriel L, Hoff KJ, Brůna T, Borodovsky M, Stanke M (2021) TSEBRA: transcript selector for BRAKER. BMC Bioinformatics 22:1–12 Zhang Z et al (2022) Improved reference genome annotation of Brassica rapa by Pacific Biosciences RNA sequencing. Front Plant Sci 13:841618 Yang R et al (2013) The reference genome of the halophytic plant Eutrema salsugineum. Front Plant Sci 4:45219 Bao W, Kojima KK, Kohany O (2015) Repbase Update, a database of repetitive elements in eukaryotic genomes. Mob DNA 6:1–6 Korf I (2004) Gene finding in novel genomes. BMC Bioinformatics 5:1–9 Wu TD, Watanabe CK (2005) GMAP: a genomic mapping and alignment program for mRNA and EST sequences. Bioinformatics 21:1859–1875 Pertea M, Pertea G (2020) GFF Utilities: GffRead and GffCompare. F1000Research 9, 304 Qiao X et al (2019) Gene duplication and evolution in recurring polyploidization-diploidization cycles in plants. Genome Biol 20:38 Lovell JT et al (2022) GENESPACE tracks regions of interest and gene copy number variation across multiple genomes. Elife 11 Goel M, Sun H, Jiao WB, Schneeberger K (2019) SyRI: finding genomic rearrangements and local sequence differences from whole-genome assemblies. Genome Biol 20:1–13 Goel M, Schneeberger K (2022) plotsr: visualizing structural similarities and rearrangements between multiple genomes. Bioinformatics 38:2922 Samarakoon H et al (2023) Flexible and efficient handling of nanopore sequencing signal data with slow5tools. Genome Biol 24:1–12 Gamaarachchi H et al (2020) GPU accelerated adaptive banded event alignment for rapid comparative nanopore signal analysis. BMC Bioinformatics 21:1–13 Cheetham SW, Kindlova M, Ewing AD (2022) Methylartist: tools for visualizing modified bases from nanopore sequence data. Bioinformatics 38:3109–3112 Mandáková T, Lysak MA (2016) Chromosome preparation for cytogenetic analyses in Arabidopsis. Curr Protoc Plant Biol 1:43–51 Li W, Godzik A (2006) Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22:1658–1659 Sievers F et al (2011) Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 7 Mandáková T, Lysak MA (2016) Painting of Arabidopsis chromosomes with chromosome-specific BAC clones. Curr Protoc plant Biol 1:359–371 Additional Declarations There is NO Competing Interest. Supplementary Files Poretti2024NatGenSuppMattbs.pdf Supplementary material Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5461468","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":381511981,"identity":"fe665fe6-2123-40bf-815f-f78c87a62f75","order_by":0,"name":"Manuel Poretti","email":"","orcid":"https://orcid.org/0000-0001-6915-2238","institution":"University of Fribourg","correspondingAuthor":false,"prefix":"","firstName":"Manuel","middleName":"","lastName":"Poretti","suffix":""},{"id":381511982,"identity":"5a943049-52bc-4708-b24e-56f979c6942f","order_by":1,"name":"Terezie Mandáková","email":"","orcid":"","institution":"CEITEC - Central European Institute of Technology, Masaryk University","correspondingAuthor":false,"prefix":"","firstName":"Terezie","middleName":"","lastName":"Mandáková","suffix":""},{"id":381511983,"identity":"02b560d9-bca7-4d94-8c1e-8b85ef5347ed","order_by":2,"name":"Rimjhim Choudhury","email":"","orcid":"","institution":"University of Fribourg","correspondingAuthor":false,"prefix":"","firstName":"Rimjhim","middleName":"","lastName":"Choudhury","suffix":""},{"id":381511984,"identity":"f95fa7a7-ecda-409b-9f09-4e0f3f44ef53","order_by":3,"name":"Martin Lysak","email":"","orcid":"https://orcid.org/0000-0003-0318-4194","institution":"Central European Institute of Technology - Masaryk University","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Lysak","suffix":""},{"id":381511980,"identity":"0c161f9f-39f5-44d7-87e7-2c44efa97894","order_by":4,"name":"Christian Parisod","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCklEQVRIie2QsWrDMBBAL3hV7FVe0l9QCTgtpPRXTgQ0ZSgUSiejH8ge6E9oKnST0eDFH2CTEpK9S+miQIeesyQF1dCtUD2Q4HT3dCcBRCJ/EQ5gj/sRQavWADiH02FQwXOloStQDStUcX7S9tGAkj2t9tXBX13MYFx9PN6Vk7R7K3Y73FJKh5u81sIx5JcvOl3kjXDTfLOcCcR7StmgIrgCR28ZGctEroWVZrMsuPRIKfxRqTzyW1KmBy1Kabqm4IjDiqXBJCkFdUmkadmwwlsajCm+MC5V17p/S6MeeoXxNqxka5W8+3l5Y+qV6/Qn/VjtnnOPOMnWYeVEwr7HLFz265pIJBL5l3wBlaJcvv0l3X4AAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-8798-0897","institution":"University of Fribourg","correspondingAuthor":true,"prefix":"","firstName":"Christian","middleName":"","lastName":"Parisod","suffix":""}],"badges":[],"createdAt":"2024-11-15 15:21:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5461468/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5461468/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":75291378,"identity":"642764a3-b1ea-4f6e-94a1-db73ea0b9ee6","added_by":"auto","created_at":"2025-02-03 05:59:17","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":347761,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGenome structure of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eBiscutella\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e species. a, \u003c/strong\u003e\u003cem\u003eIn-silico\u003c/em\u003e chromosome painting of \u003cem\u003eB. varia\u003c/em\u003e, with genes coloured along chromosomes (Bv1 to Bv9) according to their positions in the 22 genomic blocks along the eight linkage groups of the ancestral crucifer karyotype. Non-syntenic genes (in grey) mark the pericentromeres and triangles indicate the presence of telomeric repeats in most chromosomes. Given the shared structure among genomes of all taxa within the \u003cem\u003eB. laevigata\u003c/em\u003especies complex, only \u003cem\u003eB. varia\u003c/em\u003e is depicted (all taxa presented in Supplementary Fig. S1). \u003cstrong\u003eb, \u003c/strong\u003eLarge scale chromosomal rearrangements are found between chromosomes 1,3 and 4 of \u003cem\u003eB. laevigata\u003c/em\u003e species and \u003cem\u003eB. frutescens\u003c/em\u003ethat have diverged for 2.5 million years. Dashed lines highlight the loci flanking rearranged regions between different chromosomes. \u003cstrong\u003ec, \u003c/strong\u003eHi-C contact map of \u003cem\u003eB. varia\u003c/em\u003e following manual curation (all pseudochromosomes presented in Supplementary Fig. S3). The zoomed-in section focuses on scaffolds 2 to 5, corresponding to chromosomes Bv6, Bv5, Bv3, and Bv7 (blue boxes). Arrows highlight examples of anti-diagonal interactions as well as interchromosomal interactions within telomeres and centromeres which may indicate a chromatin configuration similar to Rabl. The presence of the nucleolus organizer region (NOR) on Bv3 explains the heterochromatic state of the short arm. \u003cstrong\u003ed, \u003c/strong\u003eTransposable element (TE) dynamics in\u003cem\u003e B. austriaca\u003c/em\u003e. The insertion time (age) in million years of a TE copy is based on sequence divergence to the family consensus sequence and the Y-axis represents the amount of genome space (Mb) covered by copies of different ages among TE superfamilies. Dashed lines denote whole-genome duplication (WGD) events (α, ~32.5 myr ago in blue and Bl-m, ~11.5 myr ago in red). Peaks of transpositional activity are visible following WGD events. All taxa presented in Supplementary Fig. S4.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-5461468/v1/3846077fead2b4eff8387a7c.png"},{"id":75291379,"identity":"82e8988d-1b6d-4ad9-84b3-a25ca74cabae","added_by":"auto","created_at":"2025-02-03 05:59:17","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":345838,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe abundance and distribution of pericentromeric retrotransposons in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eBiscutella\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e. a, \u003c/strong\u003eGenomic distribution of Athila (yellow) and CRM (blue) clades depicted as the number of transposable element (TE) copies along chromosomes of \u003cem\u003eB. varia\u003c/em\u003e. Dashed lines indicate pericentromeric boundaries, highlighting a distribution of CRM copies restricted to pericentromeres as compared to copies of Athila. Nucleolar chromosomes 2 and 3 show a continuous distribution of copies of Athila and CRM between centromeres and the nucleolus organizer region bearing telomeres. \u003cstrong\u003eb, \u003c/strong\u003eAlignment of chromosome 7 pericentromeres among \u003cem\u003eBiscutella\u003c/em\u003e species, with light grey and dark grey lines delineating gene positions along pericentromeres and chromosome arms, respectively. Red lines represent large arrays of centromeric satellite Cent405. Green lines indicate syntenic relationships between pericentromeres. TEs abundance was truncated at 150 copies per 400 Kbp for clarity, although it sums up to 361 copies of CRM at the location of Cent405 (37-37.2Mb) in \u003cem\u003eB. varia\u003c/em\u003e. \u003cstrong\u003ec, \u003c/strong\u003eAbundance of young (\u0026lt;11 myr; top) and old (\u0026gt;11 myr; bottom) Athila and CRM copies along chromosomes 2 and 7 in \u003cem\u003eB. varia\u003c/em\u003e, presented as in b. Putatively active centromeres show higher abundance of young CRM copies than Athila.\u003cstrong\u003e d, \u003c/strong\u003eCpG DNA methylation frequency (percentage of methylated cytosines per 200 bp windows) in \u003cem\u003eB. varia\u003c/em\u003e chromosomes 2 and 7. The methylation landscape dynamically changes, mirroring the distribution of pericentromeric TEs.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-5461468/v1/df5424a9de43427810a55304.png"},{"id":75291377,"identity":"bb185303-0d0d-480f-9711-c4e92806e837","added_by":"auto","created_at":"2025-02-03 05:59:17","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":143736,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDynamics of Athila and CRM retrotransposons in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eBiscutella\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e. \u003c/strong\u003eTop row: Distribution plots illustrate the abundance of transposable element (TE) copies having transposed over the last 40 myr, with insertion age estimated through divergence from each TE copy to its family consensus sequence. Bottom row:\u003cstrong\u003e \u003c/strong\u003ezoom-in on the last 3 myr, highlighting the recent dynamics of full-length TE copies after the divergence (~2.5 myr ago) between \u003cem\u003eB. frutescens\u003c/em\u003e and the \u003cem\u003eB. laevigata\u003c/em\u003e species, including \u003cem\u003eB. austriaca, B. prealpina\u003c/em\u003e, and \u003cem\u003eB. varia\u003c/em\u003e. Insertion age was estimated through divergence between LTR pairs. Three distinct TE categories are presented: families of Athila (yellow) and CRM (blue), showing an insertion bias towards pericentromeres, and the remaining Athila families evenly dispersed across pericentromeres and chromosome arms (grey). Notably, \u003cem\u003eB. austriaca\u003c/em\u003e and \u003cem\u003eB. frutescens\u003c/em\u003e show old and stable pericentromeres as compared to the young and dynamic ones highlighted in\u003cem\u003e B. prealpina\u003c/em\u003e and \u003cem\u003eB. varia\u003c/em\u003e. Full-length copies are accordingly younger and mostly dominated by CRM across in their pericentromeres, whereas fewer full-length TEs were characterized in species with old and stable pericentromeres; the Y-axis was adjusted to improve the visualization.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-5461468/v1/09a1ff76db6be6a587b9c882.png"},{"id":75291380,"identity":"bd8c5e73-a27d-46e5-b786-ebf142b33445","added_by":"auto","created_at":"2025-02-03 05:59:17","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":93654,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe impact of TE proximity on gene methylation. a, \u003c/strong\u003eBox plot illustrates the average frequency of CpG DNA methylation (percentage of methylated cytosines) across various genomic features in \u003cem\u003eB. varia\u003c/em\u003e, including chromosome compartments (pericentromeres and chromosome arms), gene coding sequences (CDS), and pericentromeric transposable elements (TEs) from the Athila and CRM clades. Notably, there is a large difference in CpG methylation levels between genes and TEs. \u003cstrong\u003eb-c\u003c/strong\u003e, CpG methylation levels of genes with CRM (blue) or Athila (orange) insertions within 100-2000 base pairs; \u003cem\u003eB. varia\u003c/em\u003e shows increased gene methylation when a TE is in close proximity (\u0026lt;100bp for CRM and \u0026lt;200 bp for Athila), but no detectable effect beyond 200 bp; in \u003cem\u003eB. prealpina\u003c/em\u003e increased methylation is observed beyond 200bp distance from CRM. All taxa presented in Supplementary Fig. S10.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-5461468/v1/0bcba114787a532a5e3ea987.png"},{"id":102963695,"identity":"e52fedd3-c16c-49b7-8e53-e8fcb772a422","added_by":"auto","created_at":"2026-02-19 04:20:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1920630,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5461468/v1/49ccd5ab-0c88-4e09-b6e8-442087352f53.pdf"},{"id":75291381,"identity":"80ee0ee4-2562-4539-b4cf-7f57396e86be","added_by":"auto","created_at":"2025-02-03 05:59:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":8322204,"visible":true,"origin":"","legend":"Supplementary material","description":"","filename":"Poretti2024NatGenSuppMattbs.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5461468/v1/4476a8a1386d1cbe402d0859.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"The role of centromeric transposable elements in shaping chromosome evolution","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTransposable elements (TEs) are major drivers of genome evolution, promoting high sequence turnover\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. The ability of these interspersed repeats to mobilize throughout the genome is an important source of genetic diversity and has profound direct and indirect consequences on genome size and chromosome number\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Recent evidence in \u003cem\u003eArabidopsis thaliana\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e and rice\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e demonstrates a substantial impact of TEs on chromosomal rearrangements, genome divergence, and speciation. The interactions between TEs and their host genomes however remain poorly understood, and their influence on the evolution of functional genes is yet to be fully quantified\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTEs can generate high-impact mutations when inserting into coding sequences\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, significantly affecting functional genes. Additionally, TEs are thought to regulate the expression of nearby genes through the spread of DNA methylation\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e, although the impact of TE-induced methylation on gene expression in TE-rich non-model genomes is still debated\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Purifying selection at the genomic level is expected to efficiently remove TE insertions from gene-rich chromosome arms with high recombination rates. In contrast, TEs and arrays of tandem repeats tend to accumulate in regions of low recombination, such as centromeres and flanking pericentromeres, though empirical evidence supporting this model is limited. Furthermore, certain TEs have been reported to target centromeric chromatin through a specific chromodomain (CHDCR)\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. In crucifer species (Brassicaceae), CRM elements, which present this chromodomain, show a biased insertion toward centromeric regions, whereas Athila elements lack this domain\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Similarly, in \u003cem\u003eBrassica rapa\u003c/em\u003e, Ale elements were also observed to be enriched in the centromeres\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Comparative analysis of these TEs is expected to shed light on the drivers of their genome-wide distribution.\u003c/p\u003e \u003cp\u003eCentromeres are crucial for chromosome segregation, yet their evolution remain poorly understood. Despite their conserved role, centromeric sequences evolve rapidly, a phenomenon known as the centromere paradox\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Recent technological advances have enabled the assembly of telomere-to-telomere genomes in non-model species, providing new opportunities to investigate enigmatic chromosomal regions such as pericentromeres\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. In this study, we present chromosome-scale assemblies of four closely related \u003cem\u003eBiscutella\u003c/em\u003e species (Buckler mustards; Brassicaceae), which diverged within the last 2.5\u0026nbsp;million years following a reduction in chromosome number and genome size after an 11.5-million-year-old whole-genome duplication event\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. We characterize the gene space and repetitive fraction of these \u003cem\u003eBiscutella\u003c/em\u003e genomes and assess the impact of centromeric and pericentromeric TEs on chromosome structure and genome content.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eChromosome-scale genome assembly and annotation of four\u003c/b\u003e \u003cb\u003eBiscutella\u003c/b\u003e \u003cb\u003especies\u003c/b\u003e\u003c/p\u003e \u003cp\u003eWe combined Oxford Nanopore (ONT; 60x coverage), Illumina (40x coverage), and Hi-C (80x coverage) sequencing technologies to generate high-quality chromosome-scale genome assemblies of four \u003cem\u003eBiscutella\u003c/em\u003e species (n\u0026thinsp;=\u0026thinsp;9). These included the early diverging \u003cem\u003eB. frutescens\u003c/em\u003e and three closely related taxa within the \u003cem\u003eB. laevigata\u003c/em\u003e species complex (thereafter, \u003cem\u003eB. austriaca\u003c/em\u003e, \u003cem\u003eB. prealpina\u003c/em\u003e and \u003cem\u003eB. varia\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e). These diploid-like genomes underwent the alpha whole-genome duplication (α-WGD) event, which is shared by all Brassicaceae around 32\u0026nbsp;million years myr ago\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, followed by a subsequent mesopolyploid WGD (Bl-m-WGD) around 11.5 myr ago\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. In contrast, the related \u003cem\u003eMegadenia pygmaea\u003c/em\u003e (\u003cem\u003en\u0026thinsp;=\u003c/em\u003e\u0026thinsp;6) did not experience the \u003cem\u003eBiscutella\u003c/em\u003e-specific Bl-m-WGD\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAll genomes were assembled into nine main scaffolds, spanning 710 to 1,001 Mb, consistent with genome size estimates, and presented BUSCO completeness scores ranging from 92.0\u0026ndash;99.1% (Supplementary Tables S1-3). Synteny analysis against \u003cem\u003eA. thaliana\u003c/em\u003e through the mapping of 22 Brassicaceae-specific genomic blocks\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e (GBs) onto our genomes showed high collinearity with the Ancestral Crucifer Karyotype\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea; Supplementary Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Given that GBs are restricted to chromosome arms, we utilized the extremities of the syntenic GBs flanking the centres of scaffolds to delineate the boundaries of pericentromeric regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea, Supplementary Table S4). For clarity and consistency, \u0026ldquo;pericentromere\u0026rdquo; is hence defined as the entire chromosomal region extending from the centromere to the beginning of the chromosome arm. Comparison with the comparative cytogenetic map of \u003cem\u003eB. varia\u003c/em\u003e validates the structure of our assemblies and identifies six previously missing GBs\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e, along with characterizing the assembly of highly repetitive regions, such as pericentromeres and nucleolus organizer regions (NORs)-bearing short arms of chromosomes 2 and 3. The identification of telomeric repeats in the distal regions of most scaffolds further confirmed the near-complete assemblies, with large scaffolds considered equivalent to chromosomes. Despite small-scale inversions and translocations within GBs (Supplementary Fig. S2), high chromosomal collinearity was observed among taxa within the \u003cem\u003eB. laevigata\u003c/em\u003e species complex, which diverged approximately 0.1 myr ago\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. In contrast, large-scale structural rearrangements between chromosomes 1, 3 and 4 occurred during the 2.5 myr of divergence between \u003cem\u003eB. frutescens\u003c/em\u003e and the most recent common ancestor of the \u003cem\u003eB. laevigata\u003c/em\u003e complex\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe consistency of Hi-C data across chromosomes confirmed the accuracy of the assemblies and further revealed anti-diagonal interactions between chromosome arms and interchromosomal interactions involving telomeres and centromeres (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec; Supplementary Fig. S3). Similar interactions have been documented in several large genomes, such as in Triticeae, where the Rabl configuration has been proposed\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Rabl involves the alignment of chromosome arms, with pericentromeres and telomeres clustering on opposite sides of the nucleus\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Because the interchromosomal interactions between \u003cem\u003eBiscutella\u003c/em\u003e centromeres are less distinct than in Triticeae, we cannot exclude that \u003cem\u003eBiscutella\u003c/em\u003e chromosomes may adopt a dispersed chromatin state more similar to rosette configuration\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Further analyses are necessary to clarify this matter. Additionally, Hi-C data reveal that NOR-bearing short arms of chromosomes 2 and 3\u003csup\u003e19\u003c/sup\u003e are densely compacted around the pericentromeric regions, indicating that in these chromosomes, centromeres might be located near telomeres and the nucleolus.\u003c/p\u003e \u003cp\u003eAnnotation of protein-coding genes and repeats (tandem repeats and TEs) characterized the genomes of \u003cem\u003eBiscutella\u003c/em\u003e species as highly repetitive. Protein-based BUSCO analysis supported accurate gene models with completeness scores between 89.3% and 95.2% (Supplementary Tables S2-3). Only \u003cem\u003eB. prealpina\u003c/em\u003e had a higher rate of missing BUSCO (genome-based: 7.1%; protein-based 8.5%), suggesting unassembled chromosomal regions. With 41,655 to 57,692 predicted gene models (\u003cem\u003eB. frutescens\u003c/em\u003e and \u003cem\u003eB. austriaca\u003c/em\u003e, respectively; Supplementary Table S2), the annotations of \u003cem\u003eBiscutella\u003c/em\u003e genomes, with almost twice the genes in the diploid \u003cem\u003eM. pygmaea\u003c/em\u003e (25,607)\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, are consistent with the estimates in other mesopolyploid genomes, such as \u003cem\u003eB. rapa\u003c/em\u003e (47,531 gene models)\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e, indicating advanced post-WGD diploidization.\u003c/p\u003e \u003cp\u003eIn contrast to \u003cem\u003eM. pygmaea\u003c/em\u003e, that shows a low proportion of duplicated BUSCO genes and TEs (2.2% and 42.6%, respectively)\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eBiscutella\u003c/em\u003e had substantial amounts of duplicates (i.e. between 14.3% and 24.9% BUSCO genes) and TEs (i.e. between 69.88% and 71.89% of chromosomes; Supplementary Table S5). Although the small genome of \u003cem\u003eB. frutescens\u003c/em\u003e (710 Mb) has the lowest percentage of duplicated BUSCO genes and TE content (14.3% and 69.88%, respectively), the extensive TE content in \u003cem\u003eBiscutella\u003c/em\u003e genomes is unlikely due to duplication of pre-existing repeats and is hence likely the result of TE proliferation following WGD. Indeed, the dynamics of TEs estimated from sequence divergence (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed; Supplementary Fig. S4) were consistent with different TE classes, particularly retrotransposons, showing a peak of activation following the Bl-m-WGD event that is not observed in \u003cem\u003eM. pygmaea\u003c/em\u003e.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eAnalysis of LTR Gypsy retrotransposons along chromosomes\u003c/h2\u003e \u003cp\u003eLTR retrotransposons were characterized in terms of their abundance and distribution along the \u003cem\u003eBiscutella\u003c/em\u003e chromosomes, with a particular focus on their dynamics in the pericentromeres. As expected, the distribution of their superfamilies showed distinct patterns, with Copia elements making up ca. 8% of the genomes and exhibiting a preference for distal, gene-rich chromosomal regions, whereas Gypsy elements comprising ca. 30% of the genomes were mostly localized in pericentromeric regions (Supplementary Table S5, Supplementary Fig. S5). Unlike the compact \u003cem\u003eA. thaliana\u003c/em\u003e genome, where TEs are primarily confined to the pericentromeres\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eBiscutella\u003c/em\u003e displayed a chromosome-wide distribution of TEs, similar to that observed in large, repeat-rich genomes\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eSequence homology within LTR retrotransposon protein domains (REXdb)\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e identified distinct TE clades that showed varying abundances and distribution patterns. Among the most abundant TE clades, Athila and CRM were the only LTR retrotransposons that showed a distribution closely matching pericentromeric regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea; Supplementary Fig. S6). Notably, \u003cem\u003eB. varia\u003c/em\u003e and \u003cem\u003eB. prealpina\u003c/em\u003e displayed a higher abundance of pericentromeric Athila and CRM clades (15.85% and 18.56% of their genomes, respectively) compared to \u003cem\u003eB. frutescens\u003c/em\u003e and \u003cem\u003eB. austriaca\u003c/em\u003e (11.65% and 12.15%, respectively; Supplementary Table S6). Consistent with their insertion bias\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e, CRM elements were mostly restricted to pericentromeres, whereas Athila elements, although primarily pericentromeric, also showed sparse occurrence in the gene-rich flanking GBs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea-c).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOnly the NOR-bearing chromosomes 2 and 3 in \u003cem\u003eBiscutella\u003c/em\u003e species exhibited a contrasting pattern, with both Athila and CRM clades extending beyond the pericentromeric boundaries towards NORs in the distal part of the short arms (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea; Supplementary Fig. S7). These short arms were characterized by reduced gene density and elevated CpG DNA methylation (Supplementary Fig. S7). Unlike most chromosomes, where DNA methylation gradually increases from telomeres (~\u0026thinsp;30%) to centromeres (~\u0026thinsp;85%), the short arms of chromosomes 2 and 3 maintained high DNA methylation level similar to those in the pericentromeres (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed; Supplementary Fig. S7), reflecting the enhanced chromatin interactions detected by Hi-C data (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC; Supplementary Fig. S3). NORs, which are known to co-localize with telomeres, centromeres and nucleolus\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e and to contain arrays of ribosomal DNA sequences (35S rDNA) similar to those in pericentromeres (5S rDNA)\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e, may promote the spread of pericentromeric CRM and Athila elements to the short arms. This phenomenon may lead to the heterochromatization of the short arms of NOR-bearing chromosomes. Supporting our findings, Chandrasekhara et al.\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e proposed that silencing of NOR2 in \u003cem\u003eA. thaliana\u003c/em\u003e results from a centromere-proximal TE-rich region initiating heterochromatin formation towards the short-arm telomere.\u003c/p\u003e \u003cp\u003eTo further characterize the content of pericentromeres, we identified satellite DNA (satDNA; i.e, clusters of small tandem repeats) in these regions, as active centromeres are often associated with centromeric satDNA that colocalizes with centromere-specific (CENH3) histones\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. We identified three such satDNA sequences (hereafter referred to as Cent213, Cent234, and Cent405) that are present in all four \u003cem\u003eBiscutella\u003c/em\u003e species (Supplementary Fig. S7). Cent213 arrays, likely marking the position of active centromeres, were predominantly found at the centre of pericentromeric regions across most chromosomes. In contrast, Cent234 arrays were predominantly located in the centromeres of chromosomes 6 and 7. \u003cem\u003eIn situ\u003c/em\u003e hybridization of these tandem repeats confirmed the localization of Cent213 and Cent234 within primary constrictions corresponding to centromeres in \u003cem\u003eB. frutescens\u003c/em\u003e and \u003cem\u003eB. varia\u003c/em\u003e, respectively (Supplementary Fig. S8), validating the centromeric distribution of these satellites and their colocalizing TEs. Notably, perfect sequence homology was found between Cent405 and the highly abundant CRM family TE_00002954, with 586 TE copies colocalizing with Cent405 within the putative active centromere of chromosome 7 in \u003cem\u003eB. varia\u003c/em\u003e. Similar relationships between TEs and satDNA have been reported in other plants\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e, suggesting a crucial role of the CRM clade in providing substrate for the formation and maintenance of tandem repeats in active centromeres. Further analysis of the distribution of Athila and CRM elements within the pericentromeres revealed that Athila copies generally outnumber CRM elements across all pericentromeres. However, an unusually high number of CRM copies overlap with centromeric satDNAs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb-c). This pattern was even more pronounced in younger TE insertions (\u0026lt;\u0026thinsp;11 myr; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec), suggesting that CRMs initially targeted the centromeres, followed by the subsequent colonization of these regions by Athila elements.\u003c/p\u003e \u003cp\u003eAfter aligning the pericentromeric sequences across species to detect possible structural rearrangements, we observed substantial synteny that reflects the phylogenetic relationships\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e among those species. However, we also identified considerable variation in both the size of pericentromeres and location of satDNA. Notably, chromosome 7 of \u003cem\u003eB. prealpina\u003c/em\u003e and \u003cem\u003eB. austriaca\u003c/em\u003e shared a large region (~\u0026thinsp;10 Mb) rich in CRM elements and satDNA sequences that was absent in in \u003cem\u003eB. varia\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). Further supporting the association of CRM elements with the active centromere, our results suggest that species-specific bursts of pericentromeric TEs, particularly CRMs, may lead to significant divergence within 100 000 years and contribute to variation in the pericentromere size as well as repositioning of active centromeres within the pericentromeres.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eThe evolution of pericentromeric TE families\u003c/h3\u003e\n\u003cp\u003eIn contrast to ancient TE clades, which are conserved across all plants, TE families consist of small groups of similar TE copies that descend from transpositionally-active mother copies - evolutionary young TEs encoding all the proteins necessary for genome mobilization\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Within the CRM and Athila clades, TE families exhibited species-specific differences in pericentromeres of closely related \u003cem\u003eBiscutella\u003c/em\u003e genomes. Compared to \u003cem\u003eB. austriaca\u003c/em\u003e and the early diverging \u003cem\u003eB. frutescens\u003c/em\u003e, the recently diverged \u003cem\u003eB. varia\u003c/em\u003e and \u003cem\u003eB. prealpina\u003c/em\u003e displayed a substantially higher number of CRM and Athila families, which is reflected not only in the total number of TE copies but also in the prevalence of intact copies (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Specifically, \u003cem\u003eB. frutescens\u003c/em\u003e and \u003cem\u003eB. austriaca\u003c/em\u003e had only 1.7% (358) and 2.2% (449) intact copies within the CRM clade, respectively, whereas \u003cem\u003eB. varia\u003c/em\u003e and \u003cem\u003eB. prealpina\u003c/em\u003e had 5.3% (1,360) and 8.5% (1,915) intact copies, respectively. This suggests that pericentromeres of the latter species have been evolutionarily more dynamic, featuring more diverse and younger TE communities. As estimated by coalescent modelling in Gr\u0026uuml;nig et al.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eB. varia\u003c/em\u003e and \u003cem\u003eB. prealpina\u003c/em\u003e have effective population sizes two to three times smaller than those of other species, suggesting that reduced efficiency of selection under stronger genetic drift has contributed to the significant changes observed in the distribution and accumulation of pericentromeric TEs\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDiversity of Athila and CRM retrotransposons within \u003cem\u003eBiscutella\u003c/em\u003e pericentromeres, presented as the number of TE families, along with the total number of annotated copies and the number of intact copies.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpecies\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCRM families\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAthila families\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCRM copies\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAthila copies\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eCRM intact copies\u003csup\u003e#\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAthila intact copies\u003csup\u003e#\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eB. varia\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e213\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e229\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e25,894\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e52,738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1,360\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1,659\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eB. prealpina\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e277\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22,609\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e70,640\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1,915\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1,118\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eB. frutescens\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e214\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20,788\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e42,234\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e358\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e821\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eB. austriaca\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e162\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20,483\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e52,574\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e449\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e734\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003csup\u003e#\u003c/sup\u003eStructurally intact TE copies represent young TE insertions that contain all proteins required for transpositional activity.\u003c/p\u003e \u003cp\u003eBoth CRM and Athila families exhibited variable abundances both within and between genomes (Supplementary Tables S7-S9). While each CRM family averaged between 122 copies in \u003cem\u003eB. varia\u003c/em\u003e and 273 copies in \u003cem\u003eB. austriaca\u003c/em\u003e, we identified six families with exceptionally high copy numbers, exceeding 3,000 copies and reaching up to 8,360 copies (family TE_00005306) in \u003cem\u003eB. prealpina\u003c/em\u003e. A broad array of large CRM families indeed diversified in \u003cem\u003eB. varia\u003c/em\u003e, with the ten most abundant families accounting for only\u0026thinsp;~\u0026thinsp;45% of all CRM copies, as compared to ~\u0026thinsp;70% in the other genomes. TE families also displayed distinct distribution patterns across pericentromeric regions and chromosome arms. Copies were either confined to pericentromeric or distal regions or evenly distributed across chromosomes (Supplementary Tables S7-8). By focusing on the largest TE families (the top 20% by abundance per genome) to ensure robust predictions, we found that most CRM families were restricted to pericentromeric regions, with very few showing distal or dispersed distribution (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb; Supplementary Table S9).\u003c/p\u003e \u003cp\u003e \u003cem\u003eIn situ\u003c/em\u003e hybridization of probes targeting four high-copy number CRM families, each with distinct \u003cem\u003ein silico\u003c/em\u003e distributions, confirmed that families TE_00005474 and TE_00003221 are restricted to pericentromeres. In contrast, TE_00005833 and TE_00003812 exhibited a widespread distribution along the chromosomes of \u003cem\u003eB. varia\u003c/em\u003e and \u003cem\u003eB. frutescens\u003c/em\u003e (Supplementary Fig. S9). These results validate our \u003cem\u003ein silico\u003c/em\u003e analyses and confirm that related CRM retrotransposon families have distinct distribution patterns (Supplementary Tables S7-8). In particular, \u003cem\u003eB. varia\u003c/em\u003e presented an increased proportion of distal and dispersed CRM copies, with 10 families and approximately 26% of copies distributed across chromosomes, that likely reflect recent transpositional activity. In contrast, few Athila families were restricted to pericentromeres and generally show a more dispersed distribution than CRMs (Supplementary Table S9), with multiple copies overlapping with gene-rich regions (Supplementary Fig. S7). The mechanism by which CRM copies spread from pericentromeric regions and diversified into multiple families after having invaded new genomic contexts remains elusive.\u003c/p\u003e \u003cp\u003eTo assess the evolution of pericentromeres among \u003cem\u003eBiscutella\u003c/em\u003e genomes, we further analysed the dynamics of Athila and CRM families (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). We identified two main waves of transpositional activity in the genomes of \u003cem\u003eB. austriaca\u003c/em\u003e and \u003cem\u003eB. frutescens\u003c/em\u003e. An older peak (~\u0026thinsp;30 myr ago) reflects a substantial TE burst (mode: 1,075 TEs), predominantly driven by CRM activity. This is contrasted by a more recent peak (\u0026gt;\u0026thinsp;11 myr ago), characterized by an equal contribution of Athila and CRM elements (modes: 467 and 353 TEs, respectively). These finding suggest that pericentromeres in these species are relatively stable, with older CRM copies being subsequently colonized by Athila elements. In contrast, \u003cem\u003eB. prealpina\u003c/em\u003e and \u003cem\u003eB. varia\u003c/em\u003e showed a higher turnover of CRM sequences, with many old CRM copies being replaced by more recent ones. Similar patterns were observed for Athila families with dispersed distribution. Finally, we extended our analysis to the last 3 myr of pericentromeric evolution by examining the dynamics of full-length Athila and CRM copies and estimating the insertion age based on sequence divergence between LTR pairs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Notably, full-length copies in the pericentromeres are younger than those dispersed throughout the chromosomes (~\u0026thinsp;0.7 vs 1.3 my in \u003cem\u003eB. varia\u003c/em\u003e), and are predominantly CRM rather than Athila. This supports the hypothesis that centromeres are specifically targeted by CRM elements and that Athila retrotransposons establish themselves on top of CRM copies (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eInfluence of pericentromeric TEs on gene expression\u003c/h3\u003e\n\u003cp\u003eGiven that a substantial fraction of pericentromeric TEs are found in gene-rich regions, with up to ~\u0026thinsp;49% of Athila and ~\u0026thinsp;32% of CRM copies being found in chromosome arms of \u003cem\u003eB. austriaca\u003c/em\u003e and \u003cem\u003eB. varia\u003c/em\u003e, respectively (Supplementary Tables S7 and S8), we aimed to investigate their impact on CpG DNA methylation and the expression of neighboring genes. As expected, we observed very high levels of methylation in Athila and CRM elements, with approximately 90% and 88% of methylated cytosines, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). This pattern matches the CpG methylation state typically associated with pericentromeric regions (~\u0026thinsp;85%). In contrast, genes exhibited lower methylation levels than chromosome arms (~\u0026thinsp;70%), with a median CpG methylation of ~\u0026thinsp;12%. Generally, only genes with CRM or Athila insertions within 200 bp showed increased DNA methylation, while this influence diminishes at distances ranging from 200 bp to 2000 bp (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb; Supplementary Fig. S10). These results are in line with previous findings in \u003cem\u003eBrachypodium distachion\u003c/em\u003e, where methylation extends only a few hundred base pairs from TEs\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. As an exception, we found evidence that CRM-induced methylation can spread over longer distances and have a greater impact on neighboring genes in \u003cem\u003eB. prealpina\u003c/em\u003e, with DNA methylation levels (up to 89%) generally matching those observed in pericentromeric regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec; Supplementary Fig. S10). Finally, our analysis reveals a negatively correlation between gene methylation and gene expression (Supplementary Fig. S11), suggesting that Athila and CRM elements dispersed within gene-rich regions may indeed contribute to reduced gene expression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eChromosome-scale assemblies of closely related \u003cem\u003eBiscutella\u003c/em\u003e mesopolyploid genomes highlight their highly repetitive and redundant nature, whereas Hi-C analysis reveal that such large chromosomes (\u0026gt;\u0026thinsp;100 Mb) adopt a dispersed chromatin organization. Although synteny analyses matching cytogenetic maps confirm highly collinear genomes among species having split 0.1 myr ago within the \u003cem\u003eB. laevigata\u003c/em\u003e complex, comparison with the early-diverging (2.5 myr ago) \u003cem\u003eB. frutescens\u003c/em\u003e highlights large-scale restructuring including inter-chromosomal rearrangements. The identification of different chromosomal compartments based on TE and gene distribution supports similar spatial organization among species, with shared satDNA sequences across TE-rich pericentromeric regions delineating active centromeres. Interestingly, as expected under the hypothesis that TEs form communities within genomes\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, abundant TEs exhibit both clade- and family-specific distributions along chromosomes in \u003cem\u003eBiscutella\u003c/em\u003e (i.e. pericentromeric, dispersed or distal patterns).\u003c/p\u003e \u003cp\u003eWhile Athila and CRM clades have long been recognized as abundant centromeric TEs, they have always been investigated separately\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. To our knowledge, this is the first time that these clades are shown to co-localize to such an extent and to potentially interact, suggesting an intricate interplay. Comparative analysis indeed reveals contrasting genomic distribution and activity, indicating distinct roles in genome and centromere evolution. More diverse and younger communities of Athila and CRM sequences are found in species with lower effective population sizes such as \u003cem\u003eB. varia\u003c/em\u003e and \u003cem\u003eB. prealpina.\u003c/em\u003e Furthermore, alignment of pericentromeric regions revealed substantial sequence diversity compared to chromosome arms, with size variation and relocalization of centromeric satellite within the \u003cem\u003eB. laevigata\u003c/em\u003e complex, coinciding with the recent origin of these closely related species in spatial isolation\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. This is in line with recent findings in \u003cem\u003eA. thaliana\u003c/em\u003e, where large rearrangements were identified in and near centromeres of 69 accessions\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Notably, the enrichment of young CRM insertions in regions containing centromere-specific satellites (i.e. putative active centromeres) as well as the dynamics of full-length copies within the last 3 myr, support CRMs as main drivers of the evolution of pericentromeres. We suggest that CRM elements, benefiting from their chromodomain, track active centromeres to establish new pericentromeres, which are later colonized by Athila elements that preferentially occupy these gene-poor regions primarily composed of CRM copies. Consistent with these findings, the rapidly diverging centromeres of \u003cem\u003eBrassica rapa\u003c/em\u003e also contain only a few Athila copies and are predominantly invaded by CRM elements\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. However, the interactions between co-occurring TEs that potentially compete for landing sites across pericentromeres, and how these constraints shape their diversification, remain to be investigated in more detail.\u003c/p\u003e \u003cp\u003eHighlighting the impact of pericentromeric TEs on DNA methylation and its subsequent influence on the expression of neighboring gene, our study is consistent with intricate relationships between TEs and their host genomes\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. In particular, NOR-bearing chromosomes 2 and 3 are distinct from other chromosomes due to the higher abundance of pericentromeric TEs along their chromosome arms, which correlates with increased DNA methylation levels and reduced gene density. Interestingly, \u003cem\u003eA. thaliana\u003c/em\u003e shows similar levels of high sequence diversity in the NOR-bearing short arms of chromosomes 2 and 4\u003csup\u003e39\u003c/sup\u003e. Implying a role for NOR in influencing the distribution of specialized TEs, the organization of mesopolyploid \u003cem\u003eBiscutella\u003c/em\u003e genomes shows that pericentromeric TEs can shape the structure and function of chromosomes beyond the vicinity of centromeres, and potentially affect chromosome restructuring\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThese findings lay a foundation for future investigations into the functional consequences of pericentromeric TEs, as well as other TEs, and open new avenues for understanding the interplay between transposition-related mutations and drivers of TE distribution across heterogeneous chromosomes.\u003c/p\u003e"},{"header":"Material and Methods","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003ePlant material, DNA and RNA extraction, sequencing\u003c/h2\u003e \u003cp\u003eThe following accessions were used for this study: \u003cem\u003eB. austriaca\u003c/em\u003e (A2Schnee 3B; Austria, Schneealpe Altenberg; 47.6968\u0026deg;, 15.6100\u0026deg;)\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eB. prealpina\u003c/em\u003e (RCBO_NC17; Italy, Recoaro Terme; 45.696929\u0026deg;, 11.150395\u0026deg;)\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eB. varia\u003c/em\u003e (V12-4; Germany, Beuron; 48.0516139\u0026deg;, 008.9835583\u0026deg;)\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e, and \u003cem\u003eB. frutescens\u003c/em\u003e (PI 650129; Spain; no precise geolocation available)\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHigh molecular weight (HMW) DNA extraction was adapted by combining the \u003cem\u003eA. thaliana\u003c/em\u003e leaf DNA protocol from ONT (community.nanoporetech.com/extraction_methods/arabidopsis-leaf-dna) and the HMW DNA protocol from Driguez et al\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. Leaf tissues (approximately 1 g) were ground into a fine powder, transferred to a falcon tube and pre-cooled at -20\u0026deg;C for 10\u0026ndash;30 minutes to facilitate lysis. According to the QIAGEN Genomic DNA Handbook, lysis buffer (19 ml Buffer G2\u0026thinsp;+\u0026thinsp;38 \u0026micro;l RNase A 100 mg/ml) was added to the ground powder, followed by gentle homogenization through inversion. Proteinase K (1000 \u0026micro;l) was added, and the mixture was incubated for over 3.5 hours at 50\u0026deg;C with periodic inversion to ensure homogeneity. The lysate was then centrifuged, and the supernatant was carefully transferred to pre-calibrated Genomic-tip 500/G columns (QIAGEN). The supernatant was washed twice with 15 mL buffer QC. Elution was performed with 15 mL pre-heated elution buffer QF, and the DNA was precipitated by adding 10.5 mL isopropanol. After centrifugation, DNA pellet was washed twice with 4 mL of fresh 70% ethanol. After air-drying, the DNA was resuspended in \u0026gt;\u0026thinsp;100 \u0026micro;l TE buffer and incubated for complete dissolution.\u003c/p\u003e \u003cp\u003eLong-read whole-genome sequencing was performed on Oxford Nanopore Technologies (ONT) system by Novogene Company (Beijing, China). Sequencing libraries were prepared using the ONT ligation sequencing kit V14 and later sequenced on Nanopore PromethION (estimated sequencing depth of 60X). \u003cem\u003eB. prealpina\u003c/em\u003e sequences were complemented with sequencing data produced on-site at the University of Bern (Switzerland) using Nanopore MinION. Short-read whole-genome sequencing (used for polishing ONT contigs) was performed at the Next Generation Sequencing (NGS) facility of Bern, Switzerland (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ngs.unibe.ch/\u003c/span\u003e\u003cspan address=\"https://www.ngs.unibe.ch/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eusing the\u003c/span\u003e Illumina TruSeq DNA PCR-free kit for library preparation and 150bp paired-end reads (insert size of 550 bp) for sequencing on Illumina NovaSeq 6000 (estimated sequencing depth of 40X).\u003c/p\u003e \u003cp\u003eHi-C sequencing of three species (\u003cem\u003eB. varia, B. prealpina and B. frutescens\u003c/em\u003e) was carried out by Phase Genomics company (Seattle, USA). During cross-linking and DNA digestion steps, a four restriction enzymes cocktail (DPNII, DDE1, HINF, MSEI) was used. After library preparation, 150bp Illumina paired-end reads were generated. Hi-C sequences of \u003cem\u003eB. austriaca\u003c/em\u003e were previously published in Beringer et al\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFor \u003cem\u003eB. varia, B. prealpina and B. frutescens\u003c/em\u003e, RNA samples from root, leaf and flower tissues were extracted using the miRNeasy kit (Qiagen), according to the manufacturer\u0026rsquo;s instructions. Quality of the extracted RNA were assessed with the NanoDrop ND1000 spectrophotometer based on the 260:280 ratios. The SMRTbell prep kit 3.0 from PacBio was used for generating Iso-Seq libraries, and sequencing was carried out on PacBio Sequel II at the NGS facility of Bern. For \u003cem\u003eB. austriaca\u003c/em\u003e, the comprehensive atlas of Illumina RNA-Seq data from seven different tissues (bud, leaf, senescent leaf, meristem, flower, roots, stem) from Beringer et al\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e was used.\u003c/p\u003e \u003cp\u003eSequences generated in this study are available under the NCBI BioProject PRJNA1124645.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eGenome assembly\u003c/h3\u003e\n\u003cp\u003eSequencing adapters of ONT reads that passed basecalling quality check (q-score\u0026thinsp;\u0026ge;\u0026thinsp;7) were trimmed with Porechop (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/rrwick/Porechop\u003c/span\u003e\u003cspan address=\"https://github.com/rrwick/Porechop\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) using standard parameters. Similarly, Illumina PE data were treated with Trim Galore\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e using standard parameters.\u003c/p\u003e \u003cp\u003eONT reads were initially assembled using NextDenovo\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e adapting following parameters in the run.cfg option file: \u0026ldquo;read_cutoff\u0026thinsp;=\u0026thinsp;1k\u0026rdquo; (minimum read length), \u0026ldquo;genome_size\u0026thinsp;=\u0026thinsp;1G\u0026rdquo; (estimated genome size), \u0026ldquo;seed_depth\u0026thinsp;=\u0026thinsp;30\u0026rdquo; (estimated average sequencing depth), and \u0026ldquo;sort_options = -k 30\u0026rdquo; (estimated average sequencing depth). NextPolish\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e was then used for polishing the assembly using both Illumina and ONT reads. The standard run.cfg option file for short and long reads (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://nextpolish.readthedocs.io/en/latest/TUTORIAL.html\u003c/span\u003e\u003cspan address=\"https://nextpolish.readthedocs.io/en/latest/TUTORIAL.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was modified using minimap2 \u003csup\u003e46\u003c/sup\u003e as mapping tool for Illumina reads and ONT reads, with 5 Kb as minimum read length cut-off.\u003c/p\u003e \u003cp\u003eRedundans\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e was used for reducing the heterozygosity of the polished genome assembly with following parameters: \u0026ldquo;--identity 0.90 --noscaffolding --nogapclosing\u0026rdquo;. Merqury copy number spectrum plots\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e with k-mer size of 19 were used for assessing the genome assembly quality before and after heterozygosity reduction (Supplementary Fig. S12).\u003c/p\u003e \u003cp\u003ePaired HiC reads were mapped separately to the genome using BWA-MEM\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e with standard parameters. Scripts from the E.S. Rice HiC pipeline (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/esrice/hic-pipeline\u003c/span\u003e\u003cspan address=\"https://github.com/esrice/hic-pipeline\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were used for filtering HiC mapped reads (\u003cem\u003efilter-chimeras.py\u003c/em\u003e removes experimental artifacts from the alignments and keeps uniquely mapped reads) and for combining BAM files from paired reads (\u003cem\u003ecombine_ends.py\u003c/em\u003e). Then, samtools v1.13 was used for fixing mates (samtools fixmate -m), removing PCR duplicates (samtools markdup -r) and sorting mapped reads by name (samtools sort -n). Finally, the clean and sorted BAM file was used for scaffolding the draft genome assembly (redundans output) using the HiC scaffolding tool YaHS\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. For \u003cem\u003eB. austriaca\u003c/em\u003e, YaHS was run using the parameters \u0026ldquo;-e GATC -l 10000 -q 30\u0026rdquo;, while for \u003cem\u003eB. frutescens, B. varia, and B. prealpina\u003c/em\u003e the following parameters were used \u0026ldquo;-e GATC, CTNAG, GANTC, TTAA -l 10000 -q 30\u0026rdquo;.\u003c/p\u003e \u003cp\u003eAs a last step, to manually curate the scaffolded genome assemblies, HiC contact map were generated with \u0026ldquo;\u003cem\u003ejuicer pre\u0026rdquo;\u003c/em\u003e and \u0026ldquo;\u003cem\u003ejuicer_tools pre\u0026rdquo;\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e as shown on the YaHS github page (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/c-zhou/yahs\u003c/span\u003e\u003cspan address=\"https://github.com/c-zhou/yahs\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The HiC contact map was loaded on Juicebox\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e and technical misassemblies were visually corrected. The final corrected FASTA file was generated with \u0026ldquo;\u003cem\u003ejuicer post\u0026rdquo;\u003c/em\u003e.\u003c/p\u003e\n\u003ch3\u003eTEs annotation\u003c/h3\u003e\n\u003cp\u003eThe Extensive de novo TE Annotator (EDTA v1.9.6)\u003csup\u003e \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e \u003c/sup\u003e was used for automatically annotating TEs in each \u003cem\u003eBiscutella\u003c/em\u003e species with the following parameters: \u0026ldquo;--species others --step all --anno 1\u0026rdquo;. In addition, the non-redundant coding sequences of \u003cem\u003eArabidopsis thaliana\u003c/em\u003e (TAIR10_cds_20110103)\u003csup\u003e \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e \u003c/sup\u003e were used to filter out protein-coding gene related sequences with the option \u0026ldquo;--cds\u0026rdquo;. The final non-redundant curated TE libraries (\u003cspan\u003e$\u003c/span\u003egenome.mod.EDTA.TElib.fa) were further integrated in the annotation of gene models (see below). Compared with the previously available reference genome of \u003cem\u003eB. austriaca\u003c/em\u003e \u003csup\u003e \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e \u003c/sup\u003e, all our assemblies, including the highly reduced \u003cem\u003eB. frutescens\u003c/em\u003e, presented a higher TE content, chiefly of long terminal repeat (LTR) retrotransposons from the Gypsy superfamily that raised from ca. 20% to more than 30%. Gypsy elements are known to be mainly localized within centromeres and their pericentromeres\u003csup\u003e \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e \u003c/sup\u003e, therefore suggesting that our assembly pipeline better resolved such highly repetitive regions.\u003c/p\u003e \u003cp\u003eTE families were defined as groups of TEs sharing 80% sequence homology over at least 80 bp and 80% of their length\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e, and represent evolutionary young TE clades being species specific or shared by only closely related species. The EDTA pipeline automatically classifies TE copies into TE families and select the sequences of the most representative copy for each family in the final non-redundant TE libraries. To evaluate the distribution of TE families, we quantified the number of TE copies within both pericentromeric regions and chromosome arms. This allowed us to determine whether a particular TE family presented a biased distribution. If more than three-quarters of TE copies were located within either the pericentromere or the chromosome arms, the family was classified as pericentromeric or distal, respectively. Conversely, if there was no observable preference, we categorized the TE family as evenly and randomly \u0026ldquo;dispersed\u0026rdquo;. To ensure robust predictions, we restricted our analysis to the top 20% most abundant TE families per species.\u003c/p\u003e \u003cp\u003eThe relative age of TEs, indicative of their time of insertion, was assessed by dating the divergence of each TE copy to its consensus sequence (i.e., the most representative sequence per TE family). Absolute estimates were based on a synonymous substitution rate of 8.22 \u0026times; 10\u0026thinsp;\u0026minus;\u0026thinsp;9 substitutions per synonymous site per year, as documented for \u003cem\u003eBrassicaceae\u003c/em\u003e species\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e. Divergence values, expressed as percentages, were extracted from the RepeatMasker\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e output files \"mod.out,\" which were generated through the EDTA pipeline during TE annotation.\u003c/p\u003e \u003cp\u003eTo explain the past dynamics, or activity, of various TE superfamilies, we employed the script parseRM.pl (available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/4ureliek/Parsing-RepeatMasker-Outputs\u003c/span\u003e\u003cspan address=\"https://github.com/4ureliek/Parsing-RepeatMasker-Outputs\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) with options \u0026ldquo;-l 50,1 -v.\u0026rdquo; This script allowed the calculation of the cumulative amount of DNA (in base pairs) diverging by 1% from its consensus up to 50%. Subsequently, the parseRM.pl output was used to generate TE landscape plots for each genome, where the 1% divergence bins were translated into million-year windows using the aforementioned synonymous substitution rate.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eRNA-Seq analysis\u003c/h2\u003e \u003cp\u003eAvailable Illumina RNA-Seq data from seven different tissues of \u003cem\u003eB. austriaca\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e were \u003cem\u003ede novo\u003c/em\u003e assembled with Trinity\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e using standard parameters. In addition, each paired-end dataset was mapped separately against the \u003cem\u003eB. austriaca\u003c/em\u003e genome with STAR\u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e using the parameters \u0026ldquo;--outFilterMultimapNmax 10 --outFilterMismatchNoverLmax 0.05 --alignIntronMax 10000\u0026rdquo;.\u003c/p\u003e \u003cp\u003eFor the other three \u003cem\u003eBiscutella\u003c/em\u003e species (\u003cem\u003eB. frutescens, B. varia, and B. prealpina\u003c/em\u003e), the PacBio secondary analysis tools (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/PacificBiosciences/IsoSeq\u003c/span\u003e\u003cspan address=\"https://github.com/PacificBiosciences/IsoSeq\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were used for retrieving non-redundant transcripts from PacBio Iso-Seq data. First, high-quality transcripts (supported by \u0026ge;\u0026thinsp;99% accuracy and \u0026ge;\u0026thinsp;2 full-length non-concatemer reads) from different plant tissues were mapped separately to the respective genomes using pbmm2 (v1.13.1) with \u0026ldquo;align --preset ISOSEQ --sort\u0026rdquo; parameters. Then, for each species, non-redundant PacBio isoforms were selected by merging bam files from different tissues with \u0026ldquo;samtools merge\u0026rdquo; and collapsing redundant transcripts with \u0026ldquo;isoseq3 collapse\u0026rdquo; (v4.0.0) using standard parameters.\u003c/p\u003e \u003cp\u003eFinally, PacBio full-length non-concatemer FLNC reads (\u003cem\u003eB. frutescens, B. varia, and B. prealpina\u003c/em\u003e) or Illumina RNA-Seq reads (\u003cem\u003eB. austriaca\u003c/em\u003e) from different plant tissues were used for measuring the expression of final MAKER gene models. Salmon\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e was used for estimating the expression these gene models. It was run in mapping-based mode with \u0026ldquo;-l A --validateMappings\u0026rdquo; parameters. Transcripts per million (TPM) were considered as proxy for expression.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eGene annotation\u003c/h2\u003e \u003cp\u003eGene models were predicted using a combination of two different gene prediction pipelines, namely MAKER\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e and BRAKER\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e,\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. Given that different RNA sequencing technologies were used for \u003cem\u003eB. austriaca\u003c/em\u003e compared to \u003cem\u003eB. frutescens, B. varia, and B. prealpina\u003c/em\u003e, the workflow was adapted accordingly. For \u003cem\u003eB. austriaca\u003c/em\u003e, the latest BRAKER3 pipeline\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e was used, combining bam files of the previously mapped Illumina RNA-Seq reads and the curated protein sequences from the \u003cem\u003eViridiplantae\u003c/em\u003e UniProtKB/Swiss-Prot database (release 2023)\u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. On the other side, for \u003cem\u003eB. frutescens, B. varia, and B. prealpina\u003c/em\u003e, the long-read BRAKER protocol (github.com/Gaius-Augustus/BRAKER/blob/master/docs/long_reads/long_read_protocol.md) was used to integrate PacBio Iso-Seq data and protein reference sequences into a single prediction. Here, the workflow consists of three parts. First, BRAKER2\u003csup\u003e61\u003c/sup\u003e was run integrating the \u003cem\u003eViridiplantae\u003c/em\u003e UniProtKB/Swiss-Prot database as only extrinsic evidence. Second, the script stringtie2fa.py was used to extract genomic sequences of non-redundant PacBio isoforms and GeneMarkS-T\u003csup\u003e\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e was later used to predict the protein-coding regions in these transcripts with the scripts gmst.pl and gmst2globalCoords.py, as shown in the github protocol. Finally, the long-read version of TSEBRA\u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e was used to combine the two gene sets predicted by GeneMarkS-T and BRAKER2.\u003c/p\u003e \u003cp\u003eThe MAKER pipeline (v2.31.9)\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e was used in three steps, consisting of 1) homology-based gene prediction using transcript sequences and protein sequences as extrinsic evidence, 2) training of the ab-initio gene prediction software SNAP, and 3) ab-initio gene prediction and final integration of all gene models from MAKER and BRAKER pipelines.\u003c/p\u003e \u003cp\u003eStep 1: Protein sequences from the \u003cem\u003eViridiplantae\u003c/em\u003e UniProtKB/Swiss-Prot database and from high-quality \u003cem\u003eBrassicaceae\u003c/em\u003e genomes, namely \u003cem\u003eArabidopsis thaliana\u003c/em\u003e (TAIR V10), \u003cem\u003eArabidopsis lyrata\u003c/em\u003e (cv. MN47 V2.1; phytozome V12), \u003cem\u003eMegadenia pygmaea\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eBrassica rapa\u003c/em\u003e (cv. Chiifu V3.5)\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e, and \u003cem\u003eEutrema salsugineum\u003c/em\u003e (v1.0)\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e, were used as protein evidence (MAKER option \u0026ldquo;protein=\u0026rdquo;). In addition, transcripts from previously assembled Illumina RNA-Seq reads (\u003cem\u003eB. austriaca\u003c/em\u003e) or non-redundant PacBio isoform sequences (\u003cem\u003eB. frutescens, B. varia, and B. prealpina\u003c/em\u003e) were used as transcriptomic evidence (MAKER option \u0026ldquo;est=). TE nucleotide and protein sequences from the TREP database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.botinst.uzh.ch/en/research/genetics/thomasWicker/trep-db.html\u003c/span\u003e\u003cspan address=\"https://www.botinst.uzh.ch/en/research/genetics/thomasWicker/trep-db.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and additional TE nucleotide sequences from the \u003cem\u003eBrassicaceae\u003c/em\u003e repbase database\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e, and from the previously generated EDTA TE libraries were used as TE evidence for repeat masking (MAKER options \u0026ldquo;rmlib=\u0026rdquo; and \u0026ldquo;repeat_protein=\u0026rdquo;). Additional options were specified in the MAKER option file (maker_opts.ctl): \u0026ldquo;model_org\u0026thinsp;=\u0026thinsp;all, softmask\u0026thinsp;=\u0026thinsp;1, est2genome\u0026thinsp;=\u0026thinsp;1, protein2genome\u0026thinsp;=\u0026thinsp;1\u0026rdquo;. When MAKER is finished, a GFF3 is generated using the MAKER scripts \u003cem\u003egff3_merge\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eStep 2: SNAP\u003csup\u003e \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e \u003c/sup\u003e is trained using gene models with an AED of 0.5 or better and a length of 50 or more amino acids. First, high-confidence gene models are and converted from GFF3 to ZFF format using \u0026ldquo;maker2zff -l 50 -x 0.5 -d \u003cspan\u003e$\u003c/span\u003e{base}_master_datastore_index.log\u0026rdquo;. Then, training sequences and flanking 1000bp are collected using \u0026ldquo;fathom -categorize 1000 genome.ann genome.dna\u0026rdquo; and \u0026ldquo;fathom -export 1000 -plus uni.ann uni.dna\u0026rdquo;. Finally, training parameters are created using \u0026ldquo;forge export.ann export.dna\u0026rdquo; and the script \u003cem\u003ehmm-assembler.pl\u003c/em\u003e as shown on \u003cspan class=\"ExternalRef\"\u003e \u003cspan class=\"RefSource\"\u003ehttps://github.com/KorfLab/SNAP\u003c/span\u003e \u003cspan address=\"https://github.com/KorfLab/SNAP\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e \u003c/span\u003e.\u003c/p\u003e \u003cp\u003eStep3: Second round of MAKER gene prediction including homology-based gene models (step 1), final gene models from TSEBRA (BRAKER pipeline), and training parameters for SNAP and AUGUSTUS (generated during BRAKER prediction). The MAKER option file from step 1 (maker_opts.ctl) was modified by removing FASTA sequences (protein, EST, and TEs), adding the GFF3 derived from step1 (maker_gff=) with options \u0026ldquo;est_pass\u0026thinsp;=\u0026thinsp;1, protein_pass\u0026thinsp;=\u0026thinsp;1, rm_pass\u0026thinsp;=\u0026thinsp;1, model_pass\u0026thinsp;=\u0026thinsp;1\u0026rdquo;, and setting-up gene prediction options as follow: \u0026ldquo;snaphmm=\u0026rdquo; (SNAP training parameter), \u0026ldquo;augustus_species=\u0026rdquo; (species folder with AUGUSTUS parameter), \u0026ldquo;pred_gff=\u0026rdquo; (final TSEBRA GFF3), \u0026ldquo;est2genome\u0026thinsp;=\u0026thinsp;0\u0026rdquo;, \u0026ldquo;protein2genome\u0026thinsp;=\u0026thinsp;0\u0026rdquo;. Finally, GFF3 and FASTA files of final gene models were retrieved using MAKER scripts \u003cem\u003efasta_merge\u003c/em\u003e and \u003cem\u003egff3_merge\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eSynteny analyses\u003c/h2\u003e \u003cp\u003eIn \u003cem\u003eBrassicaceae\u003c/em\u003e the presence of 22 conserved genomic blocks (GBs)\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e offers valuable insights to compare the genomic structure of related species. Because \u003cem\u003eA. thaliana\u003c/em\u003e was the first chromosome-scale plant genome assembly, GB boundaries are defined by \u003cem\u003eA. thaliana\u003c/em\u003e gene loci and colored based on the eight linkage groups of the ancestral crucifer karyotype\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. \u003cem\u003eA. thaliana\u003c/em\u003e coding sequences were mapped on each \u003cem\u003eBiscutella\u003c/em\u003e genome with GMAP\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e using \u0026ldquo;--cross-species -t 30 -f 2\u0026rdquo; parameters. Coding sequences were then extracted from GFF3 files with GffRead\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e and translated to proteins. Protein sequences were aligned (Blastp, -evalue 1e-10 -outfmt 6) against themselves and against \u003cem\u003eA. thaliana\u003c/em\u003e proteins. Finally, Dupgenfinder\u003csup\u003e\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e was used to measure pairwise syntenic relationships between \u003cem\u003eA. thaliana\u003c/em\u003e and each \u003cem\u003eBiscutella\u003c/em\u003e species using standard parameters. The collinearity output file was filtered to only consider synteny blocks containing at least 20 collinear genes. Based on homology to \u003cem\u003eA. thaliana\u003c/em\u003e, these collinear genes were assigned to the respective GBs (corresponding to eight linkage groups of the ancestral crucifer karyotype) and their genomic positions were finally used to color \u003cem\u003eBiscutella\u003c/em\u003e assembled chromosomes.\u003c/p\u003e \u003cp\u003eTo investigate synteny and structural rearrangements within \u003cem\u003eBiscutella\u003c/em\u003e genomes, we employed two computational tools: GENESPACE\u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e and the SyRI/plotsr pipeline\u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e,\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e. Protein sequences and gene coordinates in BED format (chr, start, end, name) were used as input for GENESPACE, which identifies orthologous genes within genomic regions. On the other hand, \u003cem\u003eBiscutella\u003c/em\u003e genomes were aligned pairwise using minimap2 \u003csup\u003e46\u003c/sup\u003e with the \"\u0026minus;ax asm5\" parameters. Next, SyRI, with default settings, was run to detect syntenic regions and structural rearrangements. Finally, plotsr was used to visualize such relationships among pericentromeric regions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eDNA methylation\u003c/h2\u003e \u003cp\u003eAs initial procedures, ONT sequencing files were converted from FAST5 to BLOW5 format using \u0026ldquo;slow5tools f2s\u0026rdquo;\u003csup\u003e\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e. Then, ONT reads were aligned to the respective genomes using minimap2 \u003csup\u003e46\u003c/sup\u003e with the \u0026ldquo;-a -x map-ont\u0026rdquo; options. Following alignment, the tool f5c\u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e was employed to quantify DNA methylation frequency through several steps. After indexing the BLOW5 files with \u0026ldquo;f5c index\u0026rdquo;, \u0026ldquo;f5c call-methylation\u0026rdquo; was used to call methylation in each genome with the parameters \u0026ldquo;-B 7.0M -K 800 -t 20\u0026rdquo;. Then, methylation frequency (calculated as percent of methylated Cytosines) was measured with \u0026ldquo;f5c meth-freq\u0026rdquo;. Finally, Methylartist\u003csup\u003e\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e\u003c/sup\u003e was used to estimate the average CpG methylation frequency within specific genomic features, including genes and TEs, as well as across 200bp genomic windows (corresponding for example to chromosome arms or pericentromeres). Initially, the methylation calling output was refined to contain only CpG motifs. Subsequently, Methylartist was used with the following parameters \u0026ldquo;db-nanopolish -t 2.0 -s\u0026rdquo; to create a database of the CpG-filtered methylation calling output. Finally, the \u0026ldquo;methylartist segmeth\u0026rdquo; command was used to calculate the average methylation frequency values within the regions of interest in the genome.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFluorescent\u003c/b\u003e \u003cb\u003ein situ\u003c/b\u003e \u003cb\u003ehybridization (FISH)\u003c/b\u003e\u003c/p\u003e \u003cp\u003eFISH analyses were performed to validate the putative localization of centromeric satellite repeats and CRM families showing contrasting distribution patterns in \u003cem\u003eB. varia\u003c/em\u003e and \u003cem\u003eB. frutescens.\u003c/em\u003e Mitotic chromosome spreads from fixed root tips were prepared as described previously\u003csup\u003e\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e. As FISH probes, 60nt sequences were designed to target Cent213 and Cent234 satellite repeats. For \u003cem\u003eB. varia\u003c/em\u003e, the highly abundant CRM families TE_00005474 (pericentromeric) and TE_00005833 (distal) were selected, while for \u003cem\u003eB. frutescens\u003c/em\u003e, the families TE_00003221 (pericentromeric) and TE_00003812 (dispersed) were chosen (Supplementary Table S7). To identify the most repetitive sequence of each family, LTR sequences were clustered using cd-hit\u003csup\u003e\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e\u003c/sup\u003e with the parameters \u0026ldquo;-sc 1 -sf 1 -d 0 -c 0.8\u0026rdquo;. The most representative sequences were then aligned with ClustalW\u003csup\u003e\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e\u003c/sup\u003e using standard settings to pinpoint the most conserved regions, which were subsequently used for primer design. DNA probe preparation and labelling followed the published\u003csup\u003e\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e\u003c/sup\u003e. For satellites with longer monomers, PCR primers were designed to face outward from the monomer; therefore, PCR amplification was performed only between monomers tandemly arrayed. For retrotransposons, PCR primers were designed to the GAG domain which is generally the most variable domain among different retrotransposon families. PCR products were purified using NucleoSpin Gel and PCR Clean-up kit (Macherey-Nagel) and labeled by nick translation. FISH was performed as described previously\u003csup\u003e\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e\u003c/sup\u003e. The preparations were photographed using a Zeiss Axioimager Z2 epifluorescence microscope with a CoolCube camera (MetaSystems). Images were acquired separately for all individual fluorochromes using appropriate excitation and emission filters (AHF Analysentechnik). The monochromatic images were pseudocolored, merged and cropped using Photoshop CS (Adobe Systems).\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgments\u003c/h2\u003e \u003cp\u003eWe thank Yile Huang, Hussein Anani, Marc Beringer, Matthias Heuberger and Thomas Wicker for insightful discussions during this work. This research was funded by the Czech Science Foundation (project no. 21-07748L to MAL), the Masaryk University Grant Agency (project no. MUNI/R/1268/2022 to TM) and the Swiss National Science Foundation (Grants 31003A_178938 and 310030L_197839 to CP).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eTenaillon MI, Hollister JD, Gaut B (2010) A triptych of the evolution of plant transposable elements. Trends Plant Sci 15:471\u0026ndash;478\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar A, Bennetzen JL (1999) Plant retrotransposons. Annu Rev Genet 33:479\u0026ndash;532\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLarrimore KE, Rancati G (2019) The conditional nature of gene essentiality. Curr Opin Genet Dev 58\u0026ndash;59:55\u0026ndash;61\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchubert I, Lysak MA (2011) Interpretation of karyotype evolution should consider chromosome structural constraints. Trends Genet 27:207\u0026ndash;216\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiao WB, Schneeberger K (2020) Chromosome-level assemblies of multiple Arabidopsis genomes reveal hotspots of rearrangements with altered evolutionary dynamics. Nat Commun 11:1\u0026ndash;10\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStein JC et al (2018) Genomes of 13 domesticated and wild rice relatives highlight genetic conservation, turnover and innovation across the genus Oryza. \u003cem\u003eNat. Genet.\u003c/em\u003e 50, 285\u0026ndash;296 (2018)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFedoroff NV (2012) Transposable elements, epigenetics, and genome evolution. Science 338:758\u0026ndash;767\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJoly-Lopez Z, Bureau TE (2018) Exaptation of transposable element coding sequences. Curr Opin Genet Dev 49:34\u0026ndash;42\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHollister JD et al (2011) Transposable elements and small RNAs contribute to gene expression divergence between Arabidopsis thaliana and Arabidopsis lyrata. \u003cem\u003eProc. Natl. Acad. Sci. U. S. A.\u003c/em\u003e 108, 2322\u0026ndash;2327\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang H, Lang Z, Zhu JK (2018) Dynamics and function of DNA methylation in plants. Nat Rev Mol Cell Biol 19:489\u0026ndash;506\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoi JY, Purugganan MD (2018) Evolutionary epigenomics of retrotransposon-mediated methylation spreading in Rice. Mol Biol Evol 35:365\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWyler M, Stritt C, Walser JC, Baroux C, Roulin AC (2020) Impact of transposable elements on methylation and gene expression across natural accessions of Brachypodium distachyon. Genome Biol Evol 12:1994\u0026ndash;2001\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeumann P, Nov\u0026aacute;k P, Hošt\u0026aacute;kov\u0026aacute; N, Macas J (2019) Systematic survey of plant LTR-retrotransposons elucidates phylogenetic relationships of their polyprotein domains and provides a reference for element classification. Mob DNA 10:1\u0026ndash;17\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeumann P et al (2011) Plant centromeric retrotransposons: A structural and cytogenetic perspective. Mob DNA 2:1\u0026ndash;16\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang L et al (2023) A near-complete genome assembly of Brassica rapa provides new insights into the evolution of centromeres. Plant Biotechnol J 21:1022\u0026ndash;1032\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHenikoff S, Ahmad K, Malik HS (2001) The centromere paradox: stable inheritance with rapidly evolving DNA. Science 293:1098\u0026ndash;1102\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWlodzimierz P et al (2023) Cycles of satellite and transposon evolution in Arabidopsis centromeres. Nat 618:557\u0026ndash;565\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen J et al (2023) A complete telomere-to-telomere assembly of the maize genome. Nat Genet 55:1221\u0026ndash;1231\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeiser C, Mand\u0026aacute;kov\u0026aacute; T, Arrigo N, Lysak MA, Parisod C (2016) Repeated whole-genome duplication, karyotype reshuffling, and biased retention of stress-responding genes in Buckler Mustard. Plant Cell 28:17\u0026ndash;27\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeringer M et al (2024) Biased retention of environment-responsive genes following genome fractionation. Mol Biol Evol 41:msae155\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGr\u0026uuml;nig S, Patsiou T, Parisod C (2024) Ice age-driven range shifts of diploids and expanding autotetraploids of Biscutella laevigata within a conserved niche. New Phytol 244:1616\u0026ndash;1628\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHohmann N, Wolf EM, Lysak MA, Koch MA (2015) A time-calibrated road map of Brassicaceae species radiation and evolutionary history. Plant Cell 27:2770\u0026ndash;2784\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo X et al (2021) Linked by ancestral bonds: Multiple whole-genome duplications and reticulate evolution in a Brassicaceae tribe. Mol Biol Evol 38:1695\u0026ndash;1714\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang W et al (2021) The chromosome-level genome sequence and karyotypic evolution of Megadenia pygmaea (Brassicaceae). Mol Ecol Resour 21:871\u0026ndash;879\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLysak MA, Mand\u0026aacute;kov\u0026aacute; T, Schranz ME (2016) Comparative paleogenomics of crucifers: ancestral genomic blocks revisited. Curr Opin Plant Biol 30:108\u0026ndash;115\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchranz ME, Lysak MA, Mitchell-Olds T (2006) The ABC\u0026rsquo;s of comparative genomics in the Brassicaceae: building blocks of crucifer genomes. Trends Plant Sci 11:535\u0026ndash;542\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJia J et al (2021) Homology-mediated inter-chromosomal interactions in hexaploid wheat lead to specific subgenome territories following polyploidization and introgression. Genome Biol 22:1\u0026ndash;21\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMascher M et al (2017) A chromosome conformation capture ordered sequence of the barley genome. Nature 544:427\u0026ndash;433\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTiang CL, He Y, Pawlowski WP (2012) Chromosome organization and dynamics during interphase, mitosis, and meiosis in plants. Plant Physiol 158:26\u0026ndash;34\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShan W, Kubov\u0026aacute; M, Mand\u0026aacute;kov\u0026aacute; T, Lysak MA (2021) Nuclear organization in crucifer genomes: nucleolus-associated telomere clustering is not a universal interphase configuration in Brassicaceae. Plant J 108:528\u0026ndash;540\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuesneville H (2020) Twenty years of transposable element analysis in the Arabidopsis thaliana genome. Mob DNA 11:1\u0026ndash;13\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHlouškov\u0026aacute; P, Mand\u0026aacute;kov\u0026aacute; T, Pouch M, Tr\u0026aacute;vn\u0026iacute;ček P, Lysak MA (2019) The large genome size variation in the Hesperis clade was shaped by the prevalent proliferation of DNA repeats and rarer genome downsizing. Ann Bot 124:103\u0026ndash;120\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChandrasekhara C, Mohannath G, Blevins T, Pontvianne F, Pikaard CS (2016) Chromosome-specific NOR inactivation explains selective rRNA gene silencing and dosage control in Arabidopsis. Genes Dev 30:177\u0026ndash;190\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSharma A, Wolfgruber TK, Presting GG (2013) Tandem repeats derived from centromeric retrotransposons. BMC Genomics 14:1\u0026ndash;11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBelyayev A et al (2020) Transposons and satellite DNA: On the origin of the major satellite DNA family in the Chenopodium genome. Mob DNA 11:1\u0026ndash;10\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWicker T et al (2007) A unified classification system for eukaryotic transposable elements. Nat Rev Genet 8:973\u0026ndash;982\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLynch M, Walsh B (2007) \u003cem\u003eThe origins of genome architecture\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrookfield JF (2005) Y. The ecology of the genome \u0026mdash; mobile DNA elements and their hosts. Nat Rev Genet 6:128\u0026ndash;136\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLian Q et al (2024) A pan-genome of 69 Arabidopsis thaliana accessions reveals a conserved genome structure throughout the global species range. Nat Genet 56:982\u0026ndash;991\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSigman MJ, Slotkin RK (2015) The first rule of plant transposable element silencing: Location, location, location. Plant Cell 28:304\u0026ndash;313\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKiefer C et al (2019) Interspecies association mapping links reduced CG to TG substitution rates to the loss of gene-body methylation. Nat Plants 5:846\u0026ndash;855\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDriguez P, Carty K, Putra A, Ermini L (2021) Workflow for generating HMW plant DNA for third generation sequencing with high N50 and high accuracy. \u003cem\u003eprotocols.io\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003edoi:dx.doi.org/10.17504/protocols.io.bafmibk6\u003c/span\u003e\u003cspan address=\"doi:10.17504/protocols.io.bafmibk6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrueger F, James F, Ewels P, Afyounian E (2021) \u0026amp; Schuster-Boeckler, B. FelixKrueger/TrimGalore: v0.6.7. \u003cem\u003eZenodo\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5281/ZENODO.5127899\u003c/span\u003e\u003cspan address=\"10.5281/ZENODO.5127899\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu J et al (2024) NextDenovo: an efficient error correction and accurate assembly tool for noisy long reads. Genome Biol 25:1\u0026ndash;19\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu J, Fan J, Sun Z, Liu S (2020) NextPolish: a fast and efficient genome polishing tool for long-read assembly. Bioinformatics 36:2253\u0026ndash;2255\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi H (2018) Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics 34:3094\u0026ndash;3100\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePryszcz LP, Gabald\u0026oacute;n T (2016) Redundans: an assembly pipeline for highly heterozygous genomes. Nucleic Acids Res 44:e113\u0026ndash;e113\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRhie A, Walenz BP, Koren S, Phillippy AM, Merqury (2020) Reference-free quality, completeness, and phasing assessment for genome assemblies. Genome Biol 21:1\u0026ndash;27\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi H (2013) Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou C, McCarthy SA, Durbin R (2023) YaHS: yet another Hi-C scaffolding tool. Bioinformatics 39\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurand NC et al (2016) Juicer provides a one-click system for analyzing loop-resolution Hi-C experiments. Cell Syst 3:95\u0026ndash;98\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurand NC et al (2016) Juicebox provides a visualization system for Hi-C contact maps with unlimited zoom. Cell Syst 3:99\u0026ndash;101\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOu S et al (2019) Benchmarking transposable element annotation methods for creation of a streamlined, comprehensive pipeline. Genome Biol 20:1\u0026ndash;18\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerardini TZ et al (2015) The Arabidopsis information resource: Making and mining the \u0026lsquo;gold standard\u0026rsquo; annotated reference plant genome. Genesis 53:474\u0026ndash;485\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKagale S et al (2014) Polyploid evolution of the Brassicaceae during the Cenozoic era. Plant Cell 26:2777\u0026ndash;2791\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmit A, Hubley R, Green P (2015) RepeatMasker Open-4.0\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrabherr MG et al (2011) Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol 29:644\u0026ndash;652\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDobin A et al (2013) STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29:15\u0026ndash;21\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatro R, Duggal G, Love MI, Irizarry RA, Kingsford C (2017) Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods 14:417\u0026ndash;419\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCantarel BL et al (2008) An easy-to-use annotation pipeline designed for emerging model organism genomes. Genome Res 18:188\u0026ndash;196\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrůna T, Hoff KJ, Lomsadze A, Stanke M, Borodovsky M (2021) BRAKER2: automatic eukaryotic genome annotation with GeneMark-EP\u0026thinsp;+\u0026thinsp;and AUGUSTUS supported by a protein database. NAR Genomics Bioinforma 3:1\u0026ndash;11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGabriel L et al (2024) BRAKER3: Fully automated genome annotation using RNA-seq and protein evidence with GeneMark-ETP, AUGUSTUS, and TSEBRA. Genome Res 34:769\u0026ndash;777\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBateman A et al (2023) UniProt: the Universal Protein Knowledgebase in 2023. Nucleic Acids Res 51:D523\u0026ndash;D531\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang S, Lomsadze A, Borodovsky M (2015) Identification of protein coding regions in RNA transcripts. Nucleic Acids Res 43:e78\u0026ndash;e78\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGabriel L, Hoff KJ, Brůna T, Borodovsky M, Stanke M (2021) TSEBRA: transcript selector for BRAKER. BMC Bioinformatics 22:1\u0026ndash;12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Z et al (2022) Improved reference genome annotation of Brassica rapa by Pacific Biosciences RNA sequencing. Front Plant Sci 13:841618\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang R et al (2013) The reference genome of the halophytic plant Eutrema salsugineum. Front Plant Sci 4:45219\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBao W, Kojima KK, Kohany O (2015) Repbase Update, a database of repetitive elements in eukaryotic genomes. Mob DNA 6:1\u0026ndash;6\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKorf I (2004) Gene finding in novel genomes. BMC Bioinformatics 5:1\u0026ndash;9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu TD, Watanabe CK (2005) GMAP: a genomic mapping and alignment program for mRNA and EST sequences. Bioinformatics 21:1859\u0026ndash;1875\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePertea M, Pertea G (2020) GFF Utilities: GffRead and GffCompare. \u003cem\u003eF1000Research\u003c/em\u003e 9, 304\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQiao X et al (2019) Gene duplication and evolution in recurring polyploidization-diploidization cycles in plants. Genome Biol 20:38\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLovell JT et al (2022) GENESPACE tracks regions of interest and gene copy number variation across multiple genomes. Elife 11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoel M, Sun H, Jiao WB, Schneeberger K (2019) SyRI: finding genomic rearrangements and local sequence differences from whole-genome assemblies. Genome Biol 20:1\u0026ndash;13\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoel M, Schneeberger K (2022) plotsr: visualizing structural similarities and rearrangements between multiple genomes. Bioinformatics 38:2922\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSamarakoon H et al (2023) Flexible and efficient handling of nanopore sequencing signal data with slow5tools. Genome Biol 24:1\u0026ndash;12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGamaarachchi H et al (2020) GPU accelerated adaptive banded event alignment for rapid comparative nanopore signal analysis. BMC Bioinformatics 21:1\u0026ndash;13\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCheetham SW, Kindlova M, Ewing AD (2022) Methylartist: tools for visualizing modified bases from nanopore sequence data. Bioinformatics 38:3109\u0026ndash;3112\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMand\u0026aacute;kov\u0026aacute; T, Lysak MA (2016) Chromosome preparation for cytogenetic analyses in Arabidopsis. Curr Protoc Plant Biol 1:43\u0026ndash;51\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi W, Godzik A (2006) Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22:1658\u0026ndash;1659\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSievers F et al (2011) Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMand\u0026aacute;kov\u0026aacute; T, Lysak MA (2016) Painting of Arabidopsis chromosomes with chromosome-specific BAC clones. Curr Protoc plant Biol 1:359\u0026ndash;371\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5461468/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5461468/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eTransposable elements (TEs) play pivotal roles in genome evolution, yet their impact on pericentromeric regions of chromosomes, characterized by high sequence turnover and TE abundance, remains largely unclear. This gap in knowledge limits our understanding of TEs biology and their role within host genomes. In this study, we address this gap by analysing chromosome-scale assemblies to explore the content and dynamics of pericentromeric regions in four closely related \u003cem\u003eBiscutella\u003c/em\u003e species. Although they share substantial synteny, we observe significant variability in the non-coding genome, especially within pericentromeric regions of the species affected by strongest genetic drift due to smallest population size. By comparing TEs from the CRM clade, which specifically target centromeric regions, with those from the Athila clade, we identify specialized CRMs that follow centromeres after recent repositioning, alongside an invasion by Athila copies that exhibit less insertion bias. Additionally, we find that TEs migration from pericentromeric towards distal nucleolus organizer regions correlates with increased DNA methylation and decreased gene expression. These results highlight how rapid pericentromeric evolution driven by bursts of TE activity can significantly impact genome functionality and stability. Our findings offer new insights into the evolutionary mechanisms shaping genome organization and underscore the broader implications for understanding genome dynamics and adaptation.\u003c/p\u003e","manuscriptTitle":"The role of centromeric transposable elements in shaping chromosome evolution","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-02-03 05:59:12","doi":"10.21203/rs.3.rs-5461468/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"2090c557-6940-4337-a95c-c2b8cc0ccce5","owner":[],"postedDate":"February 3rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":40638949,"name":"Biological sciences/Genetics/Genomics"},{"id":40638950,"name":"Biological sciences/Plant sciences/Plant genetics"}],"tags":[],"updatedAt":"2026-02-17T19:25:03+00:00","versionOfRecord":[],"versionCreatedAt":"2025-02-03 05:59:12","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5461468","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5461468","identity":"rs-5461468","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}