TAD conservation in vertebrate genomes is driven by stabilising selection.

TAD conservation in vertebrate genomes is driven by stabilising selection. | 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 Research Article TAD conservation in vertebrate genomes is driven by stabilising selection. Fabiana Patalano, Simen Rød Sandve, Rein Aasland, Jonas Paulsen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5809117/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Background Topologically Associating Domains (TADs) are fundamental structural and gene regulatory components of chromatin defined by regions of high intra-domain contact frequency. Though TADs are found across diverse metazoans, the extent of their evolutionary conservation is still debated. Some studies indicate significant conservation among closely related species, while others suggest considerable variability, raising questions about the evolutionary forces that preserve TAD organization. Results Here, we investigated the evolutionary conservation of TADs by analysing Hi-C data from 12 vertebrate species. We examined TAD numbers, borders, and gene positioning within TADs. We found that TAD’s features are all highly conserved across species, but this conservation tends to decrease with evolutionary distance. Closely related species show greater TAD conservation compared to more distantly related species. Yet, modelling the divergence in TAD conservation using Ornstein-Uhlenbeck (OU) process revealed considerable selective pressures on TAD number within syntenic blocks, suggesting that TAD features in these genomic regions are under stabilising selection. However, we also identified a small subset of blocks where TAD numbers evolve under genetic drift, highlighting the existence of distinct groups of blocks subject to different evolutionary dynamics. Conclusions These findings improve our understanding of TAD conservation and evolution, revealing significant conservation of TAD features, especially among closely related species. We discovered that TADs evolve both under stabilizing selection and genetic drift, highlighting the complex evolutionary dynamics of TAD evolution. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background In eukaryotes, genomic DNA is tightly packed with histones and non-histone proteins to form chromatin, which serves to protect and regulate genetic material while efficiently organising the long DNA molecules within the limited space of the nucleus. This organisation is hierarchical, with structural and regulatory processes working together to ensure genome function and stability ( 1 ). A key feature of this dynamic organisation is the formation of Topologically Associating Domains (TADs), sub-megabase-scale chromatin segments characterised by high intra-domain contact frequencies ( 2 – 5 ). TADs typically contain genes interacting with regulatory elements within the same domain ( 6 ). Interactions across TADs are restricted by border elements enriched with insulator proteins, such as CCCTC-binding factor (CTCF). The proposed mechanism for TAD formation involves loop extrusion, where cohesin interacts with CTCF, halting when cohesin encounters convergently orientated CTCF-bound sites ( 2 , 7 – 11 ). While first discovered in mammals, TADs have since been observed in many species, from Drosophila to prokaryotes and in plants ( 4 , 12 – 14 ). Vertebrates rely on CTCF for TAD borders, while other species use alternative proteins, suggesting independent evolution of some TAD mechanisms ( 15 ). Although significant progress has been made in understanding properties of TADs, the precise mechanisms underlying their evolutionary origins and conservation are not known. In particular, the degree of TAD evolutionary conservation is still debated. While some evidence supports a considerable TAD conservation ( 2 , 16 – 18 ), recent direct comparisons of TAD borders have reported a low level of conservation, challenging this notion ( 19 , 20 ). Growing evidence, direct and indirect, suggests that TAD borders could play a key role as functional elements. Indirect evidence of TAD conservation comes from studies examining genomic synteny breakpoints and the occurrence of CTCF-binding sites across species. Recent studies have revealed that synteny breakpoints in various mammals ( 17 , 18 , 21 ) and flies ( 22 ) are concentrated at TAD borders and depleted within TAD bodies, suggesting TADs are conserved during evolution ( 17 , 18 ). The conservation of TADs has also been inferred through comparative analyses of CTCF-binding sites and insulator activity across species ( 16 ), although this study does not report the number or proportion of conserved TADs observed ( 16 ). Direct evidence of TAD conservation comes from several studies that compare TAD borders across different species. Initially, TADs were directly compared between human and mouse revealing that approximately 54% of human TAD borders are conserved in mouse ( 2 ). Borders conservation was defined based on the presence or absence of borders across species. Using a similar approach with more stringent criteria, a comparison of human and chimpanzee TAD borders found that only 43% are shared between the two species ( 24 ). In a more recent study applying the same presence/absence criterion, it was estimated that only 14% of human and 15% of mouse TAD borders are conserved across 8 primate and rodent species ( 20 ), challenging earlier reports of TAD border conservation. This approach relies on whole genome alignment and liftover processes, which becomes progressively more difficult as evolutionary distance increases. Hence, when comparing multiple distantly related species, approaches that do not rely on base-pair-level orthology is necessary. In this context synteny becomes a useful framework. Synteny (meaning “same ribbon” as introduced by Renwick ( 25 )) and refers to the preserved order of genes or genomic segments across species. As gene order is preserved across much larger evolutionary distances, syntenic blocks therefore represent a practical and evolutionary meaningful unit to explore patterns of TAD conservation. Beyond gene order, some genomic regions exhibit strong structural and functional conservation, in particular some Genomic Regulatory Blocks (GRBs)—large, evolutionarily constrained regions housing key developmental genes alongside highly conserved noncoding elements (CNEs) ( 26 – 28 ). The conservation of GRBs within syntenic blocks suggests that selection acts not only to maintain individual genes and their regulatory elements but also to preserve the broader genomic architecture that enables their coordinated function. The organization of GRBs closely aligns with TADs, which form chromatin loops through cohesin and CTCF interactions, insulating regulatory landscapes and maintaining proper gene expression ( 27 , 29 ). The overlap between GRBs, synteny, and TADs suggests that TADs may act as structural scaffolds, preserving cis-regulatory interactions and ensuring the proper regulation of developmental genes across evolutionary timescales. In this study we quantify the evolutionary conservation of TADs in a broader spectrum of vertebrate species using several TAD characteristics, such as TAD number, borders, and gene positioning within TADs. We show, using an Ornstein-Uhlenbeck model, that TAD-conservation is in many cases driven by stabilizing selection. Results Identification of syntenic blocks across 12 species. To systematically explore TAD evolution, we compiled a dataset across the vertebrate lineage, including 12 species – 10 mammalian species, chicken and zebrafish (Fig. 1A). All species feature high-quality genome assemblies and available Hi-C data obtained from fibroblast cell lines. While this dataset provides valuable insight into TAD evolution within vertebrates, it is important to note that our coverage is limited, largely due to the availability of Hi-C data. Similar studies have used syntenic-block-based strategies to enhance comparative TAD analyses (2,16,24). Following this approach, we analysed TADs within syntenic blocks to ensure meaningful cross-species comparisons while minimizing alignment artifacts that increase with evolutionary distance. This method also allows for a more comprehensive investigation of TAD configurations by going beyond isolated TADs, specifically examining how TADs are organized in relation to each other, including their spatial arrangement. For each ancestral node in the phylogenetic tree, we identified conserved syntenic blocks—genomic segments with preserved order of genes across species evolving from that node—using Cyntenator (30). A synteny block consists of at least two orthologous genes across genomes. The number of syntenic blocks varies, ranging from 100 between humans and rhesus monkeys to 651 when including all mammals and chicken (Fig. 1B). As expected, the evolutionary distance between species impacts the distribution of gene counts per block; for example, the rodent’s alignment (13 million years divergence) shows up to 1,767 genes per block (median ± stdev: 31±369), while alignments across all species (429 million years divergence) reveal a maximum of only 16 genes (median ± stdev: 3±2.27). Closely related species exhibit narrower gene count distributions, with an interquartile range of 144 for rodents compared to just 2 for the vertebrate’s alignment. Alignment coverage decreases with evolutionary distance, from over 90% to under 20% (additional details can be found in Table S1). The reduction of 4194 genes within syntenic blocks when going from tetrapods to vertebrates (Table S2) may result from the additional whole-genome duplication in teleost fish ~200 million years ago, followed by selective gene retention or loss (31). Analysis of these "missing" genes reveals that 34% (1425 genes) exhibit over 70% identity with non-syntenic paralogs, suggesting their relocation to non-syntenic regions rather than their complete absence from the zebrafish genome. Nevertheless, the percentage drops to 15% for genes sharing over 80% identity. Gene Ontology (GO) analysis of genes conserved across all 12 species showed significant (p.value <0.05) enrichment in processes crucial for development and cellular regulation. Key enriched terms include organ development and anterior-posterior pattern specification (Figure S1). Identification of TADs in the vertebrate fibroblast genomes. We then performed 3D genome analysis on publicly available Hi-C data (GEO: GSE167581, SRA IDs: PRJNA482496, (31,32)) of fibroblast cells. Our study included 12 Hi-C libraries, which we merged after iterative correction and KR normalisation. This process resulted in an average of 295 million valid reads per species, with the human dataset containing the highest number of valid reads (568 million), while the zebrafish dataset had the lowest (36.6 million) (Figure S2). Using HiCExplorer (33), we identified an average of 3,278 TADs across the 12 vertebrate genomes (mean ± SD: 3,28 ± 818). The identified TADs ranged from 1,207 in zebrafish to 4,037 in rhesus (Figure S3A). This variation in TAD numbers across species may be attributed to differences in Hi-C resolution. The presence of TADs within syntenic blocks differed greatly between species, as expected, since blocks cover less of the genome, with the highest inclusion rate of 99% observed in the common ancestor of humans and rhesus macaques and the lowest rate of 12% between humans and zebrafish (Table S3, Figure S3B). TAD number is conserved across vertebrate evolution. We analysed the rate of evolutionary change in TAD features across species, comparing pairs of species without using human as a reference. This approach allowed us to assess TAD conservation independently across species with similar divergence times, such as mouse, cow, and dog, which diverged approximately 70–100 million years ago. Analyses of evolutionary conservation of TAD numbers within syntenic blocks revealed an average change of 0.018 TADs per megabase pair (Mbp) per million years (Mbp/Myr) (Figure S4). Note that these change estimates assume a somewhat unrealistic linear relationship (34), but nevertheless provide an indication of the extent of TAD change over time. TAD number conservation decreases as evolutionary distance between species increases. This is exemplified also in Figure 2A, where species sharing a more recent common ancestor, such as the mouse and rat, show an identical count of TADs within syntenic blocks in contrast to the more distantly related chicken, which does not. To assess conservation and statistical significance across all alignments, we computed the Average Relative Difference (ARD), which measures the proportional mean difference in TAD numbers, with lower values indicating stronger conservation. This provides insights into the stability of TAD numbers and helps evaluate conservation across varying evolutionary distances. To evaluate statistical significance, we compared our scores to a random distribution generated by shuffling the TAD counts within the syntenic blocks 1000 times. This method provides a baseline to determine if the observed conservation is significantly greater than what would be expected by chance. Our findings revealed that TAD number conservation is significantly higher than would be expected by chance (p.value <0.05, Figure S5). The conservation we observed ranged from 0.22 to 0.69-fold less than the random distribution (Figure 2B). Taking primates as an example, the TAD conservation was nearly0.2 times lower than the random. This pattern might suggest a strong phylogenetic signal, reflecting a non-random distribution of TAD number among these species. Divergence time correlates with TAD border position conservation in vertebrates. To determine if TAD border positions are evolutionarily conserved within syntenic blocks, we selected blocks with an equal number of TADs, averaging 40 blocks per species pair comparison. Avoiding lift-over, which loses reliability with increasing evolutionary distance, we normalized TAD border positions to a 0 to 1 scale within each syntenic block (Fig. 3A). We computed TAD border position conservation using the mean absolute differences in these relative border locations across all pairs of species. Dividing the species into three distinct clades (mammals, tetrapods and vertebrates) revealed an average difference in border position of ~9.2% within mammals, 10.7% in tetrapods, and ~14.2% in vertebrates (Figure S6). To exemplify this, we assessed mean absolute differences in the SIX2/3 cluster, a region known to be TAD-regulated in zebrafish (27). In this cluster all the border positions are conserved across different species pairs, and the absolute difference between pairs consistently stays under 0.2 (Figure S7), providing confidence to this metric. To further compare relative TAD border positions between species pairs, we calculated the R² score, which measures the proportion of variance in the border positions of one species that can be explained by the positions in another species (Figure 3A and B). Applying this analysis to the entire dataset we find that relative border positions have a statistically significant correlation for all clades (Table S4), yet with a gradually decreasing R² values over evolutionary time (Figure 3C). Additionally, we analysed the border outliers in primate alignments and found that, even though they are not statistically enriched in any GO term, these genes are involved in processes such as glucose response and immune response (Figure S8). Gene position is conserved within TADs. Because TADs serve to restrict gene regulatory interactions (6), we next investigated whether genes tend to stay within the same TADs over evolutionary time. To quantify this, we computed the average number of genes found within the same TAD across pairs of species within syntenic blocks. This revealed that mammalian synteny blocks have an average of 9.7 orthologous genes per Mbp/Myr. Given the average density of ~14 genes per Mbp in these blocks overall, this indicates a relatively high degree of gene content conservation within TADs. Similarly, for tetrapods, we observe a conservation of 9.3 genes per Mbp/Myr, yet for the whole vertebrate clade the conservation score drops to about 5 genes per Mbp per million years, reflecting a more substantial divergence in gene positional conservation as the evolutionary distance increases (Figure S9). To test if these levels of conservation were higher than expected by chance alone, we shuffled gene positions within each block 1000 times preserving the number of genes per block and the number of TADs, providing a baseline for comparison with our observed gene conservation scores. This revealed that genes generally maintained their spatial positioning across species, exceeding random expectation by factors ranging from 1.1 to 2.2 (Fig. 4A), with a p.value <0.05 (see Figure S10). For instance, in primates, gene positional conservation within TADs was observed to be 2 times higher than by chance (Fig. 4A), yet with a decreasing ratio over evolutionary time. Next, we expanded these analyses to include syntenic blocks regardless of the number of TADs they contained. For this, we computed the theoretical number of TAD borders insertions or deletions needed to ensure that orthologous genes resided within the same TADs between a pair of species. The results show that the average number of required edits (i.e. insertions or deletions) is 0.8 borders/Mbp for mammals, increasing to 1.02 borders/Mbp in tetrapods and 1.08 borders/Mbp in vertebrates (Figure S11). The increasing number of required edits to convert one conformation into another correlates with the increasing evolutionary divergence between species. Computing the required border edits relative to shuffled genes (see above), revealed that all compared clades show a significantly lower number of border edits than expected by chance (Fig. 4B, Figure S12). Most syntenic blocks show stabilizing selection while a few evolve under genetic drift. All analyses so far revealed a significant conservation of TAD features, yet with a strong phylogenetic signal. Hence, to assess whether TAD conservation is driven by stabilizing selection or evolving under genetic drift, requires us to apply comparative phylogenetic models which can distinguish between these evolutionary forces. Previous studies of conservation of gene expression (35) have made use of the Ornstein-Uhlenbeck (OU) process, which is a widely accepted model of continuous trait evolution (35–37). To analyse the evolutionary dynamics of TAD numbers within syntenic blocks, we used the OU model, which quantifies the influence of both random genetic drift and stabilizing selection. It includes two parameters: alpha (𝛼), indicating the strength of selection pressure, and sigma (𝜎), representing the rate of genetic drift. A larger 𝛼 value suggests strong stabilizing selection that pulls a trait towards an optimal value, while a larger 𝜎 value points to greater variability due to random changes. By applying the OU model, we assessed how TAD number evolution is influenced by these factors in mammals. We focused on mammals in our analysis due to the availability of Hi-C data for these species, which provided a more robust dataset for our study. The representation of other vertebrate species in the dataset is limited, with only two species from outside the mammalian group. Including these outgroup species would have further reduced the number of genomic regions available for analysis, limiting the statistical power of our results. We performed a grid search across a range of 𝛼 and 𝜎 values, selecting the best-fitting parameters based on log-likelihood (see Methods). As 𝛼 approaches zero, the process mimics Brownian motion, pointing to genetic drift, while higher 𝛼 values suggest the presence of selective pressures. The cumulative log-likelihood results show a range of optimal 𝛼 and 𝜎 parameters, primarily concentrated in regions with large 𝛼 values (Figure S13), indicating significant selective pressure on many syntenic blocks. A smaller peak at low 𝛼 values indicates that some TADs may be evolving under near-neutral conditions. These patterns highlight the diverse evolutionary dynamics across different syntenic blocks (Figure 5A). Our analysis identified the most frequent parameter combination as 𝛼 = 5.2 and 𝜎 = 2.11 (Figure S14), suggesting strong selection and low variability due to genetic drift, suggesting that stabilizing selection plays an important role in conserving TAD numbers over evolutionary time. Interestingly, 4.12% syntenic blocks displayed 𝛼<0.01, indicating that their evolutionary trajectories are more affected by genetic drift. Performing GO-analysis on blocks with low 𝛼 (𝛼 = 0.01) reveals terms like collagen catabolic processes (adjusted P = 0.02; Figure S15A), extracellular matrix disassembly (adjusted P = 0.02) and regulation of neuroinflammatory response (adjusted P = 0.02).The majority of blocks (95.88% of blocks) are concentrated at higher 𝛼 values (𝛼>0.01), indicating strong selection on TAD numbers (Figure 5B). High-𝛼 blocks are associated with various processes like intermediate filament organization, organic acid transport and glycerophospholipid metabolic process (Figure S15B). Noteworthy terms are also those related to development, including pattern specification, limb development, and gliogenesis. Many syntenic blocks under stabilizing selection (high 𝛼 values, ~8000 genes) contain developmental genes, while fewer blocks evolving through genetic drift have diverse, non-developmental functions. The presence of developmental genes in our high-𝛼 blocks suggests that regulatory regions surrounding these genes could be relevant for the conservation of TAD structure. Highly conserved noncoding elements (CNEs) are known to surround developmental genes in vertebrates (38). Clusters of these CNEs, termed genomic regulatory blocks (GRBs) (28) have been shown to overlap with TAD structure, suggesting that TADs are needed to maintain the regulatory architecture in these regions (27). To investigate whether our synteny blocks associate with GRBs, we overlapped them with GRBs from reference (27). While GRB-containing syntenic blocks span a range of α values, they do not strongly correlate with α (Figure 5B). Our low-α blocks generally show a less conserved TAD border positioning compared to high-α blocks, confirming that values of α reflect levels of TAD border position conservation (Figure 5C). Notably, GRB-containing blocks tend to have even more stable TAD border positions compared to high 𝛼 blocks in general (Figure 5C). GO term analysis reveals that GRB-containing low-α blocks are not enriched for a particular process (Figure S15C; adjusted p.value>0.35), and high α blocks are mainly enriched with various developmental processes (adjusted p.value<2.52e-04; Figure S15F). This confirms that the subset of GRBs with conserved TAD structures indeed harbours genes with developmental functions. Genes in non-GRB low α blocks, on the other hand, are enriched in processes related to collagen metabolism (adjusted p.value<2.43e-05; Figure S15C), whereas non-GRB high α blocks are not enriched for particular processes (Figure S15D). This shows that syntenic blocks with conserved TAD numbers (high α) also harbour genes with non-developmental functions, especially if they do not harbour GRBs. Comparing values of 𝛼 with average number of genes and TADs within our blocks reveals that low 𝛼 blocks have a higher gene and TAD density than intermediate 𝛼 values (0.09-6.8). For blocks with 𝛼>8.4, gene and TAD density is the highest (Table S5). Consistent with prior studies (27), TADs overlapping GRBs have lower gene density (e.g., 12.60 genes/Mbp in humans vs. 20.82 genes/Mbp in non-GRB TADs, Table S6). However, protein sequence identity between species remains comparable across low- and high-α blocks (Figure S16). Overall, our analyses reveal that most syntenic blocks are governed by stabilizing selection of their TADs. A subset of these which harbour GRBs are associated with developmental genes, consistent with prior studies. However, we also identify a large proportion of non-GRB blocks with conserved TAD structure harbouring genes with non-developmental functions. Discussion TADs are essential functional units within the genome that play a significant role in regulating gene expression. Disruptions of TAD borders have been linked to abnormal gene interactions, gene mis-regulation, and various aberrant phenotypes, including cancer ( 39 ). Additionally, such disruptions can lead to developmental disorders, such as human limb malformations ( 40 ). Our study expands the current understanding of evolutionary conservation of TADs across a range of vertebrate species. We have extended the analysis of evolutionary conservation of TADs across 12 vertebrate species, using a comprehensive quantitative framework to assess TAD-feature conservation, including the use of phylogenetic models to detect signatures of selection on TADs. Our approach integrates both a general assessment of TAD conservation across species and a focused analysis of individual blocks using the OU-model. The initial broad analysis provides essential context for understanding overall conservation patterns, while the OU-model allows us to examine the evolutionary conservation of TAD numbers in greater detail. Early studies on TAD conservation primarily relied on indirect methods or limited locus analyses, such as the Hox ( 41 ) and Six ( 42 ) clusters. These investigations suggested TAD conservation based on patterns like the enrichment of rearrangement breakpoints at TAD borders and their depletion within TAD bodies across species ( 17 , 18 ). Previous studies in vertebrates have reported various levels of TAD-conservation, ranging from 54 − 14% across mammalian lineages of different evolutionary divergence levels ( 2 , 19 , 20 ). Our findings indicate that conservation of TAD number, border positions, and gene positioning within TADs display a strong phylogenetic signal (Fig. 2 A, 3 A, 4 ). Comparing our findings directly with previous studies is challenging largely due to the lack of standardized methodologies for defining and computing TAD conservation. For instance, while some studies inferred conservation through CTCF-binding sites ( 16 ), without a formal testing against a robust null model, others focused on broader comparisons not limited to syntenic regions. These methodological differences, combined with variability in data quality, and in TAD calling algorithms ( 43 ), limit direct comparisons of conservation values between different studies. Nevertheless, our findings point towards conservation of TAD features during vertebrate evolution. TADs play a crucial role in shaping the regulatory landscapes of developmental genes ( 29 ). Their stability, size variation, and conservation across metazoans suggest strong selective pressures, particularly in cases where TADs coincide with GRBs in both humans and Drosophila ( 27 ). The alignment of TADs with regulatory elements suggests that they function by restricting the reach of gene regulatory interactions ( 27 , 44 , 45 ). Despite the evidence for evolutionary conservation, previous studies have not addressed the question whether this stability results from selection or genetic drift. In this study, we apply an OU-model with parameters that can be interpreted as the force of stabilizing selection (α) and genetic drift (σ), enabling this distinction ( 28 , 29 ). Our results revealed that a significant majority (95.88%) of syntenic blocks among mammals exhibit α > 0.01 (Fig. 5 B), indicating that they are under strong stabilizing selection. Genes in syntenic blocks with purifying selection pressure on TAD-structure were associated with developmental processes, including embryonic organ development and regionalization. This underscores the importance of conserving TAD numbers for maintaining the precise gene regulatory environment needed for establishing and maintaining organismal structure and function during vertebrate embryogenesis. The absence of enrichment for developmental genes in synteny blocks with low α parameter values reinforces the idea that TADs containing genes critical for development are subjected to stronger selective pressure. Conversely, TADs containing genes involved in more specialized or variable physiological processes may permit more flexibility in their number and arrangement, reflecting different evolutionary dynamics and greater genetic variability, despite the fact that their gene order is conserved. Achieving accurate parameter estimates in OU models can be challenging, especially when tree size is small (< 100) ( 46 ). However, our study does not hinge on individual parameter estimates. Instead, we analyse the distribution of α values across hundreds of synteny blocks, revealing that most synteny blocks exhibit α values much greater than sigma. This strongly suggests the presence of stabilizing selection acting on the TAD number. TADs associated with GRBs tend to exhibit lower gene density and higher border conservation, as developmental TADs are typically gene-poor but rich in intergenic regions that provide regulatory flexibility ( 27 ). In contrast, TADs located further from GRBs often have higher gene density and greater structural variability, likely supporting specialized, less conserved processes ( 27 ). However, GRBs are not the sole factor driving TAD conservation. Some GRB-free TADs remain highly conserved, displaying higher-than-average border similarity and strong stabilizing selection on their number. This suggests that while GRBs contribute to TAD conservation, especially for developmental genes, their presence is not always necessary for maintaining TAD structural integrity. It is crucial to acknowledge that gathering large-scale Hi-C data from the same tissue across multiple species is complex, adding another layer of difficulty to such studies. Our study using data from 12 vertebrate species reveals extensive evolutionary conservation of 3D genome organization across syntenic blocks, and highlights their likely role in gene regulation. Our results underscore the significance of using evolutionary models to study TAD conservation across species. Analysing more species will further enhance our understanding of shared patterns and lineage-specific differences in 3D genome conservation. Moreover, comprehensive mapping of conserved 3D genome structures can pinpoint regions and TADs vulnerable to disease upon disruption. Conclusions Our study enhances the current understanding of TAD conservation across a range of vertebrate species by employing a nuanced approach that includes various TAD features. We found that TAD numbers, border positions, and gene locations exhibit substantial conservation. However, this trend decreases with increasing evolutionary divergence. TAD numbers change at a rate of 0.018 TADs Mbp/Myr, while TAD border positions show a decreasing correlation with evolutionary divergence, increasing from ~9.2% difference within mammals to ~14.2% difference across vertebrates. Gene positions within TADs also show evidence of conservation, with synteny blocks containing an average of 9.7 orthologous genes per Mbp/Myr in mammals and 5 in vertebrates within the same TAD. Using the Ornstein-Uhlenbeck model, we identified that many syntenic blocks are subject to strong stabilizing selection. A subset of these blocks overlaps with GRBs and specifically harbour developmental genes, highlighting the importance of maintaining precise TAD configurations for crucial gene regulatory functions. Conversely, genes involved in specialized functions showed signs of TAD evolution primarily driven by genetic drift, suggesting TAD organization is less critical for these genes. Our findings underscore the evolutionary stability of TADs, while also illustrating the need for standardized methodologies in assessing TAD conservation. Methods Data sets and synteny identification In this study, we analysed 12 different species: human (GRCh38.p14), rhesus macaque (Mmul_10), mouse (GRCm39), rat (mRatBN7.2), rabbit (UM_NZW_1.0), cow (ARS-UCD1.3), sheep (ARS-UI_Ramb_v2.0), pig (Sscrofa11.1), cat (F.catus_Fca126_mat1.0), dog (Dog10K_Boxer_Tasha), chicken (bGalGal1.mat.broiler.GRCg7b), and zebrafish (GRCz11). To investigate the relationship between synteny and 3D chromatin architecture, we first identified syntenic blocks using the Cyntenator tool. The three primary input files for this tool are: Guide Tree: Specifies the order in which the genomes should be aligned. Genome File: Contains coordinates for protein-coding genes. Correspondence File : Details orthologous relationships among genes across all input genomes. The correspondence file was generated by performing an all-versus-all alignment of protein sequences from the twelve species using mmseq2 (47) with an e-value threshold of 10 -5 . Only the longest isoform from each protein sequence was kept. The phylogenetic tree from the TimeTree database served as the guide tree for progressive alignment. Cyntenator uses a stepwise alignment strategy, gradually adding species based on their evolutionary relationships. This method results in alignments that capture syntenic blocks shared among species within a specific node of the guide tree. Essentially, the syntenic blocks identified at each internal node represent genomic regions conserved among species branching from a common ancestor. Processing and Analysis of Hi-C Data. Raw Hi-C sequencing data was obtained from the Gene Expression Omnibus (GEO: GSE167581,SRA IDs: PRJNA482496, (31,32)) and processed using the nf-core/hic (v2.0.0) pipeline (48) to ensure quality control, sequence mapping, and filtering. The reference genomes utilized were the latest versions available as of 2023. The pipeline was executed with HindIII as the restriction enzyme and was configured with the following optional flags: --skip_maps, --skip_dist_decay, --skip_tads, --skip_compartments, --skip_balancing, --skip_mcool, and --split_fastq=false. For creating Hi-C contact matrices, we employed the Cooler software (49) (v0.8.11) to produce matrices in the cooler format. These matrices were subsequently converted to the. mcool file format using the cooler zoomify tool. Matrix balancing was then performed using Iterative Correction (ICE). Data from biological replicates were merged for comprehensive analysis. To identify TAD borders at a 32 kb resolution, we used the hicFindTADs command from HiCExplorer (33) (v3.7.2). To identify TADs located within each syntenic block, we used BEDTools (v2.30.0) (50) intersect . Measuring TAD number conservation across vertebrates. To explore the relationship between synteny and the conservation of TADs throughout vertebrate evolution, we analysed various TAD characteristics across 12 different species. Our initial goal was to assess the conservation of TAD numbers within synteny blocks. First, we identified synteny blocks for each species and counted the TADs within them, normalizing the counts by synteny block lengths in Mbp. We then calculated the absolute differences in these normalized TAD counts between species pairs, adjusting for evolutionary divergence times. By using linear regression to fit these differences against divergence times, we determined the slope, representing the average rate of TAD number change per Mbp per million years. To further quantify the conservation of TAD numbers, we also used a scoring metrics defined: Average Relative Difference (ARD): For each pair of species i and j within a given syntenic block k , the relative difference in TAD counts between species i and j within the block is: The ARD score is the average of these relative differences across all syntenic blocks. This score quantifies the average relative difference in TAD counts between species i and j in the syntenic blocks. We created these randomized distributions by shuffling the order of syntenic blocks 1000 times. For each iteration, we calculated an ARD score, and used the resulting distribution to evaluate the statistical significance of our observed ARD values. Quantifying the evolutionary conservation of TAD borders We next focused on the positioning of TAD borders within synteny blocks to quantify their conservation throughout vertebrate evolution. We normalized the positions of TAD borders within synteny blocks on a scale from 0 to 1. This normalization enabled us to compare TAD positions between different species using a consistent scale, regardless of their specific genomic coordinates. We assessed the changes in TAD borders positions over evolutionary time by calculating the absolute differences between corresponding borders within each synteny block, expressed as changes per million years. The species were grouped into three distinct evolutionary clades: mammals only, mammals and chicken, and a broader group including mammals, chicken, and zebrafish. For each group, we computed the average difference in TAD borders positions and analysed these differences in relation to species divergence. To further evaluate the conservation of TAD borders, we used the R² score. This statistical measure quantifies the extent to which TAD borders in one species can predict corresponding borders in another species. Gene-TAD conformation stability evaluation To evaluate the maintenance of syntenic genes within the same TAD across evolution, we focused on examining the positioning of orthologous syntenic genes within TADs to determine if they exhibit a tendency to reside within the same TAD. We computed the percentage of genes within the same TAD as the observed data, followed by a random permutation of TAD borders to establish a comparative baseline. This score was evaluated only for syntenic blocks with the same number of TADs. To further analyse the differences in TAD configurations between species, we introduced another score-defined Edit Distance. This score evaluates the number of edits—specifically, TAD borders insertions or deletions—needed to transition from one TAD conformation to another. Ornstein-Uhlenbeck (OU) Model In evolutionary biology, modelling how traits evolve within populations over time is essential for understanding adaptation and divergence mechanisms. We employed the OU process to model the evolution of continuous traits along a phylogeny. Specifically, we analyzed the number of TADs in each syntenic block across a mammalian alignment. Each syntenic block was analysed individually to assess how the number of TADs evolves over time in relation to the phylogeny. The OU model is governed by the following stochastic differential equation (SDE): Gene ontology enrichment analysis We performed Gene Ontology (GO) biological process enrichment analysis using the clusterProfiler R package with the enrichGO function (51). Terms with an adjusted p.value < 0.05, calculated using the Benjamini-Hochberg (BH) method, were considered significantly enriched. The top 10 enrichment terms, ranked by their Gene Ratio in descending order, were selected for visualization. GRB-including syntenic blocks identification. We obtained the list of GRBs from (27), which have identified using CNEs showing 70% identity over 50 bp between human and chicken (hg19-galGal4). Their genomic coordinates to the hg38 assembly using the UCSC LiftOver tool (52) with default parameters. To classify syntenic blocks as either GRB-including or non-GRB-including, we assessed their overlap with GRBs using BEDTools (v2.30.0) (50) intersect with no filtering based on the extent of overlap. Identification of non-syntenic paralogy To classify genes into possible non-syntenic paralogs and not, we analysed alignment blocks in the synteny dataset. For each block, we identified human genes absent in vertebrate alignment but present in tetrapods. These genes were checked against sequence similarity data to determine if alternative orthologs existed in zebrafish with no sequence identity, at least 70% sequence identity and at least 80% sequence identity. Genes from the tetrapods exclusive set that had a significant match in zebrafish based on sequence similarity have been classified as non-syntenic paralogs or not. Declarations Not applicable Ethics approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials All processed data generated in this study have been submitted to Zenodo (https://zenodo.org/) and are available for download at the following DOI: 10.5281/zenodo.14628256. The code used in this study is available in the GitHub repository at https://github.com/fapatalano/TAD_evolutionary_analysis. Competing interests The authors declare that they have no competing interests Funding FP held a fellowship from the Faculty of Mathematics and Natural Sciences, University of Oslo. This work was supported by the Norwegian Research Council projects 324137 and 343102. Authors' contributions JP, RA and SRS conceived and designed the study; JP, RA and FP designed data analyses; FP analysed data. All authors read and approved the final manuscript. Acknowledgements We gratefully acknowledge Andrea Raffo for his invaluable contributions to the mathematical aspects of the implementation of the OU model, as well as professor Thomas Hansen for valuable discussions on the use of the OU model. We also extend our thanks to Roberto Rossini for the insightful discussions on the Hi-C data, which significantly enriched this work. References Rowley MJ, Corces VG. Organizational principles of 3D genome architecture. Nat Rev Genet. 2018 Dec;19(12):789–800. Dixon JR, Selvaraj S, Yue F, Kim A, Li Y, Shen Y, et al. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature. 2012 May;485(7398):376–80. Nora EP, Lajoie BR, Schulz EG, Giorgetti L, Okamoto I, Servant N, et al. Spatial partitioning of the regulatory landscape of the X-inactivation centre. Nature. 2012 May;485(7398):381–5. Sexton T, Yaffe E, Kenigsberg E, Bantignies F, Leblanc B, Hoichman M, et al. Three-Dimensional Folding and Functional Organization Principles of the Drosophila Genome. Cell. 2012 Feb 3;148(3):458–72. Hoencamp C, Dudchenko O, Elbatsh AMO, Brahmachari S, Raaijmakers JA, van Schaik T, et al. 3D genomics across the tree of life reveals condensin II as a determinant of architecture type. Science. 2021 May 28;372(6545):984–9. Dixon JR, Gorkin DU, Ren B. Chromatin Domains: The Unit of Chromosome Organization. Mol Cell. 2016 Jun 2;62(5):668–80. Fudenberg G, Imakaev M, Lu C, Goloborodko A, Abdennur N, Mirny LA. Formation of Chromosomal Domains by Loop Extrusion. Cell Rep. 2016 May 31;15(9):2038–49. Hildebrand EM, Dekker J. Mechanisms and Functions of Chromosome Compartmentalization. Trends Biochem Sci. 2020 May 1;45(5):385–96. Mirny LA, Imakaev M, Abdennur N. Two major mechanisms of chromosome organization. Curr Opin Cell Biol. 2019 Jun;58:142–52. Nuebler J, Fudenberg G, Imakaev M, Abdennur N, Mirny LA. Chromatin organization by an interplay of loop extrusion and compartmental segregation. Proc Natl Acad Sci. 2018 Jul 17;115(29):E6697–706. Dekker C, Haering CH, Peters JM, Rowland BD. How do molecular motors fold the genome? Science. 2023 Nov 10;382(6671):646–8. Dong P, Tu X, Chu PY, Lü P, Zhu N, Grierson D, et al. 3D Chromatin Architecture of Large Plant Genomes Determined by Local A/B Compartments. Mol Plant. 2017 Dec 4;10(12):1497–509. Lioy VS, Cournac A, Marbouty M, Duigou S, Mozziconacci J, Espéli O, et al. Multiscale Structuring of the E. coli Chromosome by Nucleoid-Associated and Condensin Proteins. Cell. 2018 Feb 8;172(4):771-783.e18. Mizuguchi T, Fudenberg G, Mehta S, Belton JM, Taneja N, Folco HD, et al. Cohesin-dependent globules and heterochromatin shape 3D genome architecture in S. pombe. Nature. 2014 Dec;516(7531):432–5. Acemel RD, Lupiáñez DG. Evolution of 3D chromatin organization at different scales. Curr Opin Genet Dev. 2023 Feb 1;78:102019. Vietri Rudan M, Barrington C, Henderson S, Ernst C, Odom DT, Tanay A, et al. Comparative Hi-C Reveals that CTCF Underlies Evolution of Chromosomal Domain Architecture. Cell Rep. 2015 Mar 3;10(8):1297–309. Krefting J, Andrade-Navarro MA, Ibn-Salem J. Evolutionary stability of topologically associating domains is associated with conserved gene regulation. BMC Biol. 2018 Aug 7;16(1):87. Lazar NH, Nevonen KA, O’Connell B, McCann C, O’Neill RJ, Green RE, et al. Epigenetic maintenance of topological domains in the highly rearranged gibbon genome. Genome Res [Internet]. 2018 Jun 18 [cited 2022 Aug 29]; Available from: https://genome.cshlp.org/content/early/2018/06/16/gr.233874.117 Eres IE, Gilad Y. A TAD Skeptic: Is 3D Genome Topology Conserved? Trends Genet TIG. 2021 Mar;37(3):216–23. Okhovat M, VanCampen J, Nevonen KA, Harshman L, Li W, Layman CE, et al. TAD evolutionary and functional characterization reveals diversity in mammalian TAD boundary properties and function. Nat Commun. 2023 Dec 7;14(1):8111. Long HS, Greenaway S, Powell G, Mallon AM, Lindgren CM, Simon MM. Making sense of the linear genome, gene function and TADs. Epigenetics Chromatin. 2022 Jan 29;15(1):4. Liao Y, Zhang X, Chakraborty M, Emerson JJ. Topologically associating domains and their role in the evolution of genome structure and function in Drosophila. Genome Res. 2021 Mar 1;31(3):397–410. Whalen S, Pollard KS. Most chromatin interactions are not in linkage disequilibrium. Genome Res. 2019 Mar 1;29(3):334–43. Eres IE, Luo K, Hsiao CJ, Blake LE, Gilad Y. Reorganization of 3D genome structure may contribute to gene regulatory evolution in primates. PLOS Genet. 2019 lug;15(7):e1008278. Renwick JH. The Mapping of Human Chromosomes. Annu Rev Genet. 1971;5(1):81–120. Bejerano G, Pheasant M, Makunin I, Stephen S, Kent WJ, Mattick JS, et al. Ultraconserved elements in the human genome. Science. 2004 May 28;304(5675):1321–5. Harmston N, Ing-Simmons E, Tan G, Perry M, Merkenschlager M, Lenhard B. Topologically associating domains are ancient features that coincide with Metazoan clusters of extreme noncoding conservation. Nat Commun. 2017 Sep 5;8(1):441. Kikuta H, Laplante M, Navratilova P, Komisarczuk AZ, Engström PG, Fredman D, et al. Genomic regulatory blocks encompass multiple neighboring genes and maintain conserved synteny in vertebrates. Genome Res. 2007 May 1;17(5):545–55. Bolt CC, Duboule D. The regulatory landscapes of developmental genes. Development. 2020 Feb 3;147(3):dev171736. Rödelsperger C, Dieterich C. CYNTENATOR: Progressive Gene Order Alignment of 17 Vertebrate Genomes. PLoS ONE. 2010 Jan 28;5(1):e8861. Li D, He M, Tang Q, Tian S, Zhang J, Li Y, et al. Comparative 3D genome architecture in vertebrates. BMC Biol. 2022 May 6;20(1):99. Tian X, Li R, Fu W, Li Y, Wang X, Li M, et al. Building a sequence map of the pig pan-genome from multiple de novo assemblies and Hi-C data. Sci China Life Sci. 2020 May;63(5):750–63. Wolff J, Rabbani L, Gilsbach R, Richard G, Manke T, Backofen R, et al. Galaxy HiCExplorer 3: a web server for reproducible Hi-C, capture Hi-C and single-cell Hi-C data analysis, quality control and visualization. Nucleic Acids Res. 2020 Jul 2;48(W1):W177–84. Hunt G. Measuring rates of phenotypic evolution and the inseparability of tempo and mode. Paleobiology. 2012 Jul;38(3):351–73. Rohlfs RV, Harrigan P, Nielsen R. Modeling Gene Expression Evolution with an Extended Ornstein–Uhlenbeck Process Accounting for Within-Species Variation. Mol Biol Evol. 2014 Jan 1;31(1):201–11. Hansen TF. Stabilizing Selection and the Comparative Analysis of Adaptation. Evolution. 1997;51(5):1341–51. Chen J, Swofford R, Johnson J, Cummings BB, Rogel N, Lindblad-Toh K, et al. A quantitative framework for characterizing the evolutionary history of mammalian gene expression. Genome Res. 2019 Jan;29(1):53–63. Sandelin A, Bailey P, Bruce S, Engström PG, Klos JM, Wasserman WW, et al. Arrays of ultraconserved non-coding regions span the loci of key developmental genes in vertebrate genomes. BMC Genomics. 2004 Dec 21;5:99. Hnisz D, Weintraub AS, Day DS, Valton AL, Bak RO, Li CH, et al. Activation of proto-oncogenes by disruption of chromosome neighborhoods. Science. 2016 Mar 25;351(6280):1454–8. Lupiáñez DG, Kraft K, Heinrich V, Krawitz P, Brancati F, Klopocki E, et al. Disruptions of topological chromatin domains cause pathogenic rewiring of gene-enhancer interactions. Cell. 2015 May 21;161(5):1012–25. Woltering JM, Noordermeer D, Leleu M, Duboule D. Conservation and Divergence of Regulatory Strategies at Hox Loci and the Origin of Tetrapod Digits. PLoS Biol. 2014 Jan 21;12(1):e1001773. Gómez-Marín C, Tena JJ, Acemel RD, López-Mayorga M, Naranjo S, de la Calle-Mustienes E, et al. Evolutionary comparison reveals that diverging CTCF sites are signatures of ancestral topological associating domains borders. Proc Natl Acad Sci U S A. 2015 Jun 16;112(24):7542–7. de Wit E. TADs as the Caller Calls Them. J Mol Biol. 2020 Feb 7;432(3):638–42. Rao SSP, Huntley MH, Durand NC, Stamenova EK, Bochkov ID, Robinson JT, et al. A 3D Map of the Human Genome at Kilobase Resolution Reveals Principles of Chromatin Looping. Cell. 2014 Dec 18;159(7):1665–80. Symmons O, Uslu VV, Tsujimura T, Ruf S, Nassari S, Schwarzer W, et al. Functional and topological characteristics of mammalian regulatory domains. Genome Res. 2014 Mar;24(3):390–400. Cooper N, Thomas GH, Venditti C, Meade A, Freckleton RP. A cautionary note on the use of Ornstein Uhlenbeck models in macroevolutionary studies. Biol J Linn Soc Linn Soc Lond. 2016 May;118(1):64–77. Steinegger M, Söding J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat Biotechnol. 2017 Nov;35(11):1026–8. Servant N, bot nf core, Ewels P, Peltzer A, Garcia MU, Miller E, et al. nf-core/hic: v2.0.0 - 2023-01-12 [Internet]. Zenodo; 2023 [cited 2024 Nov 18]. Available from: https://zenodo.org/records/7556794 Abdennur N, Mirny LA. Cooler: scalable storage for Hi-C data and other genomically labeled arrays. Bioinformatics. 2020 Jan 1;36(1):311–6. Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics. 2010 Mar 15;26(6):841–2. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation [Internet]. 2021 Aug 28 [cited 2024 Dec 17];2(3). Available from: https://www.cell.com/the-innovation/abstract/S2666-6758(21)00066-7 Hinrichs AS, Karolchik D, Baertsch R, Barber GP, Bejerano G, Clawson H, et al. The UCSC Genome Browser Database: update 2006. Nucleic Acids Res. 2006 Jan 1;34(Database issue):D590-598. Additional Declarations No competing interests reported. Supplementary Files supplementarytables.docx supplementaryfigures.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 16 May, 2025 Reviews received at journal 02 May, 2025 Reviews received at journal 17 Apr, 2025 Reviewers agreed at journal 05 Apr, 2025 Reviewers agreed at journal 03 Apr, 2025 Reviewers invited by journal 03 Apr, 2025 Submission checks completed at journal 03 Apr, 2025 First submitted to journal 02 Apr, 2025 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-5809117","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":437367616,"identity":"09b61601-04e3-4912-88ff-d4585bba765b","order_by":0,"name":"Fabiana Patalano","email":"","orcid":"","institution":"University of Oslo","correspondingAuthor":false,"prefix":"","firstName":"Fabiana","middleName":"","lastName":"Patalano","suffix":""},{"id":437367617,"identity":"3c7998dc-77e0-49e3-a8fc-3a1264b2af67","order_by":1,"name":"Simen Rød Sandve","email":"","orcid":"","institution":"Norwegian University of Life Sciences","correspondingAuthor":false,"prefix":"","firstName":"Simen","middleName":"Rød","lastName":"Sandve","suffix":""},{"id":437367618,"identity":"fffa2370-ecb1-4f8b-820e-4f2507cd75f0","order_by":2,"name":"Rein Aasland","email":"","orcid":"","institution":"University of Oslo","correspondingAuthor":false,"prefix":"","firstName":"Rein","middleName":"","lastName":"Aasland","suffix":""},{"id":437367619,"identity":"18331234-ebc6-4811-acb7-464274edeac0","order_by":3,"name":"Jonas Paulsen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/UlEQVRIiWNgGAWjYBACxgYYix1EVEgkgKgDxGlhBhFniNCCACAtjG0MCYQVtvce/PiD4Y68wWH2h48L51nkGRzgMTzwgeGOHE6H9ZxLluZheGa44TCPsfHMbRLFQC0GB2cwPDPGqWVGjoE0A8NhRqAWNmnebRKJMxt4DA7zMBxObMCtxfjnD4bD9hsOsz//zTuHOC1mEiAFGw4zmDHzNkgk9jMQ0tJzxsyax+BZ8kygX6R5jkkU8zOzFRycYXAYp18M23uMb/6ouGPbd7z94Weemro8NvbmzR8+VBzGGWKGYOsNDiAJgePUAJcGBgZ5CHUAt4pRMApGwSgYBQCZ8FU7V88zkwAAAABJRU5ErkJggg==","orcid":"","institution":"University of Oslo","correspondingAuthor":true,"prefix":"","firstName":"Jonas","middleName":"","lastName":"Paulsen","suffix":""}],"badges":[],"createdAt":"2025-01-11 11:23:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5809117/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5809117/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":79806476,"identity":"76153866-6b21-495a-8fed-7a3d11a268c7","added_by":"auto","created_at":"2025-04-03 05:28:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":47881,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePhylogenetic distance affects the overall synteny conservation pattern across vertebrates.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Overview of the 12 vertebrate species included in the study. The estimated divergence times were obtained from \u003ca href=\"http://timetree.org/\"\u003ehttp://timetree.org/\u003c/a\u003e. The figure was created using R, and the icons have been downloaded from \u003ca href=\"https://www.phylopic.org/\"\u003ehttps://www.phylopic.org/\u003c/a\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e Bar plot illustrating the number of conserved syntenic blocks identified for each alignment. The minimum number of blocks observed is primates, while the maximum in tetrapods.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5809117/v1/7fe5c151705a4eb41d06e6dc.png"},{"id":79807065,"identity":"f3a6b1da-072c-4fc0-ba6f-40d05bbf7d28","added_by":"auto","created_at":"2025-04-03 05:36:37","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":143688,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe TAD number is conserved, and this conservation correlates with evolutionary divergence.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Hi-C map illustrating TAD distributions within a syntenic block shared among mouse, rat, and chicken genomes (Rat: chr9 61755803 - 66120277; Mouse: chr1 74793498 - 74938002; Chicken: chr7 21976097 - 2202637). The heatmap shows identical TAD numbers within syntenic blocks of mouse and rat, while a lower number of TADs is observed in the chicken syntenic blocks.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e Scatterplot showing TAD number conservation is significantly higher than expected by chance (p.value \u0026lt;0.05). The observed conservation ranges from 0.22 to 0.69-fold lower than a random distribution.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5809117/v1/a8e86ce4c3db1f0d506b47b0.png"},{"id":79806480,"identity":"b97b5419-14b2-47d6-ba80-b2bbe468677e","added_by":"auto","created_at":"2025-04-03 05:28:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":78039,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTAD borders are evolutionary stable in syntenic blocks across the 12 species\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Diagram illustrating TAD borders in a syntenic block. The top panel maps borders in rhesus and human along real genomic coordinates (x-axis in bp). The bottom panel normalizes these borders to a 0–1 range while preserving relative distances. Dashed lines indicate where TAD borders align between species.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e Scatter plot showing all the TAD borders of syntenic blocks with the same number of TADs along with the calculated R² value of 0.91.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e Scatter plot showing the average R² scores for each clade, where higher R² values indicate greater similarity and conservation of TAD border locations, and lower R² values indicate greater divergence in border positions.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5809117/v1/9fd961b23aa1e17547b62e94.png"},{"id":79806483,"identity":"17cba908-c25f-4e36-84c6-00b0ebb342d8","added_by":"auto","created_at":"2025-04-03 05:28:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":90281,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSyntenic gene analysis indicates stable Gene-TAD conformation across vertebrate evolution.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Scatter plot showing the frequency of syntenic genes maintaining the same spatial localisation within TADs for blocks with matching TAD counts across different species pairs. The plot demonstrates that as evolutionary distance increases, the frequencies decrease, indicating greater conservation of gene spatial organisation in closely related species.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e Scatter plot showing the relationship between the Edit Distance (TAD border insertions or deletions) and the evolutionary distance between species. The plot demonstrates that as evolutionary distance increases, more edits are needed, indicating a decrease in TAD conservation over time.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-5809117/v1/9985395a0851b244f3ac0158.png"},{"id":79807071,"identity":"d592e337-a385-4ccf-97e2-cda1fc9748bd","added_by":"auto","created_at":"2025-04-03 05:36:37","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":424725,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOrnstein-Uhlenbeck Analysis Uncovers Stabilising Selection on TAD Numbers in Vertebrates\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Heatmap showing 4 species (dog, human, cow, and mouse) for both large and small 𝛼 values.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e Stacked histogram of 𝛼 values, where the blue section represents the fraction of syntenic blocks with at least one GRB and the red section represents blocks without any GRB. Most blocks show large 𝛼 values, suggesting rapid convergence to an optimal TAD number and strong stabilizing selection. A smaller subset of blocks with lower 𝛼 values indicates weaker selection, approaching the Brownian motion model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e \u0026nbsp;Boxplot displaying the distribution of R² values for blocks with low and high 𝛼 values, as well as for blocks containing at least one GRB, regardless of 𝛼 value.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-5809117/v1/5a46089d71d31af9cc407f41.png"},{"id":79807667,"identity":"9e0a30ca-59a6-42a0-8cbc-096963ce10e3","added_by":"auto","created_at":"2025-04-03 05:44:37","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1440238,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5809117/v1/90aeb38e-2395-4b97-96f4-a80037a85bad.pdf"},{"id":79807066,"identity":"20a17de0-2a38-4385-9b19-2a205c4cc1d9","added_by":"auto","created_at":"2025-04-03 05:36:37","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":21268,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarytables.docx","url":"https://assets-eu.researchsquare.com/files/rs-5809117/v1/aeaaee6c43cd03238923a993.docx"},{"id":79806490,"identity":"8a3cb3c7-efcb-42d4-9977-e67969f24ac7","added_by":"auto","created_at":"2025-04-03 05:28:37","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":6916043,"visible":true,"origin":"","legend":"","description":"","filename":"supplementaryfigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-5809117/v1/174087411a85f437f10b846f.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"TAD conservation in vertebrate genomes is driven by stabilising selection.","fulltext":[{"header":"Background","content":"\u003cp\u003eIn eukaryotes, genomic DNA is tightly packed with histones and non-histone proteins to form chromatin, which serves to protect and regulate genetic material while efficiently organising the long DNA molecules within the limited space of the nucleus. This organisation is hierarchical, with structural and regulatory processes working together to ensure genome function and stability (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA key feature of this dynamic organisation is the formation of Topologically Associating Domains (TADs), sub-megabase-scale chromatin segments characterised by high intra-domain contact frequencies (\u003cspan additionalcitationids=\"CR3 CR4\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). TADs typically contain genes interacting with regulatory elements within the same domain (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Interactions across TADs are restricted by border elements enriched with insulator proteins, such as CCCTC-binding factor (CTCF). The proposed mechanism for TAD formation involves loop extrusion, where cohesin interacts with CTCF, halting when cohesin encounters convergently orientated CTCF-bound sites (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan additionalcitationids=\"CR8 CR9 CR10\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWhile first discovered in mammals, TADs have since been observed in many species, from \u003cem\u003eDrosophila\u003c/em\u003e to prokaryotes and in plants (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Vertebrates rely on CTCF for TAD borders, while other species use alternative proteins, suggesting independent evolution of some TAD mechanisms (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAlthough significant progress has been made in understanding properties of TADs, the precise mechanisms underlying their evolutionary origins and conservation are not known. In particular, the degree of TAD evolutionary conservation is still debated. While some evidence supports a considerable TAD conservation (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e), recent direct comparisons of TAD borders have reported a low level of conservation, challenging this notion (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGrowing evidence, direct and indirect, suggests that TAD borders could play a key role as functional elements. Indirect evidence of TAD conservation comes from studies examining genomic synteny breakpoints and the occurrence of CTCF-binding sites across species. Recent studies have revealed that synteny breakpoints in various mammals (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e) and flies (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e) are concentrated at TAD borders and depleted within TAD bodies, suggesting TADs are conserved during evolution (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). The conservation of TADs has also been inferred through comparative analyses of CTCF-binding sites and insulator activity across species (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e), although this study does not report the number or proportion of conserved TADs observed (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Direct evidence of TAD conservation comes from several studies that compare TAD borders across different species. Initially, TADs were directly compared between human and mouse revealing that approximately 54% of human TAD borders are conserved in mouse (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Borders conservation was defined based on the presence or absence of borders across species. Using a similar approach with more stringent criteria, a comparison of human and chimpanzee TAD borders found that only 43% are shared between the two species (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). In a more recent study applying the same presence/absence criterion, it was estimated that only 14% of human and 15% of mouse TAD borders are conserved across 8 primate and rodent species (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e), challenging earlier reports of TAD border conservation. This approach relies on whole genome alignment and liftover processes, which becomes progressively more difficult as evolutionary distance increases.\u003c/p\u003e \u003cp\u003eHence, when comparing multiple distantly related species, approaches that do not rely on base-pair-level orthology is necessary. In this context synteny becomes a useful framework. Synteny (meaning \u0026ldquo;same ribbon\u0026rdquo; as introduced by Renwick (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e)) and refers to the preserved order of genes or genomic segments across species. As gene order is preserved across much larger evolutionary distances, syntenic blocks therefore represent a practical and evolutionary meaningful unit to explore patterns of TAD conservation.\u003c/p\u003e \u003cp\u003eBeyond gene order, some genomic regions exhibit strong structural and functional conservation, in particular some Genomic Regulatory Blocks (GRBs)\u0026mdash;large, evolutionarily constrained regions housing key developmental genes alongside highly conserved noncoding elements (CNEs) (\u003cspan additionalcitationids=\"CR27\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). The conservation of GRBs within syntenic blocks suggests that selection acts not only to maintain individual genes and their regulatory elements but also to preserve the broader genomic architecture that enables their coordinated function. The organization of GRBs closely aligns with TADs, which form chromatin loops through cohesin and CTCF interactions, insulating regulatory landscapes and maintaining proper gene expression (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). The overlap between GRBs, synteny, and TADs suggests that TADs may act as structural scaffolds, preserving cis-regulatory interactions and ensuring the proper regulation of developmental genes across evolutionary timescales.\u003c/p\u003e \u003cp\u003eIn this study we quantify the evolutionary conservation of TADs in a broader spectrum of vertebrate species using several TAD characteristics, such as TAD number, borders, and gene positioning within TADs. We show, using an Ornstein-Uhlenbeck model, that TAD-conservation is in many cases driven by stabilizing selection.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eIdentification of syntenic blocks across 12 species.\u003c/p\u003e\n\u003cp\u003eTo systematically explore TAD evolution, we compiled a dataset across the vertebrate lineage, including 12 species \u0026ndash; 10 mammalian species, chicken and zebrafish (Fig. 1A). All species feature high-quality genome assemblies and available Hi-C data obtained from fibroblast cell lines. While this dataset provides valuable insight into TAD evolution within vertebrates, it is important to note that our coverage is limited, largely due to the availability of Hi-C data. Similar studies have used syntenic-block-based strategies to enhance comparative TAD analyses (2,16,24). Following this approach, we analysed TADs within syntenic blocks to ensure meaningful cross-species comparisons while minimizing alignment artifacts that increase with evolutionary distance. This method also allows for a more comprehensive investigation of TAD configurations by going beyond isolated TADs, specifically examining how TADs are organized in relation to each other, including their spatial arrangement.\u003c/p\u003e\n\u003cp\u003eFor each ancestral node in the phylogenetic tree, we identified conserved syntenic blocks\u0026mdash;genomic segments with preserved order of genes across species evolving from that node\u0026mdash;using Cyntenator (30). A synteny block consists of at least two orthologous genes across genomes. The number of syntenic blocks varies, ranging from 100 between humans and rhesus monkeys to 651 when including all mammals and chicken (Fig. 1B). As expected, the evolutionary distance between species impacts the distribution of gene counts per block; for example, the rodent\u0026rsquo;s alignment (13 million years divergence) shows up to 1,767 genes per block (median \u0026plusmn; stdev: 31\u0026plusmn;369), while alignments across all species (429 million years divergence) reveal a maximum of only 16 genes (median \u0026plusmn; stdev: 3\u0026plusmn;2.27). Closely related species exhibit narrower gene count distributions, with an interquartile range of 144 for rodents compared to just 2 for the vertebrate\u0026rsquo;s alignment. Alignment coverage decreases with evolutionary distance, from over 90% to under 20% (additional details can be found in Table S1). The reduction of 4194 genes within syntenic blocks when going from tetrapods to vertebrates (Table S2) may result from the additional whole-genome duplication in teleost fish ~200 million years ago, followed by selective gene retention or loss (31). Analysis of these \u0026quot;missing\u0026quot; genes reveals that 34% (1425 genes) exhibit over 70% identity with non-syntenic paralogs, suggesting their relocation to non-syntenic regions rather than their complete absence from the zebrafish genome. Nevertheless, the percentage drops to 15% for genes sharing over 80% identity.\u003c/p\u003e\n\u003cp\u003eGene Ontology (GO) analysis of genes conserved across all 12 species showed significant (p.value \u0026lt;0.05) enrichment in processes crucial for development and cellular regulation. Key enriched terms include organ development and anterior-posterior pattern specification (Figure S1).\u003c/p\u003e\n\u003cp\u003eIdentification of TADs in the vertebrate fibroblast genomes.\u003c/p\u003e\n\u003cp\u003eWe then performed 3D genome analysis on publicly available Hi-C data (GEO: GSE167581, SRA IDs: PRJNA482496, (31,32)) of fibroblast cells. Our study included 12 Hi-C libraries, which we merged after iterative correction and KR normalisation.\u003c/p\u003e\n\u003cp\u003eThis process resulted in an average of 295 million valid reads per species, with the human dataset containing the highest number of valid reads (568 million), while the zebrafish dataset had the lowest (36.6 million) (Figure S2).\u003c/p\u003e\n\u003cp\u003eUsing HiCExplorer (33), we identified an average of 3,278 TADs across the 12 vertebrate genomes (mean \u0026plusmn; SD: 3,28 \u0026plusmn; 818). The identified TADs ranged from 1,207 in zebrafish to 4,037 in rhesus (Figure S3A). This variation in TAD numbers across species may be attributed to differences in Hi-C resolution. The presence of TADs within syntenic blocks differed greatly between species, as expected, since blocks cover less of the genome, with the highest inclusion rate of 99% observed in the common ancestor of humans and rhesus macaques and the lowest rate of 12% between humans and zebrafish (Table S3, Figure S3B). \u003c/p\u003e\n\n\u003cp\u003eTAD number is conserved across vertebrate evolution.\u003c/p\u003e\n\u003cp\u003eWe analysed the rate of evolutionary change in TAD features across species, comparing pairs of species without using human as a reference. This approach allowed us to assess TAD conservation independently across species with similar divergence times, such as mouse, cow, and dog, which diverged approximately 70\u0026ndash;100 million years ago.\u003c/p\u003e\n\u003cp\u003eAnalyses of evolutionary conservation of TAD numbers within syntenic blocks revealed an average change of 0.018 TADs per megabase pair (Mbp) per million years (Mbp/Myr) (Figure S4). Note that these change estimates assume a somewhat unrealistic linear relationship (34), but nevertheless provide an indication of the extent of TAD change over time. TAD number conservation decreases as evolutionary distance between species increases. This is exemplified also in Figure 2A, where species sharing a more recent common ancestor, such as the mouse and rat, show an identical count of TADs within syntenic blocks in contrast to the more distantly related chicken, which does not.\u003c/p\u003e\n\u003cp\u003eTo assess conservation and statistical significance across all alignments, we computed the Average Relative Difference (ARD), which measures the proportional mean difference in TAD numbers, with lower values indicating stronger conservation. This provides insights into the stability of TAD numbers and helps evaluate conservation across varying evolutionary distances. To evaluate statistical significance, we compared our scores to a random distribution generated by shuffling the TAD counts within the syntenic blocks 1000 times. This method provides a baseline to determine if the observed conservation is significantly greater than what would be expected by chance.\u003c/p\u003e\n\u003cp\u003eOur findings revealed that TAD number conservation is significantly higher than would be expected by chance (p.value \u0026lt;0.05, Figure S5). The conservation we observed ranged from 0.22 to 0.69-fold less than the random distribution (Figure 2B). Taking primates as an example, the TAD conservation was nearly0.2 times lower than the random. This pattern might suggest a strong phylogenetic signal, reflecting a non-random distribution of TAD number among these species. \u003c/p\u003e\n\u003cp\u003eDivergence time correlates with TAD border position conservation in vertebrates.\u003c/p\u003e\n\u003cp\u003eTo determine if TAD border positions are evolutionarily conserved within syntenic blocks, we selected blocks with an equal number of TADs, averaging 40 blocks per species pair comparison.\u003c/p\u003e\n\u003cp\u003eAvoiding lift-over, which loses reliability with increasing evolutionary distance, we normalized TAD border positions to a 0 to 1 scale within each syntenic block (Fig. 3A). We computed TAD border position conservation using the mean absolute differences in these relative border locations across all pairs of species. Dividing the species into three distinct clades (mammals, tetrapods and vertebrates) revealed an average difference in border position of ~9.2% within mammals, 10.7% in tetrapods, and ~14.2% in vertebrates (Figure S6). To exemplify this, we assessed mean absolute differences in the \u003cem\u003eSIX2/3\u003c/em\u003e cluster, a region known to be TAD-regulated in zebrafish (27). In this cluster all the border positions are conserved across different species pairs, and the absolute difference between pairs consistently stays under 0.2 (Figure S7), providing confidence to this metric.\u003c/p\u003e\n\u003cp\u003eTo further compare relative TAD border positions between species pairs, we calculated the R\u0026sup2; score, which measures the proportion of variance in the border positions of one species that can be explained by the positions in another species (Figure 3A and B). Applying this analysis to the entire dataset we find that relative border positions have a statistically significant correlation for all clades (Table S4), yet with a gradually decreasing R\u0026sup2; values over evolutionary time (Figure 3C). Additionally, we analysed the border outliers in primate alignments and found that, even though they are not statistically enriched in any GO term, these genes are involved in processes such as glucose response and immune response (Figure S8). \u003c/p\u003e\n\u003cp\u003eGene position is conserved within TADs.\u003c/p\u003e\n\u003cp\u003eBecause TADs serve to restrict gene regulatory interactions (6), we next investigated whether genes tend to stay within the same TADs over evolutionary time. To quantify this, we computed the average number of genes found within the same TAD across pairs of species within syntenic blocks. This revealed that mammalian synteny blocks have an average of 9.7 orthologous genes per Mbp/Myr. Given the average density of ~14 genes per Mbp in these blocks overall, this indicates a relatively high degree of gene content conservation within TADs. Similarly, for tetrapods, we observe a conservation of 9.3 genes per Mbp/Myr, yet for the whole vertebrate clade the conservation score drops to about 5 genes per Mbp per million years, reflecting a more substantial divergence in gene positional conservation as the evolutionary distance increases (Figure S9).\u003c/p\u003e\n\u003cp\u003eTo test if these levels of conservation were higher than expected by chance alone, we shuffled gene positions within each block 1000 times preserving the number of genes per block and the number of TADs, providing a baseline for comparison with our observed gene conservation scores. This revealed that genes generally maintained their spatial positioning across species, exceeding random expectation by factors ranging from 1.1 to 2.2 (Fig. 4A), with a p.value \u0026lt;0.05 (see Figure S10). For instance, in primates, gene positional conservation within TADs was observed to be 2 times higher than by chance (Fig. 4A), yet with a decreasing ratio over evolutionary time. \u003c/p\u003e\n\u003cp\u003eNext, we expanded these analyses to include syntenic blocks regardless of the number of TADs they contained. For this, we computed the theoretical number of TAD borders insertions or deletions needed to ensure that orthologous genes resided within the same TADs between a pair of species. \u003c/p\u003e\n\u003cp\u003eThe results show that the average number of required edits (i.e. insertions or deletions) is 0.8 borders/Mbp for mammals, increasing to 1.02 borders/Mbp in tetrapods and 1.08 borders/Mbp in vertebrates (Figure S11). The increasing number of required edits to convert one conformation into another correlates with the increasing evolutionary divergence between species. Computing the required border edits relative to shuffled genes (see above), revealed that all compared clades show a significantly lower number of border edits than expected by chance (Fig. 4B, Figure S12). \u003c/p\u003e\n\u003cp\u003eMost syntenic blocks show stabilizing selection while a few evolve under genetic drift.\u003c/p\u003e\n\u003cp\u003eAll analyses so far revealed a significant conservation of TAD features, yet with a strong phylogenetic signal. Hence, to assess whether TAD conservation is driven by stabilizing selection or evolving under genetic drift, requires us to apply comparative phylogenetic models which can distinguish between these evolutionary forces. Previous studies of conservation of gene expression (35) have made use of the Ornstein-Uhlenbeck (OU) process, which is a widely accepted model of continuous trait evolution (35\u0026ndash;37).\u003c/p\u003e\n\u003cp\u003eTo analyse the evolutionary dynamics of TAD numbers within syntenic blocks, we used the OU model, which quantifies the influence of both random genetic drift and stabilizing selection. It includes two parameters: alpha (𝛼), indicating the strength of selection pressure, and sigma (𝜎), representing the rate of genetic drift. A larger 𝛼 value suggests strong stabilizing selection that pulls a trait towards an optimal value, while a larger 𝜎 value points to greater variability due to random changes. By applying the OU model, we assessed how TAD number evolution is influenced by these factors in mammals. We focused on mammals in our analysis due to the availability of Hi-C data for these species, which provided a more robust dataset for our study. The representation of other vertebrate species in the dataset is limited, with only two species from outside the mammalian group. Including these outgroup species would have further reduced the number of genomic regions available for analysis, limiting the statistical power of our results.\u003c/p\u003e\n\u003cp\u003eWe performed a grid search across a range of 𝛼 and 𝜎 values, selecting the best-fitting parameters based on log-likelihood (see Methods). As 𝛼 approaches zero, the process mimics Brownian motion, pointing to genetic drift, while higher 𝛼 values suggest the presence of selective pressures. The cumulative log-likelihood results show a range of optimal 𝛼 and 𝜎 parameters, primarily concentrated in regions with large 𝛼 values (Figure S13), indicating significant selective pressure on many syntenic blocks. A smaller peak at low 𝛼 values indicates that some TADs may be evolving under near-neutral conditions. These patterns highlight the diverse evolutionary dynamics across different syntenic blocks (Figure 5A). Our analysis identified the most frequent parameter combination as 𝛼 = 5.2 and 𝜎 = 2.11 (Figure S14), suggesting strong selection and low variability due to genetic drift, suggesting that stabilizing selection plays an important role in conserving TAD numbers over evolutionary time. Interestingly, 4.12% syntenic blocks displayed 𝛼\u0026lt;0.01, indicating that their evolutionary trajectories are more affected by genetic drift. Performing GO-analysis on blocks with low 𝛼 (𝛼 = 0.01) reveals terms like collagen catabolic processes (adjusted P = 0.02; Figure S15A), extracellular matrix disassembly (adjusted P = 0.02) and regulation of neuroinflammatory response (adjusted P = 0.02).The majority of blocks (95.88% of blocks) are concentrated at higher 𝛼 values (𝛼\u0026gt;0.01), indicating strong selection on TAD numbers (Figure 5B). High-𝛼 blocks are associated with various processes like intermediate filament organization, organic acid transport and glycerophospholipid metabolic process (Figure S15B). Noteworthy terms are also those related to development, including pattern specification, limb development, and gliogenesis. Many syntenic blocks under stabilizing selection (high 𝛼 values, ~8000 genes) contain developmental genes, while fewer blocks evolving through genetic drift have diverse, non-developmental functions. \u003c/p\u003e\n\u003cp\u003eThe presence of developmental genes in our high-𝛼 blocks suggests that regulatory regions surrounding these genes could be relevant for the conservation of TAD structure. Highly conserved noncoding elements (CNEs) are known to surround developmental genes in vertebrates (38). Clusters of these CNEs, termed genomic regulatory blocks (GRBs) (28) have been shown to overlap with TAD structure, suggesting that TADs are needed to maintain the regulatory architecture in these regions (27). To investigate whether our synteny blocks associate with GRBs, we overlapped them with GRBs from reference (27). While GRB-containing syntenic blocks span a range of \u0026alpha; values, they do not strongly correlate with \u0026alpha; (Figure 5B). Our low-\u0026alpha; blocks generally show a less conserved TAD border positioning compared to high-\u0026alpha; blocks, confirming that values of \u0026alpha; reflect levels of TAD border position conservation (Figure 5C). Notably, GRB-containing blocks tend to have even more stable TAD border positions compared to high 𝛼 blocks in general (Figure 5C). GO term analysis reveals that GRB-containing low-\u0026alpha; blocks are not enriched for a particular process (Figure S15C; adjusted p.value\u0026gt;0.35), and high \u0026alpha; blocks are mainly enriched with various developmental processes (adjusted p.value\u0026lt;2.52e-04; Figure S15F). This confirms that the subset of GRBs with conserved TAD structures indeed harbours genes with developmental functions. Genes in non-GRB low \u0026alpha; blocks, on the other hand, are enriched in processes related to collagen metabolism (adjusted p.value\u0026lt;2.43e-05; Figure S15C), whereas non-GRB high \u0026alpha; blocks are not enriched for particular processes (Figure S15D). This shows that syntenic blocks with conserved TAD numbers (high \u0026alpha;) also harbour genes with non-developmental functions, especially if they do not harbour GRBs.\u003c/p\u003e\n\u003cp\u003eComparing values of 𝛼 with average number of genes and TADs within our blocks reveals that low 𝛼 blocks have a higher gene and TAD density than intermediate 𝛼 values (0.09-6.8). For blocks with 𝛼\u0026gt;8.4, gene and TAD density is the highest (Table S5). Consistent with prior studies (27), TADs overlapping GRBs have lower gene density (e.g., 12.60 genes/Mbp in humans vs. 20.82 genes/Mbp in non-GRB TADs, Table S6). However, protein sequence identity between species remains comparable across low- and high-\u0026alpha; blocks (Figure S16).\u003c/p\u003e\n\u003cp\u003eOverall, our analyses reveal that most syntenic blocks are governed by stabilizing selection of their TADs. A subset of these which harbour GRBs are associated with developmental genes, consistent with prior studies. However, we also identify a large proportion of non-GRB blocks with conserved TAD structure harbouring genes with non-developmental functions.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eTADs are essential functional units within the genome that play a significant role in regulating gene expression. Disruptions of TAD borders have been linked to abnormal gene interactions, gene mis-regulation, and various aberrant phenotypes, including cancer (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e). Additionally, such disruptions can lead to developmental disorders, such as human limb malformations (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOur study expands the current understanding of evolutionary conservation of TADs across a range of vertebrate species. We have extended the analysis of evolutionary conservation of TADs across 12 vertebrate species, using a comprehensive quantitative framework to assess TAD-feature conservation, including the use of phylogenetic models to detect signatures of selection on TADs. Our approach integrates both a general assessment of TAD conservation across species and a focused analysis of individual blocks using the OU-model. The initial broad analysis provides essential context for understanding overall conservation patterns, while the OU-model allows us to examine the evolutionary conservation of TAD numbers in greater detail. Early studies on TAD conservation primarily relied on indirect methods or limited locus analyses, such as the Hox (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e) and Six (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e) clusters. These investigations suggested TAD conservation based on patterns like the enrichment of rearrangement breakpoints at TAD borders and their depletion within TAD bodies across species (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePrevious studies in vertebrates have reported various levels of TAD-conservation, ranging from 54\u0026thinsp;\u0026minus;\u0026thinsp;14% across mammalian lineages of different evolutionary divergence levels (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Our findings indicate that conservation of TAD number, border positions, and gene positioning within TADs display a strong phylogenetic signal (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Comparing our findings directly with previous studies is challenging largely due to the lack of standardized methodologies for defining and computing TAD conservation. For instance, while some studies inferred conservation through CTCF-binding sites (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e), without a formal testing against a robust null model, others focused on broader comparisons not limited to syntenic regions. These methodological differences, combined with variability in data quality, and in TAD calling algorithms (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e), limit direct comparisons of conservation values between different studies. Nevertheless, our findings point towards conservation of TAD features during vertebrate evolution.\u003c/p\u003e \u003cp\u003eTADs play a crucial role in shaping the regulatory landscapes of developmental genes (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). Their stability, size variation, and conservation across metazoans suggest strong selective pressures, particularly in cases where TADs coincide with GRBs in both humans and Drosophila (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). The alignment of TADs with regulatory elements suggests that they function by restricting the reach of gene regulatory interactions (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite the evidence for evolutionary conservation, previous studies have not addressed the question whether this stability results from selection or genetic drift. In this study, we apply an OU-model with parameters that can be interpreted as the force of stabilizing selection (α) and genetic drift (σ), enabling this distinction (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). Our results revealed that a significant majority (95.88%) of syntenic blocks among mammals exhibit α\u0026thinsp;\u0026gt;\u0026thinsp;0.01 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB), indicating that they are under strong stabilizing selection. Genes in syntenic blocks with purifying selection pressure on TAD-structure were associated with developmental processes, including embryonic organ development and regionalization. This underscores the importance of conserving TAD numbers for maintaining the precise gene regulatory environment needed for establishing and maintaining organismal structure and function during vertebrate embryogenesis.\u003c/p\u003e \u003cp\u003eThe absence of enrichment for developmental genes in synteny blocks with low α parameter values reinforces the idea that TADs containing genes critical for development are subjected to stronger selective pressure. Conversely, TADs containing genes involved in more specialized or variable physiological processes may permit more flexibility in their number and arrangement, reflecting different evolutionary dynamics and greater genetic variability, despite the fact that their gene order is conserved.\u003c/p\u003e \u003cp\u003eAchieving accurate parameter estimates in OU models can be challenging, especially when tree size is small (\u0026lt;\u0026thinsp;100) (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). However, our study does not hinge on individual parameter estimates. Instead, we analyse the distribution of α values across hundreds of synteny blocks, revealing that most synteny blocks exhibit α values much greater than sigma. This strongly suggests the presence of stabilizing selection acting on the TAD number.\u003c/p\u003e \u003cp\u003eTADs associated with GRBs tend to exhibit lower gene density and higher border conservation, as developmental TADs are typically gene-poor but rich in intergenic regions that provide regulatory flexibility (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). In contrast, TADs located further from GRBs often have higher gene density and greater structural variability, likely supporting specialized, less conserved processes (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). However, GRBs are not the sole factor driving TAD conservation. Some GRB-free TADs remain highly conserved, displaying higher-than-average border similarity and strong stabilizing selection on their number. This suggests that while GRBs contribute to TAD conservation, especially for developmental genes, their presence is not always necessary for maintaining TAD structural integrity.\u003c/p\u003e \u003cp\u003eIt is crucial to acknowledge that gathering large-scale Hi-C data from the same tissue across multiple species is complex, adding another layer of difficulty to such studies. Our study using data from 12 vertebrate species reveals extensive evolutionary conservation of 3D genome organization across syntenic blocks, and highlights their likely role in gene regulation. Our results underscore the significance of using evolutionary models to study TAD conservation across species. Analysing more species will further enhance our understanding of shared patterns and lineage-specific differences in 3D genome conservation. Moreover, comprehensive mapping of conserved 3D genome structures can pinpoint regions and TADs vulnerable to disease upon disruption.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eOur study enhances the current understanding of TAD conservation across a range of vertebrate species by employing a nuanced approach that includes various TAD features. We found that TAD numbers, border positions, and gene locations exhibit substantial conservation. However, this trend decreases with increasing evolutionary divergence. TAD numbers change at a rate of 0.018 TADs Mbp/Myr, while TAD border positions show a decreasing correlation with evolutionary divergence, increasing from ~9.2% difference within mammals to ~14.2% difference across vertebrates. \u0026nbsp;Gene positions within TADs also show evidence of conservation, with synteny blocks containing an average of 9.7 orthologous genes per Mbp/Myr in mammals and 5 in vertebrates within the same TAD. Using the Ornstein-Uhlenbeck model, we identified that many syntenic blocks are subject to strong stabilizing selection. A subset of these blocks overlaps with GRBs and specifically harbour developmental genes, highlighting the importance of maintaining precise TAD configurations for crucial gene regulatory functions. Conversely, genes involved in specialized functions showed signs of TAD evolution primarily driven by genetic drift, suggesting TAD organization is less critical for these genes. Our findings underscore the evolutionary stability of TADs, while also illustrating the need\u0026nbsp;for standardized methodologies in assessing TAD conservation.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eData sets and synteny identification\u003c/p\u003e\n\u003cp\u003eIn this study, we analysed 12 different species: human (GRCh38.p14), rhesus macaque (Mmul_10), mouse (GRCm39), rat (mRatBN7.2), rabbit (UM_NZW_1.0), cow (ARS-UCD1.3), sheep (ARS-UI_Ramb_v2.0), pig (Sscrofa11.1), cat (F.catus_Fca126_mat1.0), dog (Dog10K_Boxer_Tasha), chicken (bGalGal1.mat.broiler.GRCg7b), and zebrafish (GRCz11). To investigate the relationship between synteny and 3D chromatin architecture, we first identified syntenic blocks using the Cyntenator tool. The three primary input files for this tool are:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cstrong\u003eGuide Tree:\u003c/strong\u003e Specifies the order in which the genomes should be aligned.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eGenome File:\u003c/strong\u003e Contains coordinates for protein-coding genes.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCorrespondence File\u003c/strong\u003e: Details orthologous relationships among genes across all input genomes.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe correspondence file was generated by performing an all-versus-all alignment of protein sequences from the twelve species using mmseq2 (47) with an e-value threshold of 10\u003csup\u003e-5\u003c/sup\u003e. Only the longest isoform from each protein sequence was kept. The phylogenetic tree from the TimeTree database served as the guide tree for progressive alignment. Cyntenator uses a stepwise alignment strategy, gradually adding species based on their evolutionary relationships. This method results in alignments that capture syntenic blocks shared among species within a specific node of the guide tree. Essentially, the syntenic blocks identified at each internal node represent genomic regions conserved among species branching from a common ancestor.\u003c/p\u003e\n\u003cp\u003eProcessing and Analysis of Hi-C Data.\u003c/p\u003e\n\u003cp\u003eRaw Hi-C sequencing data was obtained from the Gene Expression Omnibus (GEO: GSE167581,SRA IDs: PRJNA482496, (31,32)) and processed using the nf-core/hic (v2.0.0) pipeline (48) to ensure quality control, sequence mapping, and filtering. The reference genomes utilized were the latest versions available as of 2023.\u003c/p\u003e\n\u003cp\u003eThe pipeline was executed with HindIII as the restriction enzyme and was configured with the following optional flags: --skip_maps, --skip_dist_decay, --skip_tads, --skip_compartments, --skip_balancing, --skip_mcool, and --split_fastq=false.\u003c/p\u003e\n\u003cp\u003eFor creating Hi-C contact matrices, we employed the Cooler software (49) (v0.8.11) to produce matrices in the cooler format. These matrices were subsequently converted to the. mcool file format using the cooler \u003cem\u003ezoomify\u003c/em\u003e tool. Matrix balancing was then performed using Iterative Correction (ICE). Data from biological replicates were merged for comprehensive analysis. To identify TAD borders at a 32 kb resolution, we used the hicFindTADs command from HiCExplorer (33) (v3.7.2).\u003c/p\u003e\n\u003cp\u003eTo identify TADs located within each syntenic block, we used BEDTools (v2.30.0) (50)\u003cem\u003e\u0026nbsp;intersect\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003eMeasuring TAD number conservation across vertebrates.\u003c/p\u003e\n\u003cp\u003eTo explore the relationship between synteny and the conservation of TADs throughout vertebrate evolution, we analysed various TAD characteristics across 12 different species. Our initial goal was to assess the conservation of TAD numbers within synteny blocks. First, we identified synteny blocks for each species and counted the TADs within them, normalizing the counts by synteny block lengths in Mbp. We then calculated the absolute differences in these normalized TAD counts between species pairs, adjusting for evolutionary divergence times. By using linear regression to fit these differences against divergence times, we determined the slope, representing the average rate of TAD number change per Mbp per million years.\u003c/p\u003e\n\u003cp\u003eTo further quantify the conservation of TAD numbers, we also used a scoring metrics defined:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cstrong\u003eAverage Relative Difference (ARD):\u003c/strong\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eFor each pair of species \u003cem\u003ei\u003c/em\u003e and\u003cem\u003e\u0026nbsp;j\u003c/em\u003e within a given syntenic block \u003cem\u003ek\u003c/em\u003e, the relative difference in TAD counts between species \u003cem\u003ei\u003c/em\u003e and\u003cem\u003e\u0026nbsp;j\u003c/em\u003e within the block is:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003eThe ARD score is the average of these relative differences across all syntenic blocks. \u0026nbsp;This score quantifies the average relative difference in TAD counts between species \u003cem\u003ei\u003c/em\u003e and\u003cem\u003e\u0026nbsp;j\u0026nbsp;\u003c/em\u003ein the syntenic blocks. We created these randomized distributions by shuffling the order of syntenic blocks 1000 times. For each iteration, we calculated an ARD score, and used the resulting distribution to evaluate the statistical significance of our observed ARD values.\u003c/p\u003e\n\u003cp\u003eQuantifying the evolutionary conservation of TAD borders\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe next focused on the positioning of TAD borders within synteny blocks to quantify their conservation throughout vertebrate evolution. We normalized the positions of TAD borders within synteny blocks on a scale from 0 to 1. This normalization enabled us to compare TAD positions between different species using a consistent scale, regardless of their specific genomic coordinates. We assessed the changes in TAD borders positions over evolutionary time by calculating the absolute differences between corresponding borders within each synteny block, expressed as changes per million years. The species were grouped into three distinct evolutionary clades: mammals only, mammals and chicken, and a broader group including mammals, chicken, and zebrafish. For each group, we computed the average difference in TAD borders positions and analysed these differences in relation to species divergence. To further evaluate the conservation of TAD borders, we used the R\u0026sup2; score. This statistical measure quantifies the extent to which TAD borders in one species can predict corresponding borders in another species.\u003c/p\u003e\n\u003cp\u003eGene-TAD conformation stability evaluation\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo evaluate the maintenance of syntenic genes within the same TAD across evolution, we focused on examining the positioning of orthologous syntenic genes within TADs to determine if they exhibit a tendency to reside within the same TAD. We computed the percentage of genes within the same TAD as the observed data, followed by a random permutation of TAD borders to establish a comparative baseline. This score was evaluated only for syntenic blocks with the same number of TADs. \u0026nbsp;To further analyse the differences in TAD configurations between species, we introduced another score-defined Edit Distance. This score evaluates the number of edits\u0026mdash;specifically, TAD borders insertions or deletions\u0026mdash;needed to transition from one TAD conformation to another.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOrnstein-Uhlenbeck (OU) Model\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn evolutionary biology, modelling how traits evolve within populations over time is essential for understanding adaptation and divergence mechanisms. We employed the OU process to model the evolution of continuous traits along a phylogeny. Specifically, we analyzed the number of TADs in each syntenic block across a mammalian alignment. Each syntenic block was analysed individually to assess how the number of TADs evolves over time in relation to the phylogeny.\u003c/p\u003e\n\u003cp\u003eThe OU model is governed by the following stochastic differential equation (SDE):\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003eGene ontology enrichment analysis\u003c/p\u003e\n\u003cp\u003eWe performed Gene Ontology (GO) biological process enrichment analysis using the \u003cem\u003eclusterProfiler\u003c/em\u003e R package with the \u003cem\u003eenrichGO\u003c/em\u003e function (51). Terms with an adjusted p.value\u0026thinsp;\u0026lt;\u0026thinsp;0.05, calculated using the Benjamini-Hochberg (BH) method, were considered significantly enriched. The top 10 enrichment terms, ranked by their Gene Ratio in descending order, were selected for visualization.\u003c/p\u003e\n\u003cp\u003eGRB-including syntenic blocks identification.\u003c/p\u003e\n\u003cp\u003eWe obtained the list of GRBs from (27), which have identified using CNEs showing 70% identity over 50\u0026thinsp;bp between human and chicken (hg19-galGal4). Their genomic coordinates to the hg38 assembly using the UCSC LiftOver tool (52) with default parameters. To classify syntenic blocks as either GRB-including or non-GRB-including, we assessed their overlap with GRBs using BEDTools (v2.30.0) (50) \u003cem\u003eintersect\u003c/em\u003e with no filtering based on the extent of overlap.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIdentification of non-syntenic paralogy\u003c/p\u003e\n\u003cp\u003eTo classify genes into possible non-syntenic paralogs and not, we analysed alignment blocks in the synteny dataset. For each block, we identified human genes absent in vertebrate alignment but present in tetrapods. These genes were checked against sequence similarity data to determine if alternative orthologs existed in zebrafish with no sequence identity, at least 70% sequence identity and at least 80% sequence identity. Genes from the tetrapods exclusive set that had a significant match in zebrafish based on sequence similarity have been classified as non-syntenic paralogs or not.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003eEthics approval and consent to participate\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003eConsent for publication\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003eAvailability of data and materials\u003c/p\u003e\n\u003cp\u003eAll processed data generated in this study have been submitted to Zenodo (https://zenodo.org/) and are available for download at the following DOI: 10.5281/zenodo.14628256. The code used in this study is available in the GitHub repository at https://github.com/fapatalano/TAD_evolutionary_analysis.\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eFP held a fellowship from the Faculty of Mathematics and Natural Sciences, University of Oslo. This work was supported by the Norwegian Research Council projects 324137 and 343102.\u003c/p\u003e\n\u003cp\u003eAuthors' contributions\u003c/p\u003e\n\u003cp\u003eJP, RA and SRS conceived and designed the study; JP, RA and FP designed data analyses; FP analysed data. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003cbr\u003e\u0026nbsp;We gratefully acknowledge Andrea Raffo for his invaluable contributions to the mathematical aspects of the implementation of the OU model, as well as professor Thomas Hansen for valuable discussions on the use of the OU model. We also extend our thanks to Roberto Rossini for the insightful discussions on the Hi-C data, which significantly enriched this work.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eRowley MJ, Corces VG. Organizational principles of 3D genome architecture. Nat Rev Genet. 2018 Dec;19(12):789\u0026ndash;800.\u003c/li\u003e\n\u003cli\u003eDixon JR, Selvaraj S, Yue F, Kim A, Li Y, Shen Y, et al. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature. 2012 May;485(7398):376\u0026ndash;80.\u003c/li\u003e\n\u003cli\u003eNora EP, Lajoie BR, Schulz EG, Giorgetti L, Okamoto I, Servant N, et al. Spatial partitioning of the regulatory landscape of the X-inactivation centre. Nature. 2012 May;485(7398):381\u0026ndash;5.\u003c/li\u003e\n\u003cli\u003eSexton T, Yaffe E, Kenigsberg E, Bantignies F, Leblanc B, Hoichman M, et al. Three-Dimensional Folding and Functional Organization Principles of the \u003cem\u003eDrosophila\u003c/em\u003e Genome. Cell. 2012 Feb 3;148(3):458\u0026ndash;72.\u003c/li\u003e\n\u003cli\u003eHoencamp C, Dudchenko O, Elbatsh AMO, Brahmachari S, Raaijmakers JA, van Schaik T, et al. 3D genomics across the tree of life reveals condensin II as a determinant of architecture type. Science. 2021 May 28;372(6545):984\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eDixon JR, Gorkin DU, Ren B. Chromatin Domains: The Unit of Chromosome Organization. Mol Cell. 2016 Jun 2;62(5):668\u0026ndash;80.\u003c/li\u003e\n\u003cli\u003eFudenberg G, Imakaev M, Lu C, Goloborodko A, Abdennur N, Mirny LA. Formation of Chromosomal Domains by Loop Extrusion. Cell Rep. 2016 May 31;15(9):2038\u0026ndash;49.\u003c/li\u003e\n\u003cli\u003eHildebrand EM, Dekker J. Mechanisms and Functions of Chromosome Compartmentalization. Trends Biochem Sci. 2020 May 1;45(5):385\u0026ndash;96.\u003c/li\u003e\n\u003cli\u003eMirny LA, Imakaev M, Abdennur N. Two major mechanisms of chromosome organization. Curr Opin Cell Biol. 2019 Jun;58:142\u0026ndash;52.\u003c/li\u003e\n\u003cli\u003eNuebler J, Fudenberg G, Imakaev M, Abdennur N, Mirny LA. Chromatin organization by an interplay of loop extrusion and compartmental segregation. Proc Natl Acad Sci. 2018 Jul 17;115(29):E6697\u0026ndash;706.\u003c/li\u003e\n\u003cli\u003eDekker C, Haering CH, Peters JM, Rowland BD. How do molecular motors fold the genome? Science. 2023 Nov 10;382(6671):646\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eDong P, Tu X, Chu PY, L\u0026uuml; P, Zhu N, Grierson D, et al. 3D Chromatin Architecture of Large Plant Genomes Determined by Local A/B Compartments. Mol Plant. 2017 Dec 4;10(12):1497\u0026ndash;509.\u003c/li\u003e\n\u003cli\u003eLioy VS, Cournac A, Marbouty M, Duigou S, Mozziconacci J, Esp\u0026eacute;li O, et al. Multiscale Structuring of the E. coli Chromosome by Nucleoid-Associated and Condensin Proteins. Cell. 2018 Feb 8;172(4):771-783.e18.\u003c/li\u003e\n\u003cli\u003eMizuguchi T, Fudenberg G, Mehta S, Belton JM, Taneja N, Folco HD, et al. Cohesin-dependent globules and heterochromatin shape 3D genome architecture in S. pombe. Nature. 2014 Dec;516(7531):432\u0026ndash;5.\u003c/li\u003e\n\u003cli\u003eAcemel RD, Lupi\u0026aacute;\u0026ntilde;ez DG. Evolution of 3D chromatin organization at different scales. Curr Opin Genet Dev. 2023 Feb 1;78:102019.\u003c/li\u003e\n\u003cli\u003eVietri Rudan M, Barrington C, Henderson S, Ernst C, Odom DT, Tanay A, et al. Comparative Hi-C Reveals that CTCF Underlies Evolution of Chromosomal Domain Architecture. Cell Rep. 2015 Mar 3;10(8):1297\u0026ndash;309.\u003c/li\u003e\n\u003cli\u003eKrefting J, Andrade-Navarro MA, Ibn-Salem J. Evolutionary stability of topologically associating domains is associated with conserved gene regulation. BMC Biol. 2018 Aug 7;16(1):87.\u003c/li\u003e\n\u003cli\u003eLazar NH, Nevonen KA, O\u0026rsquo;Connell B, McCann C, O\u0026rsquo;Neill RJ, Green RE, et al. Epigenetic maintenance of topological domains in the highly rearranged gibbon genome. Genome Res [Internet]. 2018 Jun 18 [cited 2022 Aug 29]; Available from: https://genome.cshlp.org/content/early/2018/06/16/gr.233874.117\u003c/li\u003e\n\u003cli\u003eEres IE, Gilad Y. A TAD Skeptic: Is 3D Genome Topology Conserved? Trends Genet TIG. 2021 Mar;37(3):216\u0026ndash;23.\u003c/li\u003e\n\u003cli\u003eOkhovat M, VanCampen J, Nevonen KA, Harshman L, Li W, Layman CE, et al. TAD evolutionary and functional characterization reveals diversity in mammalian TAD boundary properties and function. Nat Commun. 2023 Dec 7;14(1):8111.\u003c/li\u003e\n\u003cli\u003eLong HS, Greenaway S, Powell G, Mallon AM, Lindgren CM, Simon MM. Making sense of the linear genome, gene function and TADs. Epigenetics Chromatin. 2022 Jan 29;15(1):4.\u003c/li\u003e\n\u003cli\u003eLiao Y, Zhang X, Chakraborty M, Emerson JJ. Topologically associating domains and their role in the evolution of genome structure and function in Drosophila. Genome Res. 2021 Mar 1;31(3):397\u0026ndash;410.\u003c/li\u003e\n\u003cli\u003eWhalen S, Pollard KS. Most chromatin interactions are not in linkage disequilibrium. Genome Res. 2019 Mar 1;29(3):334\u0026ndash;43.\u003c/li\u003e\n\u003cli\u003eEres IE, Luo K, Hsiao CJ, Blake LE, Gilad Y. Reorganization of 3D genome structure may contribute to gene regulatory evolution in primates. PLOS Genet. 2019 lug;15(7):e1008278.\u003c/li\u003e\n\u003cli\u003eRenwick JH. The Mapping of Human Chromosomes. Annu Rev Genet. 1971;5(1):81\u0026ndash;120.\u003c/li\u003e\n\u003cli\u003eBejerano G, Pheasant M, Makunin I, Stephen S, Kent WJ, Mattick JS, et al. Ultraconserved elements in the human genome. Science. 2004 May 28;304(5675):1321\u0026ndash;5.\u003c/li\u003e\n\u003cli\u003eHarmston N, Ing-Simmons E, Tan G, Perry M, Merkenschlager M, Lenhard B. Topologically associating domains are ancient features that coincide with Metazoan clusters of extreme noncoding conservation. Nat Commun. 2017 Sep 5;8(1):441.\u003c/li\u003e\n\u003cli\u003eKikuta H, Laplante M, Navratilova P, Komisarczuk AZ, Engstr\u0026ouml;m PG, Fredman D, et al. Genomic regulatory blocks encompass multiple neighboring genes and maintain conserved synteny in vertebrates. Genome Res. 2007 May 1;17(5):545\u0026ndash;55.\u003c/li\u003e\n\u003cli\u003eBolt CC, Duboule D. The regulatory landscapes of developmental genes. Development. 2020 Feb 3;147(3):dev171736.\u003c/li\u003e\n\u003cli\u003eR\u0026ouml;delsperger C, Dieterich C. CYNTENATOR: Progressive Gene Order Alignment of 17 Vertebrate Genomes. PLoS ONE. 2010 Jan 28;5(1):e8861.\u003c/li\u003e\n\u003cli\u003eLi D, He M, Tang Q, Tian S, Zhang J, Li Y, et al. Comparative 3D genome architecture in vertebrates. BMC Biol. 2022 May 6;20(1):99.\u003c/li\u003e\n\u003cli\u003eTian X, Li R, Fu W, Li Y, Wang X, Li M, et al. Building a sequence map of the pig pan-genome from multiple de novo assemblies and Hi-C data. Sci China Life Sci. 2020 May;63(5):750\u0026ndash;63.\u003c/li\u003e\n\u003cli\u003eWolff J, Rabbani L, Gilsbach R, Richard G, Manke T, Backofen R, et al. Galaxy HiCExplorer 3: a web server for reproducible Hi-C, capture Hi-C and single-cell Hi-C data analysis, quality control and visualization. Nucleic Acids Res. 2020 Jul 2;48(W1):W177\u0026ndash;84.\u003c/li\u003e\n\u003cli\u003eHunt G. Measuring rates of phenotypic evolution and the inseparability of tempo and mode. Paleobiology. 2012 Jul;38(3):351\u0026ndash;73.\u003c/li\u003e\n\u003cli\u003eRohlfs RV, Harrigan P, Nielsen R. Modeling Gene Expression Evolution with an Extended Ornstein\u0026ndash;Uhlenbeck Process Accounting for Within-Species Variation. Mol Biol Evol. 2014 Jan 1;31(1):201\u0026ndash;11.\u003c/li\u003e\n\u003cli\u003eHansen TF. Stabilizing Selection and the Comparative Analysis of Adaptation. Evolution. 1997;51(5):1341\u0026ndash;51.\u003c/li\u003e\n\u003cli\u003eChen J, Swofford R, Johnson J, Cummings BB, Rogel N, Lindblad-Toh K, et al. A quantitative framework for characterizing the evolutionary history of mammalian gene expression. Genome Res. 2019 Jan;29(1):53\u0026ndash;63.\u003c/li\u003e\n\u003cli\u003eSandelin A, Bailey P, Bruce S, Engstr\u0026ouml;m PG, Klos JM, Wasserman WW, et al. Arrays of ultraconserved non-coding regions span the loci of key developmental genes in vertebrate genomes. BMC Genomics. 2004 Dec 21;5:99.\u003c/li\u003e\n\u003cli\u003eHnisz D, Weintraub AS, Day DS, Valton AL, Bak RO, Li CH, et al. Activation of proto-oncogenes by disruption of chromosome neighborhoods. Science. 2016 Mar 25;351(6280):1454\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eLupi\u0026aacute;\u0026ntilde;ez DG, Kraft K, Heinrich V, Krawitz P, Brancati F, Klopocki E, et al. Disruptions of topological chromatin domains cause pathogenic rewiring of gene-enhancer interactions. Cell. 2015 May 21;161(5):1012\u0026ndash;25.\u003c/li\u003e\n\u003cli\u003eWoltering JM, Noordermeer D, Leleu M, Duboule D. Conservation and Divergence of Regulatory Strategies at Hox Loci and the Origin of Tetrapod Digits. PLoS Biol. 2014 Jan 21;12(1):e1001773.\u003c/li\u003e\n\u003cli\u003eG\u0026oacute;mez-Mar\u0026iacute;n C, Tena JJ, Acemel RD, L\u0026oacute;pez-Mayorga M, Naranjo S, de la Calle-Mustienes E, et al. Evolutionary comparison reveals that diverging CTCF sites are signatures of ancestral topological associating domains borders. Proc Natl Acad Sci U S A. 2015 Jun 16;112(24):7542\u0026ndash;7.\u003c/li\u003e\n\u003cli\u003ede Wit E. TADs as the Caller Calls Them. J Mol Biol. 2020 Feb 7;432(3):638\u0026ndash;42.\u003c/li\u003e\n\u003cli\u003eRao SSP, Huntley MH, Durand NC, Stamenova EK, Bochkov ID, Robinson JT, et al. A 3D Map of the Human Genome at Kilobase Resolution Reveals Principles of Chromatin Looping. Cell. 2014 Dec 18;159(7):1665\u0026ndash;80.\u003c/li\u003e\n\u003cli\u003eSymmons O, Uslu VV, Tsujimura T, Ruf S, Nassari S, Schwarzer W, et al. Functional and topological characteristics of mammalian regulatory domains. Genome Res. 2014 Mar;24(3):390\u0026ndash;400.\u003c/li\u003e\n\u003cli\u003eCooper N, Thomas GH, Venditti C, Meade A, Freckleton RP. A cautionary note on the use of Ornstein Uhlenbeck models in macroevolutionary studies. Biol J Linn Soc Linn Soc Lond. 2016 May;118(1):64\u0026ndash;77.\u003c/li\u003e\n\u003cli\u003eSteinegger M, S\u0026ouml;ding J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat Biotechnol. 2017 Nov;35(11):1026\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eServant N, bot nf core, Ewels P, Peltzer A, Garcia MU, Miller E, et al. nf-core/hic: v2.0.0 - 2023-01-12 [Internet]. Zenodo; 2023 [cited 2024 Nov 18]. Available from: https://zenodo.org/records/7556794\u003c/li\u003e\n\u003cli\u003eAbdennur N, Mirny LA. Cooler: scalable storage for Hi-C data and other genomically labeled arrays. Bioinformatics. 2020 Jan 1;36(1):311\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eQuinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics. 2010 Mar 15;26(6):841\u0026ndash;2.\u003c/li\u003e\n\u003cli\u003eWu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation [Internet]. 2021 Aug 28 [cited 2024 Dec 17];2(3). Available from: https://www.cell.com/the-innovation/abstract/S2666-6758(21)00066-7\u003c/li\u003e\n\u003cli\u003eHinrichs AS, Karolchik D, Baertsch R, Barber GP, Bejerano G, Clawson H, et al. The UCSC Genome Browser Database: update 2006. Nucleic Acids Res. 2006 Jan 1;34(Database issue):D590-598.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [BMC Biology](https://bmcbiol.biomedcentral.com/)","snPcode":"12915","submissionUrl":"https://submission.springernature.com/new-submission/12915/3","title":"BMC Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5809117/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5809117/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eTopologically Associating Domains (TADs) are fundamental structural and gene regulatory components of chromatin defined by regions of high intra-domain contact frequency. Though TADs are found across diverse metazoans, the extent of their evolutionary conservation is still debated. Some studies indicate significant conservation among closely related species, while others suggest considerable variability, raising questions about the evolutionary forces that preserve TAD organization.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eHere, we investigated the evolutionary conservation of TADs by analysing Hi-C data from 12 vertebrate species. We examined TAD numbers, borders, and gene positioning within TADs. We found that TAD\u0026rsquo;s features are all highly conserved across species, but this conservation tends to decrease with evolutionary distance. Closely related species show greater TAD conservation compared to more distantly related species. Yet, modelling the divergence in TAD conservation using Ornstein-Uhlenbeck (OU) process revealed considerable selective pressures on TAD number within syntenic blocks, suggesting that TAD features in these genomic regions are under stabilising selection. However, we also identified a small subset of blocks where TAD numbers evolve under genetic drift, highlighting the existence of distinct groups of blocks subject to different evolutionary dynamics.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThese findings improve our understanding of TAD conservation and evolution, revealing significant conservation of TAD features, especially among closely related species. We discovered that TADs evolve both under stabilizing selection and genetic drift, highlighting the complex evolutionary dynamics of TAD evolution.\u003c/p\u003e","manuscriptTitle":"TAD conservation in vertebrate genomes is driven by stabilising selection.","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-03 05:28:32","doi":"10.21203/rs.3.rs-5809117/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-05-16T12:04:54+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-05-02T13:04:51+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-17T12:25:38+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"269566703058999008442981135280259274224","date":"2025-04-05T14:44:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"20950822484871555822218355217606978312","date":"2025-04-03T14:38:05+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-03T14:19:21+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-03T07:57:30+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Biology","date":"2025-04-02T10:25:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [BMC Biology](https://bmcbiol.biomedcentral.com/)","snPcode":"12915","submissionUrl":"https://submission.springernature.com/new-submission/12915/3","title":"BMC Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b0f8a8c8-6de3-4e14-8c86-25edf28975cf","owner":[],"postedDate":"April 3rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-07-25T17:23:13+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-03 05:28:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5809117","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5809117","identity":"rs-5809117","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}