Negative selection on baboon admixture is strongest on chromosome X

Negative selection on baboon admixture is strongest on chromosome X | 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 Negative selection on baboon admixture is strongest on chromosome X Erik Fogh Sørensen, Garrett Hellenthal, Kasper Munch This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7456420/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 10 You are reading this latest preprint version Abstract Background The six admixing baboon species offer a natural experiment to study negative selection on admixture and the nature of genetic incompatibility. In Tanzania, a secondary contact between olive and yellow baboons allows admixture despite 1.3 million years of divergence. An independent secondary contact occurred in Ethiopia, upon which olive baboons invaded and displaced an ancient Hamadryas-like population separated by 0.6 million years, mirroring the displacement of Neanderthals by modern humans. Results We analyze 156 high-coverage genomes sampled from seven olive and four yellow baboon populations in East Africa. Analyzing local ancestry across the whole genome, we find evidence of negative selection on minor parent ancestry in both Tanzanian yellow and olive baboon populations that reaches far beyond the hybrid zone. Across populations, we find that selection on minor parent ancestry is stronger on the X chromosome than on autosomes, most extremely in one yellow baboon population, which shows a seven-fold difference. The proportion of minor parent ancestry (MPA) is substantially higher on the X chromosome in Ethiopian olive and yellow baboon populations, which both displaced the populations now representing their minor parent ancestry, owing mainly to a few genomic regions with MPA at very high frequencies. We hypothesize that strong negative selection on MPA allowed these X chromosome regions to retain the original ancestry, as this was slowly displaced across the remaining genome. Conclusions Our findings provide deeper insights into admixture dynamics in primates, highlighting the persistence of selection against admixture across various levels of admixture, and underscoring the need to include chromosome X in admixture analyses. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Background Speciation is a gradual process in which populations diverge because of geographic, behavioural or ecological isolation. This isolation fosters differentiation as populations accumulate unique mutations and experience distinct selective pressures and genetic drift[1–4]. In cases of prolonged isolation, genetic incompatibilities can emerge[5–7], eventually leading to complete reproductive isolation, preventing further admixture[8–10]. Variants experience selection if they are less fit. This can occur if they arise in an inbred population[11], are maladapted to the local environment[2,12], or cause sexual incompatibilities, reducing the reproductive success of hybrids[4,13,14]. A more complex source of selection is genetic incompatibility in the form of Bateson–Dobzhansky–Muller Incompatibility (BDMI), which may arise from negative epistatic interactions when genes of different ancestry are mixed in hybrids. In that case, negative selection is expected to gradually remove the minor parent ancestry (MPA). Selection against admixture from the rarer ancestry across swordfish populations has been reported as evidence of BDMI[3] by detecting negative selection on admixture[6,15–17]. In baboons (genus Papio), hybridization between distinct populations is commonly observed. Still, our previous report[18] showed that genetic variants from foreign ancestry have to travel through the hybrid zone before reaching populations further from the hybrid zone[18–20], even in the face of extensive ongoing admixture. With its multiple species, the baboon species complex thus presents an ideal system for investigating negative selection mechanisms on admixture. Baboons have six distinct species, with the deepest split dated at least 1.3 million years ago between southern baboons (yellow, kinda, and chacma baboons; P. cynocephalus, P. kindae, and P. ursinus) and northern baboons (olive, hamadryas, and guinea baboons; P. anubis, P. hamadryas, and P. papio)[18,21]. Baboon species exhibit a striking discordance between autosomal and mitochondrial phylogenies[22], with a mitochondrial clade extending across olive, yellow, and hamadryas baboon ranges in East Africa (dotted lines in Fig. 1 ), even though olive and yellow baboons also each carry species-specific, deeply diverged mitochondria in other parts of their species range. Our recent analysis[18] supported that this discordance arose when an ancient widespread hamadryas-like baboon was subsumed in male-driven range expansions of yellow and olive baboons, displacing the nuclear genome of the resident population but not its mitochondria. In this process, often called nuclear swamping[23,24], continuous displacement of original ancestry will result in the invading species going from minor to major ancestry. Any hybrid incompatibility would thus initially manifest as negative selection on the invading ancestry. As the invading ancestry reaches the majority, selection would change to favour the invading ancestry. The relative strengths of migration and negative selection on the minor parent ancestry thus determine which ancestry ends up fixed and could result in swamped populations retaining high-frequency original ancestry at incompatibility loci. Haldane’s rule that genetic incompatibility first emerges in the hemizygous sex is supported by a century of observations. It suggests an outsize role of the mammalian X chromosomes in hybrid incompatibility[10,25–28]. The smaller effective population size (Ne) of the X chromosome results in faster divergence by genetic drift, a larger sensitivity to recent changes in population size[29], and sensitivity to sex-biased dispersion and reproductive variance[27,30]. Dynamic co-amplification of multicopy ampliconic genes in Mus musculus and Bos taurus suggests that mammalian X and Y chromosomes participate in an arms race to control the X/Y ratio of fertile sperm[31–33]. This hypothesis predicts an accelerated evolution supported by multiple reports of positive selection on primate X chromosomes[34–37]. If admixture is neutral, it will be distributed randomly across the genome. On the other hand, if linked selection purges specific admixture tracts, there will be an association between proxies for linked selection efficacy and admixture levels[3,20]. This correlation can be interpreted as a sign of selection against admixture if regions with high potential linked selection carry less admixture than neutral regions. The rate of admixture removal will be highest in the first generations, where large numbers of negatively selected variants remain linked. However, the correlation with recombination rate and MPA will not be strong for very recent admixture, where introgressed segment sizes remain orders of magnitude larger than that of fine-scale recombination rate and the 100kb scale at which we quantify MPA[38]. In these initial generations, removal will instead be much more indiscriminate due to the high linkage across the chromosomes[11,39]. We expect the strongest association between recombination rate and MPA as admixture disperses behind the hybrid zone and recombination reduces segment lengths to sizes relevant to medium-scale recombination rates and selection coefficients. However, the correlation will gradually weaken based on time since admixture, as the deleterious minor parent ancestry variants are removed and the MPA proportion is reduced. A previous study by Vilgalys et al. on yellow and olive baboons found evidence of selection against admixture in a baboon hybrid zone in the Amboseli basin[20]. They found that gene-dense and low recombination regions had significantly less admixture than expected. If the introgressed sequence carries variants under negative selection, we expect that more admixture is retained in high-recombination regions where neutral introgressed sequence segments are more readily unlinked from the deleterious variants that are eventually purged from the population[3,20]. In this hybrid zone, baboons with older admixture account for most of this correlation. In contrast, recent hybrids did not show this correlation pattern, indicating that it takes at least 4–5 generations for introgressing chromosomes to break into segments of a size similar to the scale of recombination rate and gene density variation[38]. This study aims to elucidate the dynamics of selection against admixture in the baboon species complex, focusing on identifying genome-wide patterns and the role of the X chromosome. Chromosome X is known to have more admixture than the autosomes for both olive and yellow baboons[18], even though many other admixture cases exhibit decreased admixture on chromosome X[6,25,40]. We investigate the relationship between admixture and linked selection across eleven sampling locations with varying distances from hybrid zones. The current species boundary represents a secondary contact between olive and yellow baboons[22–24,41]. Globetrotter was used to date the most recent admixture events to between 10 and 100 generations for the various populations[18,42,43] and found extensive migration between the olive baboon populations in Tanzania. The Amboseli population is similar to the Arusha and Tarangire populations, which all contain large amounts of minor parent ancestry. All three national parks are located near the boundary between olive and yellow baboons, with high variance in admixture proportions in sampled individuals. Improved reference population size allows for better admixture inference, as parentage of rare variants, frequency of common variants, and species-specific haplotype combinations are all better resolved with more samples. All individuals are 30X sequenced, allowing for higher accuracy and resolution methods, and are sampled across 19 localities. Sampling of a gradient of olive baboon populations shows how admixture patterns change as they penetrate deeper into the relatively pure olive baboon ranges. Unlike the previous studies[3,20], the X chromosome is also investigated, showing up to seven times stronger removal of minor parent ancestry in low recombination and low diversity regions. Results Minor Parent Ancestry declines with distance from the hybrid zone We infer local ancestry along each high-coverage P. anubis (olive) and P.cynocephalus (yellow) baboon genome using RFMix[44]. The four other baboon species were used as reference panels. We assign olive baboon ancestry ("northern ancestry") to genomic segments closer to P. hamadryas and P. papio (hamadryas and Guinea baboons) and yellow baboon ancestry ("southern ancestry") to segments closer to P. kindae and P. ursinus (Kinda and chacma baboons). We consider minor parent ancestry (MPA) the rarer ancestry in each population and represent the MPA proportion along the genome as its mean in nonoverlapping, contiguous 100-kilobase windows. To assess the accuracy of RFMix, we used Haptools[45] to simulate admixture from the current standing variation of two of the sampled populations. The Pearson correlation between the simulated and inferred admixture by RFMix is 98.9%, with no bias towards major parent ancestry and high diversity regions. See Supplementary Simulation of admixture for details. The autosomal MPA proportion of sampled populations differs widely (Fig. 1 , Supplementary Table 1). In yellow baboons, it ranges from 8.0% in Ruaha, closest to the hybrid zone, to 0.36% in Selous, furthest away from the hybrid zone. In olive baboons, the MPA proportion drops from 20.9% in the Tarangire population, which borders the yellow range, to 3.4% in the Serengeti population, located 250 km into the olive range. This relationship with distance from the hybridization zone suggests that the introgressed sequence is removed by selection as it disperses by animal movement between populations close to the species contact zone[22–24,41]. Southern ancestry is virtually absent in the Ethiopian Gog olive baboon population (autosomal MPA proportion 0.065%), which is located in Ethiopia, 1100 km from the olive-yellow hybrid zone. This proportion is consistent with no introgression from southern baboons into Gog olive baboons, as this small signal can be caused by incomplete lineage sorting and/or noise. Correlation with recombination rate reveals negative selection on MPA far behind hybrid zones To obtain fine-scale recombination rates, we created a genetic map with Pyrho [46,47] and SMC++ [48,49] using the 38 individuals from Mikumi. See Methods and Materials. The distribution of recombination in 100 kb windows has a rightward skew (see Supplementary Distribution and regressions with outliers), and 0.5% of 100kb windows have an estimated recombination rate above 0.408cM. When included, these outliers will have an outsized impact on the regressions. To restrict the analysis to a representative range of recombination rates, we removed the upper and lower 0.5 percentiles of recombination rate. We use variance-weighted linear regressions to account for heteroscedasticity (Supplementary Table 2 and Supplementary Fig. 1) and find a significant association between the population mean MPA proportion in 100kb windows and the 100kb window recombination rate. This is true for all populations (p-values < 2e-5) except Tarangire, Arusha, and Lake Manyara (See Fig. 2 and Supplementary Table 3), consistent with negative selection on MPA in both olive and yellow baboon populations far from the hybrid zone. However, in line with expectations, the correlation is absent in Tarangire and Arusha, where admixture was most recently introduced, which is evident as longer MPA segments in the local ancestry paintings shown in Fig. 1 . Lake Manyara baboons show no significant correlation despite a MPA proportion smaller than the baboons in Ngorongoro and Ruaha. Ruaha, the most admixed yellow population, also has the steepest regression slope (0.0414), significantly larger than the slope for the Selous, Mikumi, and Udzungwa populations (p-value 0.00067), which range from 0.0288 to 0.0325 (See Fig. 2 A, Supplementary Table 3). For an impression of the effect sizes, we ranked autosomal 100kb windows by their recombination rate and computed the mean MPA for windows in five equally sized bins ordered based on recombination rate (see Supplementary Fig. 3 and Supplementary Table 4). Comparing mean MPA proportions for Mikumi between the 20% windows with the lowest recombination rates to the 20% with the highest rates shows a difference of 0.25% (0.896–1.15%), corresponding to a relative difference of 28%. Ruaha, the most admixed population, shows a similar absolute difference of 0.23 percentage points but a relative difference of only 3%. The degree of admixture also varies among Tanzanian olive baboon populations (Fig. 2 B and Supplementary Table 1). The Gombe population has the steepest regression slope of 0.0633 (p-value 5.6e-19) for recombination rate and MPA proportion, and the Serengeti and Ngorongoro populations also have significant regression slopes (p-values 4.07e-07 and 0.00177). We find no significant correlation between recombination rate and MPA proportion in Lake Manyara after Bonferroni correction (p-value 0.0306). The populations in Tarangire and Arusha at the active contact zone do not show a significant relationship between recombination rate and MPA. Arusha has the largest absolute and relative difference between low and high-recombination windows (6.82% in the lowest 20–7.22% in the highest 20%, a 6% increase). Ngorongoro has 4.92% MPA in the lowest 20–5.16% MPA in the highest 20%, a 5% increase. See Supplementary Table 4 for all quintile differences. Correlation with diversity offers a better proxy for negative selection on MPA As a proxy for linked selection, the recombination rate does not account for variation in the density of sites under negative or positive selection. A more direct quantification of linked selection is relative genetic diversity, measured as π, the mean pairwise differences among haplotypes in the population. Although recent bottlenecks may exacerbate variation in diversity along the chromosomes[50], the relative degree of diversity is primarily shaped by linked selection[39,51–54] and variation in mutation rate across the genome[55–57]. However, as admixture itself contributes to 100kb mean diversity, we must obtain independent estimates of π from “background” populations, which do not share historical admixture events with the studied population. In the absence of shared admixture in the background populations, this estimate of π quantifies linked selection as a proxy for the functional importance of each 100kb window. If independent admixture contributes to background population diversity, its distribution across 100kb windows will additionally reflect the strength of negative selection on MPA; i.e., diversity measured in admixed background populations is expected to correlate more strongly with MPA proportion in the study population if they also experience similar selective events[18]. We calculated the average diversity across Guinea, Hamadryas, Kinda, and Chacma baboon reference populations in 100kb windows and will refer to this as “background diversity” below. (See Supplementary Fig. 4 for distribution). Background diversity strongly, but not perfectly, correlates with the recombination rate (Pearson correlation: 0.602, p-value: 0.0, Supplementary Fig. 5). Among yellow baboons, the Ruaha population again shows the steepest regression slope between background diversity and MPA (9.25, p-value 1.91e-19), significantly higher than other yellow populations (p-value: 1.01e-10), while the Mikumi (slope 2.19, p-value 1.21e-16) and Udzungwa populations also have significant regression slopes and slope 3.79, p-value 9.51e-18) (Supplementary Table 5). Further from the hybrid zone, the Selous population shows no significant correlation after Bonferroni correction to account for testing multiple populations (see Fig. 2 C). In all populations, the difference in MPA proportions between the bottom and top 20% windows ranked by background diversity is larger than when ranking by recombination rate. In Ruaha (see Supplementary Fig. 6 and Supplementary Table 6), the absolute difference is 16% (7.68–8.92%), and Udzungwa shows the largest relative difference, 44%, between the bottom and top 20% windows (1.22–1.76%). Among the olive baboon populations, the correlation between MPA proportion and background diversity is significantly stronger in populations away from the hybrid zone. Gombe (p-value 4.35e-34) and Ngorongoro (p-value 4.03e-13) show the strongest correlation between background diversity and MPA (see Fig. 2 D). The Gombe population has the largest absolute and relative difference in mean MPA proportions for windows in the bottom and top diversity quintiles (MPA 4.46% in the lowest quintile and 5.74% in the highest, a 29% increase). The Ngorongoro population has the second largest absolute difference (5.14% in the lowest 20–5.65% in the highest 20%, a 10% increase). The Tarangire and Arusha populations closest to the contact zone show no significant correlation between background diversity and MPA. To formally compare the predictive strengths of background diversity and recombination, we z-normalized them and used a generalized linear model to explain MPA proportion as a function of both recombination rate and background diversity (see Supplementary Table 7 for all coefficients and p-values). The Ruaha, Gombe, Lake Manyara, and Serengeti baboon populations all had a significant negative association with recombination for MPA prediction (p-values 0.000113, 2.43e-07, 2.78e-07, and 6.86e-07) while having a larger positive slope based on background diversity. As an example of the different regression slopes, the slope of the Ruaha population, when using only recombination, was 0.0414. When using only background diversity, it was 9.25, whilst the regression using both had slopes of -0.0559 and 10.5, respectively. That is, while the recombination rate is positively correlated with the MPA proportion, its correlation becomes negative when modelled together with background diversity. This may in part reflect that genomic regions with very low recombination rates have more admixture than expected when adjusting for background diversity, indicating that admixture might persist more easily in low-recombination, high-diversity regions than in high-recombination, high-diversity regions. The correlation between mutation rate and recombination rate allows diversity to incorporate most of the recombination rate’s predictive power. Chromosome X diversity is strongly reduced In all studied populations, the diversity on chromosome X relative to autosomes is significantly lower than the ¾ expected from hemizygosity[58]. This is partly explained by recent bottlenecks experienced by many populations (See Supplementary Fig. 7), most severely in northern baboons, which more strongly affect chromosome X diversity. However, even after adjusting for the effect of historical population sizes and the lower mutation rate on chromosome X[55], the X/autosome ratio remains significantly below ¾ (See Supplementary Table 8) (Mann-Whitney U test, p-values from 1.78e-26 to 2.17e-272). Adjusted X/autosome ratios range from 0.346 to 0.521, corresponding to ratios between 73.4% and 88.2% of the expected values. All baboon species exhibit higher reproductive variance in males[23,27,59–63], which should increase rather than reduce the X/autosome ratio, leaving only stronger linked selection on the X chromosome and sex-biased demographics[30] to explain the lower ratios. See Supplementary Fig. 8 for the distribution of background diversity on chromosome X. Linked selection on MPA is up to seven times stronger on the X chromosome On the X chromosome, all olive baboon populations except Tarangire and Arusha show significant correlations of MPA proportion with both background diversity (Fig. 3 B and Supplementary Table 9) and recombination rate (Supplementary Table 10). The Ngorongoro population has an association between background diversity and MPA proportion that is an order of magnitude larger than that for the autosomes (slope 81.1, p-value 8.19e-09). Here, the mean MPA proportion goes from 0.47–7.27% between the bottom and top 20% 100kb windows ranked by background diversity, a 14.4-fold difference. The similarly strong association in the Serengeti olive baboons (slope 68, p-value 2.47e-08) shows a span of 0.70% in the lowest quintile to 6.37% in the highest quintile in MPA proportion, an 8.04-fold difference. Among the yellow baboon populations (Fig. 3 A, Table 3), only the Ruaha population shows a significant association between background diversity and MPA proportion (slope 156, p-val 1.79e-16). Here, the 100kb windows with a background diversity in the top and bottom quintiles are 7.94% and 20.1%, a 2.5-fold difference (See Supplementary Fig. 10 and Supplementary Table 11). The overall admixture in low-diversity regions is similar among all four yellow baboon populations. Instead, the Ruaha population has much more admixture in high-diversity regions than the other yellow baboons. (See Supplementary Tables 9 and 10 for results on all populations). To formally test whether the correlation between background diversity and MPA proportion is stronger on the X chromosome than on autosomes, we Z-score normalized background diversity and performed a weighted linear regression with MPA proportion against normalized diversity, chromosome type, and their interaction, keeping the variance weights the same as in the previous models. A linear model with an interaction term can determine whether there is a significant difference between the regression slopes for the X chromosome and the autosomes. Among olive baboons, the regression slopes for the Ngorongoro, Lake Manyara, Serengeti, and Gombe populations are significantly higher on the X chromosome (p-values 6.29e-06, 2.53e-06, 1.42e-09, 0.00235). In Ngorongoro and Serengeti, the slopes are 5.53 and 7.13 times steeper on the X chromosome. As expected, from their location near the hybrid zone, the Tarangire and Arusha populations (Fig. 4 and Supplementary Table 12) do not show significant differences between regression slopes for chromosome X and the autosomes. The Ruaha yellow population (Fig. 4 ) also shows a significantly larger slope on chromosome X, 7.4 times steeper than for autosomes (p-val 1.42e-14). (See Supplementary Table 12 for all slopes and p-values). Many populations show higher MPA proportions on the X chromosome In all yellow baboon populations, the MPA proportion is significantly higher on chromosome X than on the autosomes ( Fig. 5 , Mann-Whitney U-test p-values below 8.06e-05, see Supplementary Table 13), whereas the olive baboon populations show no significant difference (See Fig. 5 ). MPA proportions for the X chromosome vary between 25.0% in Tarangire and 4.2% in Serengeti (see Supplementary Table 1), and in the yellow baboon populations vary between 12.9% in Ruaha and 4.5% in Selous. See Supplementary Fig. 11 for the distribution of MPA proportions across individual chromosome haplotypes, stratified by chromosome type. The higher MPA proportions on the X chromosome of many populations seem to contradict our observation of stronger negative linked selection on admixture on the X chromosome. On the X chromosome, the distribution of MPA proportions is skewed toward higher frequencies (See Supplementary Fig. 12), particularly in the yellow baboons, where this is evident as vertical patterns in the local ancestry painting (Fig. 6 ). To investigate if high-frequency MPA is responsible for the stronger correlations on chromosome X, we repeated our analyses after masking 100kb windows with MPA proportions above 25% (See Supplementary Fig. 13). Individual X and autosome regressions remain significant, but the association on X is strengthened in yellow baboons and weakened in olive baboons (Supplementary Table 14): In Udzungwa and Mikumi yellow baboons, the regression slopes for chromosome X are now significantly higher than for autosomes (p-values 7.26e-15 and 2.35e-05), whereas Lake Manyara and Serengeti olive baboons no longer show higher slopes for chromosome X than for autosomes. This observation suggests that any negative selection on high-frequency MPA in yellow baboons is not represented by background diversity levels. To investigate if high-frequency MPA could be explained as recent adaptive introgressions, we used Relate to scan for recent selective sweeps. The sample sizes of the individual populations are not sufficient for this analysis. However, an analysis of the pooled Tanzanian olive baboon populations yielded significant evidence of positive selection across the autosome but none on the X chromosome. (See Supplementary Fig. 14, Supplementary Section Positive Selection Scan). These sweeps on the autosomes were significantly more common in areas with low MPA, and we therefore find no evidence that adaptive introgressions cause high MPA areas. High-frequency MPA is most common on X chromosomes in the yellow populations that represent nuclear swamping of a resident northern lineage, and could represent haplotypes from the original displaced olive population. If these haplotypes were incompatible with gradually introgressing yellow haplotypes, strong negative selection would maintain these as the major parent ancestry. Hamadryas MPA in Ethiopian olive baboons reveals strong selection on hybrid ancestry To further explore this incompatibility scenario, we also analyzed the separate hamadryas MPA component of the Gog olive baboon population in Ethiopia. Olive and hamadryas baboons belong to the northern clade of baboon species and are thus more similar than olive and yellow baboons. Their divergence is comparable to that of anatomically modern humans and Neanderthals[21,64]. The mitochondria of the Gog olive baboon population are more similar to those of hamadryas baboons than to the mitochondria of other sampled olive populations, including the Tanzanian olive baboons, which also carry hamadryas-like mitochondria[22,24]. This discordance has been explained by nuclear swamping, where an invasion of primarily olive males displaced the autosomal but not mitochondrial ancestry of a resident hamadryas-like population[18]. We ran RFMix to infer hamadryas-like ancestry along chromosomes of the Gog individuals. We used Tanzanian olive and Hamadryas baboons as the two reference panels. Minor Parent Ancestry (MPA) is calculated as the proportion of Hamadryas ancestry in 100-kilobase windows after filtering areas with less than 75% callable bases (See Materials and Methods). On the autosomes, Gog olive baboons have the steepest regression slopes of all investigated baboon populations against both recombination (Fig. 7 A, slope 0.251, p-value 1.03e-76) and background diversity (Fig. 7 B, slope 38.8, p-value 1.58e-132) (Supplementary Table 15), as well as the largest absolute difference between the lowest 20% background diversity quintile and the highest (7.05% in the lowest 20–12.6% in the highest 20%, a 78% increase). With a regression using both recombination rate and background diversity, it is only background diversity that has a significant association (p-values 0.393 and 1.41e-65). Wide genomic regions retaining pure Hamadryas ancestry suggest large-scale hybrid incompatibility. Despite a large MPA proportion (Fig. 7 C and D), the X chromosome of Gog olive baboons also shows the steepest slope in the background diversity and MPA proportion regression of all investigated populations, with a slope of 214, and a p-value of 7.18e-10, significantly higher than that for autosomes (p-value 8.88e-06). After normalization, the X chromosome regression is 54% steeper than the autosomal one. See Supplementary Table 15 for all regressions. Gog olive baboons also show the largest absolute difference between the 20% low diversity quantile chromosome X (14.5–42.1%, a 190% increase). Despite showing the strongest evidence of negative selection on MPA, the Gog X chromosome also shows the largest proportion of MPA in our study: 33.1% compared to only 8.8% on autosomes. (Fig. 7 D and Fig. 8 ). These contrasting observations are qualitatively similar to those for yellow and olive admixture, yet greatly amplified. The evidence of negative selection is strongest here despite very long admixture tracts, suggesting a selection regime different from autosomes. These observations are all consistent with nuclear swamping and point to strong selection retaining original ancestry on the X chromosome. Discussion Our comprehensive analysis of 156 high-coverage baboon genomes reveals that negative selection on admixture is pervasive across both olive and yellow baboon populations, extending far beyond hybrid zones and persisting over evolutionary timescales. The striking finding that linked selection against minor parent ancestry (MPA) is up to seven times stronger on the X chromosome than autosomes provides compelling evidence for the disproportionate role of sex chromosomes in maintaining species boundaries, consistent with Haldane's rule and theoretical predictions about hybrid incompatibility. Perhaps the most intriguing finding is the apparent paradox that many populations show significantly higher MPA proportions on the X chromosome despite experiencing stronger negative selection on MPA. This pattern is particularly pronounced in yellow baboon populations and Ethiopian olive baboons, where X chromosome MPA can exceed 30% while showing the steepest correlations between MPA and proxies for linked selection. The absence of evidence for adaptive introgression on the X chromosome, despite high-frequency MPA in some regions, argues against positive selection driving these patterns. Instead, the retention of original ancestry at high frequency likely reflects strong negative selection against hybrid genotypes at these loci. During nuclear swamping events[23,65], the invading ancestry transitions from minor to major status as it spreads through the population. Our observations suggest that strong selection on the invading minor parent ancestry blocks the turnover of ancestry in genomic regions spanning several megabases. This would explain why Ethiopian olive baboons show 33.1% hamadryas ancestry on the X chromosome compared to only 8.8% on autosomes, despite showing the strongest evidence for negative selection against MPA in our entire dataset. We propose that the wide genomic regions retaining pure hamadryas ancestry on the X chromosome represent incompatibility loci where hybrid genotypes face severe fitness costs. At incompatibility loci where such negative selection was insufficient to prevent the invading ancestry from reaching the majority, selection would instead purge the resident ancestry, which is now the minority, possibly explaining the pure olive ancestry extending across several megabases. The gradient of MPA proportions from hybrid zones into the species' core ranges provides a natural experiment for understanding how selection acts on admixture over time. Populations closest to contact zones (Tarangire, Arusha) show no correlation between MPA and recombination rate or background diversity, consistent with recent admixture where introgressed segments remain too large for fine-scale recombination to effectively unlink neutral from deleterious variants. This observation aligns with theoretical predictions and previous empirical work[38,54,66], suggesting that 4–5 generations are required before recombination can effectively purge deleterious variants while preserving neutral admixed segments. The progressive strengthening of the correlation between MPA and linked selection proxies with distance from hybrid zones demonstrates that selection continues to shape admixture patterns long after initial hybridization events. The Gombe olive baboon population, located furthest from active admixture zones, shows the steepest regression slopes, indicating ongoing purging of deleterious admixed variants even in populations with relatively ancient admixture histories (estimated at 10–100 generations by Globetrotter analysis). Our finding that background diversity consistently outperforms recombination rate as a predictor of MPA proportions has important implications for understanding the mechanisms of selection against admixture. While recombination rate has been a commonly used proxy for linked selection efficiency, it fails to account for variation in the density of functional sites under selection. Background diversity, measured in non-admixed reference populations, integrates the effects of both recombination rate and functional constraint, providing a more comprehensive measure of linked selection intensity. The negative association between recombination rate and MPA when controlling for background diversity suggests that low-recombination, high-diversity regions may harbour incompatibility loci where admixture persists at intermediate frequencies. This could occur if these regions contain balanced polymorphisms or if the fitness costs of admixture are frequency-dependent. Alternatively, structural variants such as inversions, which suppress recombination while maintaining diversity, might play a role in maintaining admixture in these regions. The ubiquity of negative selection against admixture across all studied populations, including those hundreds of kilometres from hybrid zones, challenges the notion that gene flow readily homogenizes genomes between hybridizing species. Even in the face of ongoing gene flow, selection maintains species boundaries by continuously purging incompatible genetic combinations. The much stronger linked selection on the X chromosome MPA in some populations underscores the critical role of sex chromosomes in speciation, supporting theoretical predictions about the evolution of reproductive isolation. The reduced X/autosome diversity ratios (34.6–52.1% of expected values) across all populations suggest that the X chromosome experiences stronger linked selection even within species. This could result from the exposure of recessive deleterious mutations in hemizygous males, the smaller effective population size of the X chromosome, the accumulation of sexually antagonistic alleles, or meiotic drive. The combination of stronger within-species selection and stronger negative selection on MPA makes the X chromosome a particularly effective barrier to gene flow between species. Materials and Methods A mask was created based on the depth of coverage per individual per site. For each individual, the most common read depth (mode) was calculated using bcftools (see Supplementary Fig. 15 for distribution, Supplementary Section Commands). Sites were filtered using bcftools, removing sites that were uncalled, had a heterozygous call and alternate allele depth less than 3, coverage less than a third of the mean coverage or more than twice, or a GQ score less than or equal to 30. A species-wide mask was created by calling a base as passing if 95% or more of the samples had a passing state with bedtools multiIntersectBed. For chromosome X, the mask was created using only the 98 females. Diversity was calculated using scikit-allels' windowed_diversity function with a window size of 100kb. All windows in the analysis are 100kb, starting from base 1 of the reference genome unless stated otherwise. Windows were removed from the analysis if less than 75% of the window was set as passing in the filter step, resulting in 25397 100kb windows for the autosome and 1063 100kb windows for chromosome X. Analysis was performed on a per-population basis as identified in the metadata. See Supplementary Table 16 for a count of males and females. The historical effective population size for every population was inferred with SMC++. The autosome mutation rate was set at 0.57e-08 per base. With 4.5 times more mutations in males[72], the chromosome X was set at 0.45e-08. This mutation rate was calculated based on the higher mutation rate in males corresponding to (4.5/3 + 2/3)/(4.5/2 + 1/2), the mutation rate in chromosome X divided by the autosome mutation rate. Generation time was set at 11 years[55]. The Autosomal and chromosome X SMC + + run was run with a piecewise spline and ten expectation maximization iterations. Chromosome X Ne was inferred only using females. A fine-scale recombination map was inferred using Pyrho and the Mikumi Yellow population, as it had the largest sample size. The large number of samples from this population allows for a more accurate inference of the fine-scale recombination rate. However, the modest admixture from olives will slightly bias our map toward lower rates in regions with higher MPA proportions, as admixture increases LD in the affected region[67,68]. The recombination rate in 100kb windows has a mean and median of 0.0834 and 0.0668 centiMorgan (cM), scaled to a size equal to a previously published map for olive baboons of 2293 cM for the autosome[69]. The Pyrho workflow consists of three steps: making a lookup table, inferring optimal hyperparameters, and then inferring the fine-scale recombination. Pyrho maketable was run with standard options except for the approximate setting for computational feasibility. The autosomal and chromosome X population size histories for Mikumi yellow baboons were used. Pyrho hyperparam was run with the following possible block penalties and window sizes: 10,25,50,100, and 10,25,50,100. Then, Pyrho optimize was run with optimal hyperparameters as measured by the L2 norm for each chromosome. Lastly, the autosome-wide recombination rate was scaled to be equal to the estimate of 2293cM published by [69], as Pyrho is optimized to infer fine-scale recombination and not total genetic distance. The scaling factor used was 1.3, as the total autosome-wide recombination rate inferred from pyrho was 1761. The assembly used for this estimate is Panubis1.0, while the one used in [18] is panu3. The recombination rate per window was calculated with the Pyrho genetic map, based on the genetic distance between the first and last base in the window, with interpolation performed assuming an average recombination rate between SNPs. Local Ancestry Inference was done using RFMix version 2, removing all alleles with a Minor Allele Frequency of less than 1% for computational efficiency. RFMix was run with the Pyrho genetic map and an assumed admixture date of 100 generations ago. Relate was run with a mutation rate of 0.57e-8 and an effective initial population size of 50000 for the autosomes, and chromosome X with a mutational rate of 0.46e-8 and an effective population size of 25000, in both cases with a generation time of 11 generations. The callability mask is the same as previously detailed. Otherwise, standard options were used for the Relate workflow, with the convert from VCF and prepare input script used to transform the VCF into haps/sample format, followed by Relate, EstimatePopulationSize, and then DetectSelection. We used weighted regressions to account for heteroscedasticity, as identified with the Breusch-Pagan test[70]. We used the statsmodels python package to run the regression models. The significance threshold is set at 0.005 to correct for multiple testing (Bonferroni correction) for the ten olive and yellow populations in Tanzania. Declarations Competing Interests Erik Fogh Sørensen and Kasper Munch have published an article together with Jeffrey Rogers (https://doi.org/10.1126/science.abn8153) Funding This work was funded by Novo Nordisk Foundation grant 0058553 (E.F.S. and K.M.). Author Contribution Conceptualization: All authors.Data curation: E.F.S.Investigation: E.F.S. and K.M.Formal analysis: E.F.S. and K.M.Supervision: G.H. and K.M.Visualization: E.F.S. (All Figures) and K.M. (Figure 1).Writing – original draft: E.F.S. and K.M.Writing – review & editing: All authors. Acknowledgement We thank Mikkel Heide Schierup and Dietmar Zinner for feedback on the manuscript, and Jeffrey Rogers, Christian Roos and Clifford J. Jolly for insightful discussion. Data Availability Data is sourced from accession code PRJEB59576 [18,71], which is publicly available in the European Nucleotide Archive. References Marques DA, Meier JI, Seehausen O. A Combinatorial View on Speciation and Adaptive Radiation. Trends Ecol Evol. 2019;34:531–44. Worsham MLD, Julius EP, Nice CC, Diaz PH, Huffman DG. Geographic isolation facilitates the evolution of reproductive isolation and morphological divergence. Ecol Evol. 2017;7:10278–88. Schumer M, Xu C, Powell DL, Durvasula A, Skov L, Holland C, et al. Natural selection interacts with recombination to shape the evolution of hybrid genomes. Science. 2018;360:656–60. Abbott R, Albach D, Ansell S, Arntzen JW, Baird SJE, Bierne N, et al. Hybridization and speciation*. J Evol Biol. 2013;26:229–46. Gompert Z, Parchman TL, Buerkle CA. Genomics of isolation in hybrids. Philosophical Transactions Royal Soc B Biological Sci. 2012;367:439–50. Poikela N, Laetsch DR, Kankare M, Hoikkala A, Lohse K. Experimental introgression in Drosophila: Asymmetric postzygotic isolation associated with chromosomal inversions and an incompatibility locus on the X chromosome. Mol Ecol. 2023;32:854–66. Cutter AD. The polymorphic prelude to Bateson–Dobzhansky–Muller incompatibilities. Trends Ecol Evol. 2012;27:209–18. Langmore NE, Grealy A, Noh H-J, Medina I, Skeels A, Grant J, et al. Coevolution with hosts underpins speciation in brood-parasitic cuckoos. Science. 2024;384:1030–6. Orr HA. Dobzhansky, Bateson, and the Genetics of Speciation. Genetics. 1996;144:1331–5. Presgraves DC. Evaluating genomic signatures of “the large X‐effect” during complex speciation. Mol Ecol. 2018;27:3822–30. Harris K, Nielsen R. The Genetic Cost of Neanderthal Introgression. Genetics. 2016;203:881–91. Zeller E, Timmermann A. The evolving three-dimensional landscape of human adaptation. Sci Adv. 2024;10:eadq3613. Vamosi SM, Schluter D. SEXUAL SELECTION AGAINST HYBRIDS BETWEEN SYMPATRIC STICKLEBACK SPECIES: EVIDENCE FROM A FIELD EXPERIMENT. Evolution. 1999;53:874–9. Sánchez-Ramírez S, Weiss JG, Thomas CG, Cutter AD. Sex-specific and sex-chromosome regulatory evolution underlie widespread misregulation of inter-species hybrid transcriptomes. Biorxiv. 2020;2020.05.04.076505. Larson EL, Vanderpool D, Sarver BAJ, Callahan C, Keeble S, Provencio LL, et al. The Evolution of Polymorphic Hybrid Incompatibilities in House Mice. Genetics. 2018;209:845–59. Larson EL, Keeble S, Vanderpool D, Dean MD, Good JM. The composite regulatory basis of the large X-effect in mouse speciation. Mol Biol Evol. 2016;msw243. Schield DR, Scordato ESC, Smith CCR, Carter JK, Cherkaoui SI, Gombobaatar S, et al. Sex‐linked genetic diversity and differentiation in a globally distributed avian species complex. Mol Ecol. 2021;30:2313–32. Sørensen EF, Harris RA, Zhang L, Raveendran M, Kuderna LFK, Walker JA, et al. Genome-wide coancestry reveals details of ancient and recent male-driven reticulation in baboons. Science. 2023;380:eabn8153. Chiou KL, Bergey CM, Burrell AS, Disotell TR, Rogers J, Jolly CJ, et al. Genome‐wide ancestry and introgression in a Zambian baboon hybrid zone. Mol Ecol. 2021; Vilgalys TP, Fogel AS, Anderson JA, Mututua RS, Warutere JK, Siodi IL, et al. Selection against admixture and gene regulatory divergence in a long-term primate field study. Science. 2022;377:635–41. Rogers J, Raveendran M, Harris RA, Mailund T, Leppälä K, Athanasiadis G, et al. The comparative genomics and complex population history of Papio baboons. Sci Adv. 2019;5:eaau6947. Kopp GH, Sithaldeen R, Trede F, Grathwol F, Roos C, Zinner D. A Comprehensive Overview of Baboon Phylogenetic History. Genes. 2023;14:614. Jolly CJ. Philopatry at the frontier: A demographically driven scenario for the evolution of multilevel societies in baboons (Papio). J Hum Evol. 2020;146:102819. Zinner D, Wertheimer J, Liedigk R, Groeneveld LF, Roos C. Baboon phylogeny as inferred from complete mitochondrial genomes. Am J Phys Anthropol. 2013;150:133–40. Skov L, Macià MC, Lucotte E, Cavassim MIA, Castellano D, Schierup MH, et al. Extraordinary selection on the human X chromosome associated with archaic admixture. Biorxiv. 2022;2022.09.19.508556. Lucotte EA, Skov L, Jensen JM, Macià MC, Munch K, Schierup MH. Dynamic Copy Number Evolution of X- and Y-Linked Ampliconic Genes in Human Populations. Genetics. 2018;209:907–20. Vicoso B, Charlesworth B. Effective Population Size and the Faster-X Effect: An Extended Model. Evolution. 2009;63:2413–26. Meisel RP, Connallon T. The faster-X effect: integrating theory and data. Trends Genet. 2013;29:537–44. Pool JE, Nielsen R. Inference of Historical Changes in Migration Rate From the Lengths of Migrant Tracts. Genetics. 2009;181:711–9. Keinan A, Reich D. Can a Sex-Biased Human Demography Account for the Reduced Effective Population Size of Chromosome X in Non-Africans? Mol Biol Evol. 2010;27:2312–21. Larson EL, Keeble S, Vanderpool D, Dean MD, Good JM. The composite regulatory basis of the large X-effect in mouse speciation. Molecular biology and evolution. 2016;msw243. Rathje C, Johnson E, Drage D, Patinioti C, Silvestri G, Affara N, et al. Differential Sperm Motility Mediates the Sex Ratio Drive Shaping Mouse Sex Chromosome Evolution. Curr Biol. 2019;29:3692-3698.e4. Hughes JF, Skaletsky H, Pyntikova T, Koutseva N, Raudsepp T, Brown LG, et al. Sequence analysis in Bos taurus reveals pervasiveness of X–Y arms races in mammalian lineages. Genome Res. 2020; Skov L, Macià MC, Lucotte EA, Cavassim MIA, Castellano D, Schierup MH, et al. Extraordinary selection on the human X chromosome associated with archaic admixture. Cell Genom. 2023;100274. Dutheil JY, Munch K, Nam K, Mailund T, Schierup MH. Strong Selective Sweeps on the X Chromosome in the Human-Chimpanzee Ancestor Explain Its Low Divergence. PLOS Genetics. 2015;11:e1005451. Nam K, Munch K, Hobolth A, Dutheil J, Veeramah KR, Woerner AE, et al. Extreme selective sweeps independently targeted the X chromosomes of the great apes. Proceedings of the National Academy of Sciences. 2015;112:6413–8. Nam K, Munch K, Mailund T, Nater A, Greminger M, Krützen M, et al. Evidence that the rate of strong selective sweeps increases with population size in the great apes. Proceedings of the National Academy of Sciences. 2017;114:201605660. Veller C, Edelman NB, Muralidhar P, Nowak MA. Recombination and selection against introgressed DNA. Evolution. 2023;77:1131–44. Matheson J, Masel J. Background Selection From Unlinked Sites Causes Nonindependent Evolution of Deleterious Mutations. Genome Biol Evol. 2024;16:evae050. Liu KJ, Steinberg E, Yozzo A, Song Y, Kohn MH, Nakhleh L. Interspecific introgressive origin of genomic diversity in the house mouse. Proc Natl Acad Sci. 2015;112:196–201. Vilgalys TP, Fogel AS, Mututua RS, Warutere JK, Siodi L, Anderson JA, et al. Selection against admixture and gene regulatory divergence in a long-term primate field study. Biorxiv. 2021;2021.08.19.456711. Hellenthal G, Busby GBJ, Band G, Wilson JF, Capelli C, Falush D, et al. A Genetic Atlas of Human Admixture History. Science. 2014;343:747–51. Wangkumhang P, Greenfield M, Hellenthal G. An efficient method to identify, date and describe admixture events using haplotype information. Biorxiv. 2021;2021.08.12.455263. Maples BK, Gravel S, Kenny EE, Bustamante CD. RFMix: A Discriminative Modeling Approach for Rapid and Robust Local-Ancestry Inference. Am J Hum Genetics. 2013;93:278–88. Massarat AR, Lamkin M, Reeve C, Williams AL, D’Antonio M, Gymrek M. Haptools: a toolkit for admixture and haplotype analysis. Bioinformatics. 2023;39:btad104. Spence JP, Song YS. Inference and analysis of population-specific fine-scale recombination maps across 26 diverse human populations. Sci Adv. 2019;5:eaaw9206. Kamm JA, Spence JP, Chan J, Song YS. Two-Locus Likelihoods Under Variable Population Size and Fine-Scale Recombination Rate Estimation. Genetics. 2016;203:1381–99. Ronquist F, Kudlicka J, Senderov V, Borgström J, Lartillot N, Lundén D, et al. Universal probabilistic programming offers a powerful approach to statistical phylogenetics. Commun Biology. 2021;4:244. Terhorst J, Kamm JA, Song YS. Robust and scalable inference of population history from hundreds of unphased whole genomes. Nat Genet. 2017;49:303–9. Pool JE, Nielsen R. POPULATION SIZE CHANGES RESHAPE GENOMIC PATTERNS OF DIVERSITY. Evolution. 2007;61:3001–6. Charlesworth B, Jensen JD. Effects of Selection at Linked Sites on Patterns of Genetic Variability. Annu Rev Ecol, Evol, Syst. 2021;52:177–97. Ellegren H, Galtier N. Determinants of genetic diversity. Nat Rev Genet. 2016;17:422–33. Burri R, Nater A, Kawakami T, Mugal CF, Olason PI, Smeds L, et al. Linked selection and recombination rate variation drive the evolution of the genomic landscape of differentiation across the speciation continuum of Ficedula flycatchers. Genome Res. 2015;25:1656–65. Nordborg M, Charlesworth B, Charlesworth D. The effect of recombination on background selection*. Genet Res. 1996;67:159–74. Wu FL, Strand AI, Cox LA, Ober C, Wall JD, Moorjani P, et al. A comparison of humans and baboons suggests germline mutation rates do not track cell divisions. Plos Biol. 2020;18:e3000838. Bergeron LA, Besenbacher S, Zheng J, Li P, Bertelsen MF, Quintard B, et al. Evolution of the germline mutation rate across vertebrates. Nature. 2023;615:285–91. Bergeron LA, Besenbacher S, Bakker J, Zheng J, Li P, Pacheco G, et al. The germline mutational process in rhesus macaque and its implications for phylogenetic dating. GigaScience. 2021;10:giab029. Pool JE, Nielsen R. POPULATION SIZE CHANGES RESHAPE GENOMIC PATTERNS OF DIVERSITY. Evolution. 2007;61:3001–6. Alberts SC, Watts HE, Altmann J. Queuing and queue-jumping: long-term patterns of reproductive skew in male savannah baboons, Papio cynocephalus. Anim Behav. 2003;65:821–40. Alberts SC, Buchan JC, Altmann J. Sexual selection in wild baboons: from mating opportunities to paternity success. Anim Behav. 2006;72:1177–96. Alberts SC. Social influences on survival and reproduction: Insights from a long‐term study of wild baboons. J Anim Ecol. 2019;88:47–66. Städele V, Roberts ER, Barrett BJ, Strum SC, Vigilant L, Silk JB. Male–female relationships in olive baboons (Papio anubis): Parenting or mating effort? J Hum Evol. 2019;127:81–92. Kopp GH, Silva MJF da, Fischer J, Brito JC, Regnaut S, Roos C, et al. The Influence of Social Systems on Patterns of Mitochondrial DNA Variation in Baboons. Int J Primatol. 2014;35:210–25. Hubisz MJ, Williams AL, Siepel A. Mapping gene flow between ancient hominins through demography-aware inference of the ancestral recombination graph. Plos Genet. 2020;16:e1008895. Jiang J, Yu J, Li J, Li P, Fan Z, Niu L, et al. Mitochondrial Genome and Nuclear Markers Provide New Insight into the Evolutionary History of Macaques. PLoS ONE. 2016;11:e0154665. McVicker G, Gordon D, Davis C, Green P. Widespread Genomic Signatures of Natural Selection in Hominid Evolution. Plos Genet. 2009;5:e1000471. Loh P-R, Lipson M, Patterson N, Moorjani P, Pickrell JK, Reich D, et al. Inferring Admixture Histories of Human Populations Using Linkage Disequilibrium. Genetics. 2013;193:1233–54. Baran Y, Pasaniuc B, Sankararaman S, Torgerson DG, Gignoux C, Eng C, et al. Fast and accurate inference of local ancestry in Latino populations. Bioinformatics. 2012;28:1359–67. Wall JD, Robinson JA, Cox LA. High-Resolution Estimates of Crossover and Noncrossover Recombination from a Captive Baboon Colony. Genome Biol Evol. 2022;14:evac040. Breusch TS, Pagan AR. A Simple Test for Heteroscedasticity and Random Coefficient Variation. Econometrica. 1979;47:1287. Kuderna LFK, Gao H, Janiak MC, Kuhlwilm M, Orkin JD, Bataillon T, et al. A global catalog of whole-genome diversity from 233 primate species. Science. 2023;380:906–13. Additional Declarations Competing interest reported. Erik Fogh Sørensen and Kasper Munch have published an article together with Jeffrey Rogers ( https://doi.org/10.1126/science.abn8153 ) Supplementary Files Supplementaryandfiguresbaboonadmixture.docx Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 01 May, 2026 Reviews received at journal 28 Apr, 2026 Reviewers agreed at journal 10 Apr, 2026 Reviewers agreed at journal 14 Dec, 2025 Reviews received at journal 13 Nov, 2025 Reviewers agreed at journal 06 Oct, 2025 Reviewers invited by journal 01 Oct, 2025 Editor assigned by journal 26 Aug, 2025 Submission checks completed at journal 26 Aug, 2025 First submitted to journal 25 Aug, 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-7456420","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":525495550,"identity":"da16a3cd-e6e1-429c-bd9b-abb34002bd88","order_by":0,"name":"Erik Fogh Sørensen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvUlEQVRIiWNgGAWjYFACxmYGhooDPFAeswGRWs6QpoWBmYGx7QCcQ1gLv/ThZsOf8+7ImLcfYPzMw2BtTFCLZF9iczLvtmc8MmcSmKV5GNLNCGoxOMPYfJhx22EeCQkGBqCWwzYEtdgDtRz8OQeshfk3UVoMeBibE3gbwFrYQLYQdpgE0BZjnmPPeCR4Etss5xikE/Y+fw/7Y8kfNXfsJdgPH77xpsLasIGgHgRgBComKiJHwSgYBaNgFBAEADMuMiiwylyBAAAAAElFTkSuQmCC","orcid":"","institution":"Aarhus University","correspondingAuthor":true,"prefix":"","firstName":"Erik","middleName":"Fogh","lastName":"Sørensen","suffix":""},{"id":525495553,"identity":"948539c9-9591-46d1-a17b-aa7a1f06540c","order_by":1,"name":"Garrett Hellenthal","email":"","orcid":"","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Garrett","middleName":"","lastName":"Hellenthal","suffix":""},{"id":525495554,"identity":"4ace1e80-93a3-4b8e-a767-327b13c7ae75","order_by":2,"name":"Kasper Munch","email":"","orcid":"","institution":"Aarhus University","correspondingAuthor":false,"prefix":"","firstName":"Kasper","middleName":"","lastName":"Munch","suffix":""}],"badges":[],"createdAt":"2025-08-25 18:53:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7456420/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7456420/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93553413,"identity":"e86f526d-bd0f-4a2e-a0e1-85165bdd4b62","added_by":"auto","created_at":"2025-10-15 06:03:48","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1557866,"visible":true,"origin":"","legend":"","description":"","filename":"NegativeselectiononbaboonadmixtureisstrongestonchromosomeX.docx","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/8a5d18f0e5050513906f5e06.docx"},{"id":93553300,"identity":"8d6af97a-7d7e-4962-a341-5918743a94f6","added_by":"auto","created_at":"2025-10-15 05:55:46","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5705,"visible":true,"origin":"","legend":"","description":"","filename":"8332fe13612f4c0cb0be579d44d6004f.json","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/779a5bc2d9464d0c19ed1499.json"},{"id":93552634,"identity":"449fe4ef-5fd1-4d97-9050-20fbfaf50931","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1771735,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryandfiguresbaboonadmixture.docx","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/a8835fb81a11ae9963135447.docx"},{"id":93552632,"identity":"9a3f66d4-e154-4cac-a4cb-02ff92395b90","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":84520,"visible":true,"origin":"","legend":"","description":"","filename":"8332fe13612f4c0cb0be579d44d6004f1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/2eb8bff4c5b28541e6e0a1a0.xml"},{"id":93553411,"identity":"25e9b493-cb83-42d8-9252-d561cb737cf5","added_by":"auto","created_at":"2025-10-15 06:03:46","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":368062,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/00ac1988b98f9c3e8e4a31ac.jpeg"},{"id":93552638,"identity":"2bbefa3f-72be-4101-8a7a-8ad44a97b693","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105904,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/94411b51ca8438048d5fd26c.jpeg"},{"id":93552639,"identity":"23ad8788-7a53-415c-82b4-adfa1df1d425","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":155984,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/513cda1eff55560fa12404f0.png"},{"id":93552640,"identity":"19fa1191-e2e0-48d1-a451-b0f3b5e2ed96","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":146336,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/d6986eecf2715e3d9620d70d.png"},{"id":93553302,"identity":"4dfcaff7-d25d-4f97-8db3-c74da8d706f2","added_by":"auto","created_at":"2025-10-15 05:55:46","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58214,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/cde9d73b5c7c24739a72f481.png"},{"id":93552651,"identity":"7dbd7530-bd1f-4088-9244-87603cc9398c","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":828105,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/dcb04e270fd43b7ecd88f352.jpeg"},{"id":93552644,"identity":"c798739d-a87a-471a-8fbb-67d0d0f40ab4","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":139249,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/693ea672df96395ae6c8adee.png"},{"id":93552655,"identity":"dadb2212-1217-47ac-9e69-8c6d33242064","added_by":"auto","created_at":"2025-10-15 05:47:47","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":214284,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/89eba937d4aac5f6a855672e.png"},{"id":93552647,"identity":"8f314507-f1aa-475c-b566-78b7f18fe7e2","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":314055,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/af64bb096e28441f3602fab7.png"},{"id":93552649,"identity":"00866547-4e7d-4eb2-a734-ab94595bc20a","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56400,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/187fbaf5ab2aeaa768ccc90a.png"},{"id":93552646,"identity":"e6043cff-1172-48e6-81d7-ee88d5ed0c35","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36400,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/41b9917250de8dd2ff979c46.png"},{"id":93553306,"identity":"fcfc903f-a01a-4ca3-b28b-ff83ba3dd7d8","added_by":"auto","created_at":"2025-10-15 05:55:46","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":44001,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/d1b7ed362476a458ab55e4f0.png"},{"id":93553304,"identity":"6ad3aad8-8f7c-46f5-8be6-6feaed31d3c2","added_by":"auto","created_at":"2025-10-15 05:55:46","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26901,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/cb58f70fdccbc7939114225b.png"},{"id":93552652,"identity":"9e01ec37-e87c-41f4-9d0b-f1f5aa1e5ca2","added_by":"auto","created_at":"2025-10-15 05:47:47","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":501716,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/ea55b2456b85fc79f0863bcb.png"},{"id":93552648,"identity":"82d67025-9da4-4432-8937-3f5143ea841e","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38296,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/82985402f1f9fe35366423d6.png"},{"id":93552636,"identity":"a05b9065-d397-43a0-a227-c737f010547d","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":62599,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/82c79832ba8f0381b67e6581.png"},{"id":93553307,"identity":"6ff618dd-127c-4459-871f-a338b443906d","added_by":"auto","created_at":"2025-10-15 05:55:46","extension":"xml","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":83291,"visible":true,"origin":"","legend":"","description":"","filename":"8332fe13612f4c0cb0be579d44d6004f1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/97c0d9dea2a34b8c52bc885b.xml"},{"id":93552641,"identity":"fab48b0e-63e1-4293-a722-71ad80e99753","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"html","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":88506,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/451d19db93b04ce13aa72a19.html"},{"id":93552624,"identity":"91ca5c1b-43fb-4f04-b2c9-e76c13df986e","added_by":"auto","created_at":"2025-10-15 05:47:45","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":368062,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eOverview of sampling locations and RFMix ancestry paintings of each individual's phased chromosome 1 haplotypes. Olive and yellow colors represent northern and southern baboon ancestry, respectively. Pie charts at the sample locations show the MPA proportion of each population. The current olive/yellow baboon contact zone is shown with a dashed line. Dotted lines on Africa and insert maps show the extent of the mitochondrial G clade spanning the East African populations of yellow, olive, and hamadryas (hamadryas range shown in grey).\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/eb80f533a1e24ff1971a2e67.jpeg"},{"id":93552626,"identity":"9bf32f55-8c83-49ae-82ff-aeb2404df4dd","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":105904,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eWeighted Linear Regressions for MPA proportion against recombination rate (A, B) and background diversity (C, D). Tarangire showed no correlation and is excluded for legibility. Tarangire is left out for improved legibility of the other populations, see Supplementary Figure 2 with Tarangire included.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/a5045462682ad34660e08fbe.jpeg"},{"id":93553412,"identity":"5b2f9822-dcc8-453d-865a-cef41e78de45","added_by":"auto","created_at":"2025-10-15 06:03:46","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":197284,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eA-B) Weighted Linear Regression of MPA proportion against background diversity on the X chromosome for Tanzanian yellow and olive baboons, respectively. Tarangire is omitted for improved legibility of the other populations; see Supplementary Figure 9 with Tarangire included.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/0c6cfc5832eecf7506978536.png"},{"id":93553298,"identity":"aa81fba1-baa3-46f7-8d0a-b7b0e0f4de4d","added_by":"auto","created_at":"2025-10-15 05:55:46","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":395039,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eA truncated, unnormalized plot of background diversity and MPA on the autosomes and chromosome X for all Tanzanian baboon populations except Tarangire.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/07edc619c5ba19b4ed9ac69b.png"},{"id":93552628,"identity":"00a2eecb-f8ef-4c86-b52b-7aec3c01457c","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":216222,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eBarplot of autosomal and X-linked Minor Parent Ancestry Percentage for yellow and olive baboons in Tanzania.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/a4fd6a6ab894db487a5cfd13.png"},{"id":93553303,"identity":"06de7765-02ab-4b40-ae53-31a2272c34e0","added_by":"auto","created_at":"2025-10-15 05:55:46","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":828105,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA) \u003c/strong\u003e\u003cem\u003eRFMix ancestry paintings of each individual's phased chromosome X haplotypes. Olive and yellow colors represent northern and southern baboon ancestry. The haplotypes are clustered on similarity using UPGMA and otherwise sorted by sampling location. Windows with less than 75 % callability are filtered out. \u003c/em\u003e\u003cstrong\u003eB) \u003c/strong\u003e\u003cem\u003eSame as A), but here the similarly sized chromosome 8 for comparison. Note that there are fewer haplotypes for Chromosome X because males only carry one.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/0aacf261b4ffb71fbb961a84.jpeg"},{"id":93552650,"identity":"b56e9331-6359-49ad-b70f-a44244cd7e98","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":255117,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA)\u003c/strong\u003e\u003cem\u003e Weighted Linear Regressions of MPA proportion along Gog olive autosome against recombination rate.. \u003c/em\u003e\u003cstrong\u003eB)\u003c/strong\u003e\u003cem\u003eWeighted Linear Regressions of MPA proportion along autosomes and the X chromosome against background diversity.\u003c/em\u003e\u003cstrong\u003eC)\u003c/strong\u003e\u003cem\u003eMPA proportions for autosomes and the X chromosome in bins of background diversity. \u003c/em\u003e\u003cstrong\u003eD)\u003c/strong\u003e\u003cem\u003e Histogram of log-scaled MPA proportion across autosomes and the X chromosome.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/48df7fb748036d1fa662af0c.png"},{"id":93552645,"identity":"839b6693-99df-45c1-aa41-45a864867729","added_by":"auto","created_at":"2025-10-15 05:47:46","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":560908,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eRFMix paintings of chrX and chr8 for the Gog olives. The haplotypes are ordered based on their similarity using UPGMA, and clustered based on the sampling location. Windows with less than 75 % callability are filtered out. Note that there are fewer haplotypes for Chromosome X, as males only contain 1 copy instead of 2.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/3702992250bff817bd1781bc.png"},{"id":93554337,"identity":"223e643b-b20a-4a76-9389-dcadeb449fb5","added_by":"auto","created_at":"2025-10-15 06:11:50","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3385959,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/a30a3cb1-d045-4d65-a11a-41d15652b9ac.pdf"},{"id":93553296,"identity":"7eec2872-0413-4b80-80bc-7f7f33abb3bc","added_by":"auto","created_at":"2025-10-15 05:55:45","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1771735,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryandfiguresbaboonadmixture.docx","url":"https://assets-eu.researchsquare.com/files/rs-7456420/v1/14246e81ab62ab23aae0ab2c.docx"}],"financialInterests":"Competing interest reported. Erik Fogh Sørensen and Kasper Munch have published an article together with Jeffrey Rogers (https://doi.org/10.1126/science.abn8153)","formattedTitle":"Negative selection on baboon admixture is strongest on chromosome X","fulltext":[{"header":"Background","content":"\u003cp\u003eSpeciation is a gradual process in which populations diverge because of geographic, behavioural or ecological isolation. This isolation fosters differentiation as populations accumulate unique mutations and experience distinct selective pressures and genetic drift[1\u0026ndash;4]. In cases of prolonged isolation, genetic incompatibilities can emerge[5\u0026ndash;7], eventually leading to complete reproductive isolation, preventing further admixture[8\u0026ndash;10].\u003c/p\u003e\u003cp\u003eVariants experience selection if they are less fit. This can occur if they arise in an inbred population[11], are maladapted to the local environment[2,12], or cause sexual incompatibilities, reducing the reproductive success of hybrids[4,13,14]. A more complex source of selection is genetic incompatibility in the form of Bateson\u0026ndash;Dobzhansky\u0026ndash;Muller Incompatibility (BDMI), which may arise from negative epistatic interactions when genes of different ancestry are mixed in hybrids. In that case, negative selection is expected to gradually remove the minor parent ancestry (MPA). Selection against admixture from the rarer ancestry across swordfish populations has been reported as evidence of BDMI[3] by detecting negative selection on admixture[6,15\u0026ndash;17].\u003c/p\u003e\u003cp\u003eIn baboons (genus Papio), hybridization between distinct populations is commonly observed. Still, our previous report[18] showed that genetic variants from foreign ancestry have to travel through the hybrid zone before reaching populations further from the hybrid zone[18\u0026ndash;20], even in the face of extensive ongoing admixture. With its multiple species, the baboon species complex thus presents an ideal system for investigating negative selection mechanisms on admixture. Baboons have six distinct species, with the deepest split dated at least 1.3\u0026nbsp;million years ago between southern baboons (yellow, kinda, and chacma baboons; P. cynocephalus, P. kindae, and P. ursinus) and northern baboons (olive, hamadryas, and guinea baboons; P. anubis, P. hamadryas, and P. papio)[18,21].\u003c/p\u003e\u003cp\u003eBaboon species exhibit a striking discordance between autosomal and mitochondrial phylogenies[22], with a mitochondrial clade extending across olive, yellow, and hamadryas baboon ranges in East Africa (dotted lines in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), even though olive and yellow baboons also each carry species-specific, deeply diverged mitochondria in other parts of their species range. Our recent analysis[18] supported that this discordance arose when an ancient widespread hamadryas-like baboon was subsumed in male-driven range expansions of yellow and olive baboons, displacing the nuclear genome of the resident population but not its mitochondria. In this process, often called nuclear swamping[23,24], continuous displacement of original ancestry will result in the invading species going from minor to major ancestry. Any hybrid incompatibility would thus initially manifest as negative selection on the invading ancestry. As the invading ancestry reaches the majority, selection would change to favour the invading ancestry. The relative strengths of migration and negative selection on the minor parent ancestry thus determine which ancestry ends up fixed and could result in swamped populations retaining high-frequency original ancestry at incompatibility loci.\u003c/p\u003e\u003cp\u003eHaldane\u0026rsquo;s rule that genetic incompatibility first emerges in the hemizygous sex is supported by a century of observations. It suggests an outsize role of the mammalian X chromosomes in hybrid incompatibility[10,25\u0026ndash;28]. The smaller effective population size (Ne) of the X chromosome results in faster divergence by genetic drift, a larger sensitivity to recent changes in population size[29], and sensitivity to sex-biased dispersion and reproductive variance[27,30]. Dynamic co-amplification of multicopy ampliconic genes in \u003cem\u003eMus musculus\u003c/em\u003e and \u003cem\u003eBos taurus\u003c/em\u003e suggests that mammalian X and Y chromosomes participate in an arms race to control the X/Y ratio of fertile sperm[31\u0026ndash;33]. This hypothesis predicts an accelerated evolution supported by multiple reports of positive selection on primate X chromosomes[34\u0026ndash;37].\u003c/p\u003e\u003cp\u003eIf admixture is neutral, it will be distributed randomly across the genome. On the other hand, if linked selection purges specific admixture tracts, there will be an association between proxies for linked selection efficacy and admixture levels[3,20]. This correlation can be interpreted as a sign of selection against admixture if regions with high potential linked selection carry less admixture than neutral regions. The rate of admixture removal will be highest in the first generations, where large numbers of negatively selected variants remain linked. However, the correlation with recombination rate and MPA will not be strong for very recent admixture, where introgressed segment sizes remain orders of magnitude larger than that of fine-scale recombination rate and the 100kb scale at which we quantify MPA[38]. In these initial generations, removal will instead be much more indiscriminate due to the high linkage across the chromosomes[11,39]. We expect the strongest association between recombination rate and MPA as admixture disperses behind the hybrid zone and recombination reduces segment lengths to sizes relevant to medium-scale recombination rates and selection coefficients. However, the correlation will gradually weaken based on time since admixture, as the deleterious minor parent ancestry variants are removed and the MPA proportion is reduced.\u003c/p\u003e\u003cp\u003eA previous study by Vilgalys et al. on yellow and olive baboons found evidence of selection against admixture in a baboon hybrid zone in the Amboseli basin[20]. They found that gene-dense and low recombination regions had significantly less admixture than expected. If the introgressed sequence carries variants under negative selection, we expect that more admixture is retained in high-recombination regions where neutral introgressed sequence segments are more readily unlinked from the deleterious variants that are eventually purged from the population[3,20]. In this hybrid zone, baboons with older admixture account for most of this correlation. In contrast, recent hybrids did not show this correlation pattern, indicating that it takes at least 4\u0026ndash;5 generations for introgressing chromosomes to break into segments of a size similar to the scale of recombination rate and gene density variation[38].\u003c/p\u003e\u003cp\u003eThis study aims to elucidate the dynamics of selection against admixture in the baboon species complex, focusing on identifying genome-wide patterns and the role of the X chromosome. Chromosome X is known to have more admixture than the autosomes for both olive and yellow baboons[18], even though many other admixture cases exhibit decreased admixture on chromosome X[6,25,40]. We investigate the relationship between admixture and linked selection across eleven sampling locations with varying distances from hybrid zones. The current species boundary represents a secondary contact between olive and yellow baboons[22\u0026ndash;24,41]. Globetrotter was used to date the most recent admixture events to between 10 and 100 generations for the various populations[18,42,43] and found extensive migration between the olive baboon populations in Tanzania. The Amboseli population is similar to the Arusha and Tarangire populations, which all contain large amounts of minor parent ancestry. All three national parks are located near the boundary between olive and yellow baboons, with high variance in admixture proportions in sampled individuals. Improved reference population size allows for better admixture inference, as parentage of rare variants, frequency of common variants, and species-specific haplotype combinations are all better resolved with more samples. All individuals are 30X sequenced, allowing for higher accuracy and resolution methods, and are sampled across 19 localities. Sampling of a gradient of olive baboon populations shows how admixture patterns change as they penetrate deeper into the relatively pure olive baboon ranges. Unlike the previous studies[3,20], the X chromosome is also investigated, showing up to seven times stronger removal of minor parent ancestry in low recombination and low diversity regions.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eMinor Parent Ancestry declines with distance from the hybrid zone\u003c/p\u003e\u003cp\u003eWe infer local ancestry along each high-coverage P. anubis (olive) and P.cynocephalus (yellow) baboon genome using RFMix[44]. The four other baboon species were used as reference panels. We assign olive baboon ancestry (\"northern ancestry\") to genomic segments closer to P. hamadryas and P. papio (hamadryas and Guinea baboons) and yellow baboon ancestry (\"southern ancestry\") to segments closer to P. kindae and P. ursinus (Kinda and chacma baboons). We consider minor parent ancestry (MPA) the rarer ancestry in each population and represent the MPA proportion along the genome as its mean in nonoverlapping, contiguous 100-kilobase windows. To assess the accuracy of RFMix, we used Haptools[45] to simulate admixture from the current standing variation of two of the sampled populations. The Pearson correlation between the simulated and inferred admixture by RFMix is 98.9%, with no bias towards major parent ancestry and high diversity regions. See Supplementary Simulation of admixture for details.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe autosomal MPA proportion of sampled populations differs widely (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Supplementary Table\u0026nbsp;1). In yellow baboons, it ranges from 8.0% in Ruaha, closest to the hybrid zone, to 0.36% in Selous, furthest away from the hybrid zone. In olive baboons, the MPA proportion drops from 20.9% in the Tarangire population, which borders the yellow range, to 3.4% in the Serengeti population, located 250 km into the olive range. This relationship with distance from the hybridization zone suggests that the introgressed sequence is removed by selection as it disperses by animal movement between populations close to the species contact zone[22\u0026ndash;24,41].\u003c/p\u003e\u003cp\u003eSouthern ancestry is virtually absent in the Ethiopian Gog olive baboon population (autosomal MPA proportion 0.065%), which is located in Ethiopia, 1100 km from the olive-yellow hybrid zone. This proportion is consistent with no introgression from southern baboons into Gog olive baboons, as this small signal can be caused by incomplete lineage sorting and/or noise.\u003c/p\u003e\u003cp\u003eCorrelation with recombination rate reveals negative selection on MPA far behind hybrid zones\u003c/p\u003e\u003cp\u003eTo obtain fine-scale recombination rates, we created a genetic map with Pyrho [46,47] and SMC++ [48,49] using the 38 individuals from Mikumi. See Methods and Materials. The distribution of recombination in 100 kb windows has a rightward skew (see Supplementary Distribution and regressions with outliers), and 0.5% of 100kb windows have an estimated recombination rate above 0.408cM. When included, these outliers will have an outsized impact on the regressions. To restrict the analysis to a representative range of recombination rates, we removed the upper and lower 0.5 percentiles of recombination rate.\u003c/p\u003e\u003cp\u003eWe use variance-weighted linear regressions to account for heteroscedasticity (Supplementary Table\u0026nbsp;2 and Supplementary Fig.\u0026nbsp;1) and find a significant association between the population mean MPA proportion in 100kb windows and the 100kb window recombination rate. This is true for all populations (p-values\u0026thinsp;\u0026lt;\u0026thinsp;2e-5) except Tarangire, Arusha, and Lake Manyara (See Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and Supplementary Table\u0026nbsp;3), consistent with negative selection on MPA in both olive and yellow baboon populations far from the hybrid zone. However, in line with expectations, the correlation is absent in Tarangire and Arusha, where admixture was most recently introduced, which is evident as longer MPA segments in the local ancestry paintings shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Lake Manyara baboons show no significant correlation despite a MPA proportion smaller than the baboons in Ngorongoro and Ruaha.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eRuaha, the most admixed yellow population, also has the steepest regression slope (0.0414), significantly larger than the slope for the Selous, Mikumi, and Udzungwa populations (p-value 0.00067), which range from 0.0288 to 0.0325 (See Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, Supplementary Table\u0026nbsp;3). For an impression of the effect sizes, we ranked autosomal 100kb windows by their recombination rate and computed the mean MPA for windows in five equally sized bins ordered based on recombination rate (see Supplementary Fig.\u0026nbsp;3 and Supplementary Table\u0026nbsp;4). Comparing mean MPA proportions for Mikumi between the 20% windows with the lowest recombination rates to the 20% with the highest rates shows a difference of 0.25% (0.896\u0026ndash;1.15%), corresponding to a relative difference of 28%. Ruaha, the most admixed population, shows a similar absolute difference of 0.23 percentage points but a relative difference of only 3%.\u003c/p\u003e\u003cp\u003eThe degree of admixture also varies among Tanzanian olive baboon populations (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB and Supplementary Table\u0026nbsp;1). The Gombe population has the steepest regression slope of 0.0633 (p-value 5.6e-19) for recombination rate and MPA proportion, and the Serengeti and Ngorongoro populations also have significant regression slopes (p-values 4.07e-07 and 0.00177). We find no significant correlation between recombination rate and MPA proportion in Lake Manyara after Bonferroni correction (p-value 0.0306). The populations in Tarangire and Arusha at the active contact zone do not show a significant relationship between recombination rate and MPA. Arusha has the largest absolute and relative difference between low and high-recombination windows (6.82% in the lowest 20\u0026ndash;7.22% in the highest 20%, a 6% increase). Ngorongoro has 4.92% MPA in the lowest 20\u0026ndash;5.16% MPA in the highest 20%, a 5% increase. See Supplementary Table\u0026nbsp;4 for all quintile differences.\u003c/p\u003e\u003cp\u003eCorrelation with diversity offers a better proxy for negative selection on MPA\u003c/p\u003e\u003cp\u003eAs a proxy for linked selection, the recombination rate does not account for variation in the density of sites under negative or positive selection. A more direct quantification of linked selection is relative genetic diversity, measured as π, the mean pairwise differences among haplotypes in the population. Although recent bottlenecks may exacerbate variation in diversity along the chromosomes[50], the relative degree of diversity is primarily shaped by linked selection[39,51\u0026ndash;54] and variation in mutation rate across the genome[55\u0026ndash;57].\u003c/p\u003e\u003cp\u003eHowever, as admixture itself contributes to 100kb mean diversity, we must obtain independent estimates of π from \u0026ldquo;background\u0026rdquo; populations, which do not share historical admixture events with the studied population. In the absence of shared admixture in the background populations, this estimate of π quantifies linked selection as a proxy for the functional importance of each 100kb window. If independent admixture contributes to background population diversity, its distribution across 100kb windows will additionally reflect the strength of negative selection on MPA; i.e., diversity measured in admixed background populations is expected to correlate more strongly with MPA proportion in the study population if they also experience similar selective events[18]. We calculated the average diversity across Guinea, Hamadryas, Kinda, and Chacma baboon reference populations in 100kb windows and will refer to this as \u0026ldquo;background diversity\u0026rdquo; below. (See Supplementary Fig.\u0026nbsp;4 for distribution). Background diversity strongly, but not perfectly, correlates with the recombination rate (Pearson correlation: 0.602, p-value: 0.0, Supplementary Fig.\u0026nbsp;5).\u003c/p\u003e\u003cp\u003eAmong yellow baboons, the Ruaha population again shows the steepest regression slope between background diversity and MPA (9.25, p-value 1.91e-19), significantly higher than other yellow populations (p-value: 1.01e-10), while the Mikumi (slope 2.19, p-value 1.21e-16) and Udzungwa populations also have significant regression slopes and slope 3.79, p-value 9.51e-18) (Supplementary Table\u0026nbsp;5). Further from the hybrid zone, the Selous population shows no significant correlation after Bonferroni correction to account for testing multiple populations (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). In all populations, the difference in MPA proportions between the bottom and top 20% windows ranked by background diversity is larger than when ranking by recombination rate. In Ruaha (see Supplementary Fig.\u0026nbsp;6 and Supplementary Table\u0026nbsp;6), the absolute difference is 16% (7.68\u0026ndash;8.92%), and Udzungwa shows the largest relative difference, 44%, between the bottom and top 20% windows (1.22\u0026ndash;1.76%).\u003c/p\u003e\u003cp\u003eAmong the olive baboon populations, the correlation between MPA proportion and background diversity is significantly stronger in populations away from the hybrid zone. Gombe (p-value 4.35e-34) and Ngorongoro (p-value 4.03e-13) show the strongest correlation between background diversity and MPA (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). The Gombe population has the largest absolute and relative difference in mean MPA proportions for windows in the bottom and top diversity quintiles (MPA 4.46% in the lowest quintile and 5.74% in the highest, a 29% increase). The Ngorongoro population has the second largest absolute difference (5.14% in the lowest 20\u0026ndash;5.65% in the highest 20%, a 10% increase). The Tarangire and Arusha populations closest to the contact zone show no significant correlation between background diversity and MPA.\u003c/p\u003e\u003cp\u003eTo formally compare the predictive strengths of background diversity and recombination, we z-normalized them and used a generalized linear model to explain MPA proportion as a function of both recombination rate and background diversity (see Supplementary Table\u0026nbsp;7 for all coefficients and p-values). The Ruaha, Gombe, Lake Manyara, and Serengeti baboon populations all had a significant negative association with recombination for MPA prediction (p-values 0.000113, 2.43e-07, 2.78e-07, and 6.86e-07) while having a larger positive slope based on background diversity. As an example of the different regression slopes, the slope of the Ruaha population, when using only recombination, was 0.0414. When using only background diversity, it was 9.25, whilst the regression using both had slopes of -0.0559 and 10.5, respectively. That is, while the recombination rate is positively correlated with the MPA proportion, its correlation becomes negative when modelled together with background diversity. This may in part reflect that genomic regions with very low recombination rates have more admixture than expected when adjusting for background diversity, indicating that admixture might persist more easily in low-recombination, high-diversity regions than in high-recombination, high-diversity regions. The correlation between mutation rate and recombination rate allows diversity to incorporate most of the recombination rate\u0026rsquo;s predictive power.\u003c/p\u003e\u003cp\u003eChromosome X diversity is strongly reduced\u003c/p\u003e\u003cp\u003eIn all studied populations, the diversity on chromosome X relative to autosomes is significantly lower than the \u0026frac34; expected from hemizygosity[58]. This is partly explained by recent bottlenecks experienced by many populations (See Supplementary Fig.\u0026nbsp;7), most severely in northern baboons, which more strongly affect chromosome X diversity. However, even after adjusting for the effect of historical population sizes and the lower mutation rate on chromosome X[55], the X/autosome ratio remains significantly below \u0026frac34; (See Supplementary Table\u0026nbsp;8) (Mann-Whitney U test, p-values from 1.78e-26 to 2.17e-272). Adjusted X/autosome ratios range from 0.346 to 0.521, corresponding to ratios between 73.4% and 88.2% of the expected values. All baboon species exhibit higher reproductive variance in males[23,27,59\u0026ndash;63], which should increase rather than reduce the X/autosome ratio, leaving only stronger linked selection on the X chromosome and sex-biased demographics[30] to explain the lower ratios. See Supplementary Fig.\u0026nbsp;8 for the distribution of background diversity on chromosome X.\u003c/p\u003e\u003cp\u003eLinked selection on MPA is up to seven times stronger on the X chromosome\u003c/p\u003e\u003cp\u003eOn the X chromosome, all olive baboon populations except Tarangire and Arusha show significant correlations of MPA proportion with both background diversity (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB and Supplementary Table\u0026nbsp;9) and recombination rate (Supplementary Table\u0026nbsp;10). The Ngorongoro population has an association between background diversity and MPA proportion that is an order of magnitude larger than that for the autosomes (slope 81.1, p-value 8.19e-09). Here, the mean MPA proportion goes from 0.47\u0026ndash;7.27% between the bottom and top 20% 100kb windows ranked by background diversity, a 14.4-fold difference. The similarly strong association in the Serengeti olive baboons (slope 68, p-value 2.47e-08) shows a span of 0.70% in the lowest quintile to 6.37% in the highest quintile in MPA proportion, an 8.04-fold difference.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAmong the yellow baboon populations (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, Table\u0026nbsp;3), only the Ruaha population shows a significant association between background diversity and MPA proportion (slope 156, p-val 1.79e-16). Here, the 100kb windows with a background diversity in the top and bottom quintiles are 7.94% and 20.1%, a 2.5-fold difference (See Supplementary Fig.\u0026nbsp;10 and Supplementary Table\u0026nbsp;11). The overall admixture in low-diversity regions is similar among all four yellow baboon populations. Instead, the Ruaha population has much more admixture in high-diversity regions than the other yellow baboons. (See Supplementary Tables\u0026nbsp;9 and 10 for results on all populations).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo formally test whether the correlation between background diversity and MPA proportion is stronger on the X chromosome than on autosomes, we Z-score normalized background diversity and performed a weighted linear regression with MPA proportion against normalized diversity, chromosome type, and their interaction, keeping the variance weights the same as in the previous models. A linear model with an interaction term can determine whether there is a significant difference between the regression slopes for the X chromosome and the autosomes. Among olive baboons, the regression slopes for the Ngorongoro, Lake Manyara, Serengeti, and Gombe populations are significantly higher on the X chromosome (p-values 6.29e-06, 2.53e-06, 1.42e-09, 0.00235). In Ngorongoro and Serengeti, the slopes are 5.53 and 7.13 times steeper on the X chromosome. As expected, from their location near the hybrid zone, the Tarangire and Arusha populations (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and Supplementary Table\u0026nbsp;12) do not show significant differences between regression slopes for chromosome X and the autosomes. The Ruaha yellow population (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) also shows a significantly larger slope on chromosome X, 7.4 times steeper than for autosomes (p-val 1.42e-14). (See Supplementary Table\u0026nbsp;12 for all slopes and p-values).\u003c/p\u003e\u003cp\u003eMany populations show higher MPA proportions on the X chromosome\u003c/p\u003e\u003cp\u003eIn all yellow baboon populations, the MPA proportion is significantly higher on chromosome X than on the autosomes ( Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, Mann-Whitney U-test p-values below 8.06e-05, see Supplementary Table\u0026nbsp;13), whereas the olive baboon populations show no significant difference (See Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). MPA proportions for the X chromosome vary between 25.0% in Tarangire and 4.2% in Serengeti (see Supplementary Table\u0026nbsp;1), and in the yellow baboon populations vary between 12.9% in Ruaha and 4.5% in Selous. See Supplementary Fig.\u0026nbsp;11 for the distribution of MPA proportions across individual chromosome haplotypes, stratified by chromosome type.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe higher MPA proportions on the X chromosome of many populations seem to contradict our observation of stronger negative linked selection on admixture on the X chromosome. On the X chromosome, the distribution of MPA proportions is skewed toward higher frequencies (See Supplementary Fig.\u0026nbsp;12), particularly in the yellow baboons, where this is evident as vertical patterns in the local ancestry painting (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). To investigate if high-frequency MPA is responsible for the stronger correlations on chromosome X, we repeated our analyses after masking 100kb windows with MPA proportions above 25% (See Supplementary Fig.\u0026nbsp;13). Individual X and autosome regressions remain significant, but the association on X is strengthened in yellow baboons and weakened in olive baboons (Supplementary Table\u0026nbsp;14): In Udzungwa and Mikumi yellow baboons, the regression slopes for chromosome X are now significantly higher than for autosomes (p-values 7.26e-15 and 2.35e-05), whereas Lake Manyara and Serengeti olive baboons no longer show higher slopes for chromosome X than for autosomes. This observation suggests that any negative selection on high-frequency MPA in yellow baboons is not represented by background diversity levels.\u003c/p\u003e\u003cp\u003eTo investigate if high-frequency MPA could be explained as recent adaptive introgressions, we used Relate to scan for recent selective sweeps. The sample sizes of the individual populations are not sufficient for this analysis. However, an analysis of the pooled Tanzanian olive baboon populations yielded significant evidence of positive selection across the autosome but none on the X chromosome. (See Supplementary Fig.\u0026nbsp;14, Supplementary Section Positive Selection Scan). These sweeps on the autosomes were significantly more common in areas with low MPA, and we therefore find no evidence that adaptive introgressions cause high MPA areas.\u003c/p\u003e\u003cp\u003eHigh-frequency MPA is most common on X chromosomes in the yellow populations that represent nuclear swamping of a resident northern lineage, and could represent haplotypes from the original displaced olive population. If these haplotypes were incompatible with gradually introgressing yellow haplotypes, strong negative selection would maintain these as the major parent ancestry.\u003c/p\u003e\u003cp\u003eHamadryas MPA in Ethiopian olive baboons reveals strong selection on hybrid ancestry\u003c/p\u003e\u003cp\u003eTo further explore this incompatibility scenario, we also analyzed the separate hamadryas MPA component of the Gog olive baboon population in Ethiopia. Olive and hamadryas baboons belong to the northern clade of baboon species and are thus more similar than olive and yellow baboons. Their divergence is comparable to that of anatomically modern humans and Neanderthals[21,64]. The mitochondria of the Gog olive baboon population are more similar to those of hamadryas baboons than to the mitochondria of other sampled olive populations, including the Tanzanian olive baboons, which also carry hamadryas-like mitochondria[22,24]. This discordance has been explained by nuclear swamping, where an invasion of primarily olive males displaced the autosomal but not mitochondrial ancestry of a resident hamadryas-like population[18]. We ran RFMix to infer hamadryas-like ancestry along chromosomes of the Gog individuals. We used Tanzanian olive and Hamadryas baboons as the two reference panels. Minor Parent Ancestry (MPA) is calculated as the proportion of Hamadryas ancestry in 100-kilobase windows after filtering areas with less than 75% callable bases (See Materials and Methods).\u003c/p\u003e\u003cp\u003eOn the autosomes, Gog olive baboons have the steepest regression slopes of all investigated baboon populations against both recombination (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA, slope 0.251, p-value 1.03e-76) and background diversity (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB, slope 38.8, p-value 1.58e-132) (Supplementary Table\u0026nbsp;15), as well as the largest absolute difference between the lowest 20% background diversity quintile and the highest (7.05% in the lowest 20\u0026ndash;12.6% in the highest 20%, a 78% increase). With a regression using both recombination rate and background diversity, it is only background diversity that has a significant association (p-values 0.393 and 1.41e-65).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWide genomic regions retaining pure Hamadryas ancestry suggest large-scale hybrid incompatibility.\u003c/p\u003e\u003cp\u003eDespite a large MPA proportion (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC and D), the X chromosome of Gog olive baboons also shows the steepest slope in the background diversity and MPA proportion regression of all investigated populations, with a slope of 214, and a p-value of 7.18e-10, significantly higher than that for autosomes (p-value 8.88e-06). After normalization, the X chromosome regression is 54% steeper than the autosomal one. See Supplementary Table\u0026nbsp;15 for all regressions. Gog olive baboons also show the largest absolute difference between the 20% low diversity quantile chromosome X (14.5\u0026ndash;42.1%, a 190% increase).\u003c/p\u003e\u003cp\u003eDespite showing the strongest evidence of negative selection on MPA, the Gog X chromosome also shows the largest proportion of MPA in our study: 33.1% compared to only 8.8% on autosomes. (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD and Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e). These contrasting observations are qualitatively similar to those for yellow and olive admixture, yet greatly amplified. The evidence of negative selection is strongest here despite very long admixture tracts, suggesting a selection regime different from autosomes. These observations are all consistent with nuclear swamping and point to strong selection retaining original ancestry on the X chromosome.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur comprehensive analysis of 156 high-coverage baboon genomes reveals that negative selection on admixture is pervasive across both olive and yellow baboon populations, extending far beyond hybrid zones and persisting over evolutionary timescales. The striking finding that linked selection against minor parent ancestry (MPA) is up to seven times stronger on the X chromosome than autosomes provides compelling evidence for the disproportionate role of sex chromosomes in maintaining species boundaries, consistent with Haldane's rule and theoretical predictions about hybrid incompatibility.\u003c/p\u003e\u003cp\u003ePerhaps the most intriguing finding is the apparent paradox that many populations show significantly higher MPA proportions on the X chromosome despite experiencing stronger negative selection on MPA. This pattern is particularly pronounced in yellow baboon populations and Ethiopian olive baboons, where X chromosome MPA can exceed 30% while showing the steepest correlations between MPA and proxies for linked selection. The absence of evidence for adaptive introgression on the X chromosome, despite high-frequency MPA in some regions, argues against positive selection driving these patterns. Instead, the retention of original ancestry at high frequency likely reflects strong negative selection against hybrid genotypes at these loci.\u003c/p\u003e\u003cp\u003eDuring nuclear swamping events[23,65], the invading ancestry transitions from minor to major status as it spreads through the population. Our observations suggest that strong selection on the invading minor parent ancestry blocks the turnover of ancestry in genomic regions spanning several megabases. This would explain why Ethiopian olive baboons show 33.1% hamadryas ancestry on the X chromosome compared to only 8.8% on autosomes, despite showing the strongest evidence for negative selection against MPA in our entire dataset. We propose that the wide genomic regions retaining pure hamadryas ancestry on the X chromosome represent incompatibility loci where hybrid genotypes face severe fitness costs. At incompatibility loci where such negative selection was insufficient to prevent the invading ancestry from reaching the majority, selection would instead purge the resident ancestry, which is now the minority, possibly explaining the pure olive ancestry extending across several megabases.\u003c/p\u003e\u003cp\u003eThe gradient of MPA proportions from hybrid zones into the species' core ranges provides a natural experiment for understanding how selection acts on admixture over time. Populations closest to contact zones (Tarangire, Arusha) show no correlation between MPA and recombination rate or background diversity, consistent with recent admixture where introgressed segments remain too large for fine-scale recombination to effectively unlink neutral from deleterious variants. This observation aligns with theoretical predictions and previous empirical work[38,54,66], suggesting that 4\u0026ndash;5 generations are required before recombination can effectively purge deleterious variants while preserving neutral admixed segments. The progressive strengthening of the correlation between MPA and linked selection proxies with distance from hybrid zones demonstrates that selection continues to shape admixture patterns long after initial hybridization events. The Gombe olive baboon population, located furthest from active admixture zones, shows the steepest regression slopes, indicating ongoing purging of deleterious admixed variants even in populations with relatively ancient admixture histories (estimated at 10\u0026ndash;100 generations by Globetrotter analysis).\u003c/p\u003e\u003cp\u003eOur finding that background diversity consistently outperforms recombination rate as a predictor of MPA proportions has important implications for understanding the mechanisms of selection against admixture. While recombination rate has been a commonly used proxy for linked selection efficiency, it fails to account for variation in the density of functional sites under selection. Background diversity, measured in non-admixed reference populations, integrates the effects of both recombination rate and functional constraint, providing a more comprehensive measure of linked selection intensity. The negative association between recombination rate and MPA when controlling for background diversity suggests that low-recombination, high-diversity regions may harbour incompatibility loci where admixture persists at intermediate frequencies. This could occur if these regions contain balanced polymorphisms or if the fitness costs of admixture are frequency-dependent. Alternatively, structural variants such as inversions, which suppress recombination while maintaining diversity, might play a role in maintaining admixture in these regions.\u003c/p\u003e\u003cp\u003eThe ubiquity of negative selection against admixture across all studied populations, including those hundreds of kilometres from hybrid zones, challenges the notion that gene flow readily homogenizes genomes between hybridizing species. Even in the face of ongoing gene flow, selection maintains species boundaries by continuously purging incompatible genetic combinations. The much stronger linked selection on the X chromosome MPA in some populations underscores the critical role of sex chromosomes in speciation, supporting theoretical predictions about the evolution of reproductive isolation. The reduced X/autosome diversity ratios (34.6\u0026ndash;52.1% of expected values) across all populations suggest that the X chromosome experiences stronger linked selection even within species. This could result from the exposure of recessive deleterious mutations in hemizygous males, the smaller effective population size of the X chromosome, the accumulation of sexually antagonistic alleles, or meiotic drive. The combination of stronger within-species selection and stronger negative selection on MPA makes the X chromosome a particularly effective barrier to gene flow between species.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003eA mask was created based on the depth of coverage per individual per site. For each individual, the most common read depth (mode) was calculated using bcftools (see Supplementary Fig.\u0026nbsp;15 for distribution, Supplementary Section Commands). Sites were filtered using bcftools, removing sites that were uncalled, had a heterozygous call and alternate allele depth less than 3, coverage less than a third of the mean coverage or more than twice, or a GQ score less than or equal to 30. A species-wide mask was created by calling a base as passing if 95% or more of the samples had a passing state with bedtools multiIntersectBed. For chromosome X, the mask was created using only the 98 females.\u003c/p\u003e\u003cp\u003eDiversity was calculated using scikit-allels' windowed_diversity function with a window size of 100kb. All windows in the analysis are 100kb, starting from base 1 of the reference genome unless stated otherwise. Windows were removed from the analysis if less than 75% of the window was set as passing in the filter step, resulting in 25397 100kb windows for the autosome and 1063 100kb windows for chromosome X. Analysis was performed on a per-population basis as identified in the metadata. See Supplementary Table\u0026nbsp;16 for a count of males and females.\u003c/p\u003e\u003cp\u003eThe historical effective population size for every population was inferred with SMC++. The autosome mutation rate was set at 0.57e-08 per base. With 4.5 times more mutations in males[72], the chromosome X was set at 0.45e-08. This mutation rate was calculated based on the higher mutation rate in males corresponding to (4.5/3\u0026thinsp;+\u0026thinsp;2/3)/(4.5/2\u0026thinsp;+\u0026thinsp;1/2), the mutation rate in chromosome X divided by the autosome mutation rate. Generation time was set at 11 years[55]. The Autosomal and chromosome X SMC\u0026thinsp;+\u0026thinsp;+\u0026thinsp;run was run with a piecewise spline and ten expectation maximization iterations. Chromosome X Ne was inferred only using females.\u003c/p\u003e\u003cp\u003eA fine-scale recombination map was inferred using Pyrho and the Mikumi Yellow population, as it had the largest sample size. The large number of samples from this population allows for a more accurate inference of the fine-scale recombination rate. However, the modest admixture from olives will slightly bias our map toward lower rates in regions with higher MPA proportions, as admixture increases LD in the affected region[67,68]. The recombination rate in 100kb windows has a mean and median of 0.0834 and 0.0668 centiMorgan (cM), scaled to a size equal to a previously published map for olive baboons of 2293 cM for the autosome[69].\u003c/p\u003e\u003cp\u003eThe Pyrho workflow consists of three steps: making a lookup table, inferring optimal hyperparameters, and then inferring the fine-scale recombination. Pyrho maketable was run with standard options except for the approximate setting for computational feasibility. The autosomal and chromosome X population size histories for Mikumi yellow baboons were used. Pyrho hyperparam was run with the following possible block penalties and window sizes: 10,25,50,100, and 10,25,50,100. Then, Pyrho optimize was run with optimal hyperparameters as measured by the L2 norm for each chromosome. Lastly, the autosome-wide recombination rate was scaled to be equal to the estimate of 2293cM published by [69], as Pyrho is optimized to infer fine-scale recombination and not total genetic distance. The scaling factor used was 1.3, as the total autosome-wide recombination rate inferred from pyrho was 1761. The assembly used for this estimate is Panubis1.0, while the one used in [18] is panu3. The recombination rate per window was calculated with the Pyrho genetic map, based on the genetic distance between the first and last base in the window, with interpolation performed assuming an average recombination rate between SNPs.\u003c/p\u003e\u003cp\u003eLocal Ancestry Inference was done using RFMix version 2, removing all alleles with a Minor Allele Frequency of less than 1% for computational efficiency. RFMix was run with the Pyrho genetic map and an assumed admixture date of 100 generations ago.\u003c/p\u003e\u003cp\u003eRelate was run with a mutation rate of 0.57e-8 and an effective initial population size of 50000 for the autosomes, and chromosome X with a mutational rate of 0.46e-8 and an effective population size of 25000, in both cases with a generation time of 11 generations. The callability mask is the same as previously detailed. Otherwise, standard options were used for the Relate workflow, with the convert from VCF and prepare input script used to transform the VCF into haps/sample format, followed by Relate, EstimatePopulationSize, and then DetectSelection.\u003c/p\u003e\u003cp\u003eWe used weighted regressions to account for heteroscedasticity, as identified with the Breusch-Pagan test[70]. We used the statsmodels python package to run the regression models. The significance threshold is set at 0.005 to correct for multiple testing (Bonferroni correction) for the ten olive and yellow populations in Tanzania.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003cp\u003eErik Fogh S\u0026oslash;rensen and Kasper Munch have published an article together with Jeffrey Rogers (https://doi.org/10.1126/science.abn8153)\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis work was funded by Novo Nordisk Foundation grant 0058553 (E.F.S. and K.M.).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eConceptualization: All authors.Data curation: E.F.S.Investigation: E.F.S. and K.M.Formal analysis: E.F.S. and K.M.Supervision: G.H. and K.M.Visualization: E.F.S. (All Figures) and K.M. (Figure 1).Writing \u0026ndash; original draft: E.F.S. and K.M.Writing \u0026ndash; review \u0026amp; editing: All authors.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe thank Mikkel Heide Schierup and Dietmar Zinner for feedback on the manuscript, and Jeffrey Rogers, Christian Roos and Clifford J. Jolly for insightful discussion.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eData is sourced from accession code PRJEB59576 [18,71], which is publicly available in the European Nucleotide Archive.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMarques DA, Meier JI, Seehausen O. A Combinatorial View on Speciation and Adaptive Radiation. Trends Ecol Evol. 2019;34:531\u0026ndash;44.\u003c/li\u003e\n\u003cli\u003eWorsham MLD, Julius EP, Nice CC, Diaz PH, Huffman DG. Geographic isolation facilitates the evolution of reproductive isolation and morphological divergence. Ecol Evol. 2017;7:10278\u0026ndash;88.\u003c/li\u003e\n\u003cli\u003eSchumer M, Xu C, Powell DL, Durvasula A, Skov L, Holland C, et al. Natural selection interacts with recombination to shape the evolution of hybrid genomes. Science. 2018;360:656\u0026ndash;60.\u003c/li\u003e\n\u003cli\u003eAbbott R, Albach D, Ansell S, Arntzen JW, Baird SJE, Bierne N, et al. Hybridization and speciation*. J Evol Biol. 2013;26:229\u0026ndash;46.\u003c/li\u003e\n\u003cli\u003eGompert Z, Parchman TL, Buerkle CA. Genomics of isolation in hybrids. Philosophical Transactions Royal Soc B Biological Sci. 2012;367:439\u0026ndash;50.\u003c/li\u003e\n\u003cli\u003ePoikela N, Laetsch DR, Kankare M, Hoikkala A, Lohse K. Experimental introgression in Drosophila: Asymmetric postzygotic isolation associated with chromosomal inversions and an incompatibility locus on the X chromosome. Mol Ecol. 2023;32:854\u0026ndash;66.\u003c/li\u003e\n\u003cli\u003eCutter AD. The polymorphic prelude to Bateson\u0026ndash;Dobzhansky\u0026ndash;Muller incompatibilities. Trends Ecol Evol. 2012;27:209\u0026ndash;18.\u003c/li\u003e\n\u003cli\u003eLangmore NE, Grealy A, Noh H-J, Medina I, Skeels A, Grant J, et al. Coevolution with hosts underpins speciation in brood-parasitic cuckoos. Science. 2024;384:1030\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eOrr HA. Dobzhansky, Bateson, and the Genetics of Speciation. Genetics. 1996;144:1331\u0026ndash;5.\u003c/li\u003e\n\u003cli\u003ePresgraves DC. Evaluating genomic signatures of \u0026ldquo;the large X‐effect\u0026rdquo; during complex speciation. Mol Ecol. 2018;27:3822\u0026ndash;30.\u003c/li\u003e\n\u003cli\u003eHarris K, Nielsen R. The Genetic Cost of Neanderthal Introgression. Genetics. 2016;203:881\u0026ndash;91.\u003c/li\u003e\n\u003cli\u003eZeller E, Timmermann A. The evolving three-dimensional landscape of human adaptation. Sci Adv. 2024;10:eadq3613.\u003c/li\u003e\n\u003cli\u003eVamosi SM, Schluter D. SEXUAL SELECTION AGAINST HYBRIDS BETWEEN SYMPATRIC STICKLEBACK SPECIES: EVIDENCE FROM A FIELD EXPERIMENT. Evolution. 1999;53:874\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eS\u0026aacute;nchez-Ram\u0026iacute;rez S, Weiss JG, Thomas CG, Cutter AD. Sex-specific and sex-chromosome regulatory evolution underlie widespread misregulation of inter-species hybrid transcriptomes. Biorxiv. 2020;2020.05.04.076505.\u003c/li\u003e\n\u003cli\u003eLarson EL, Vanderpool D, Sarver BAJ, Callahan C, Keeble S, Provencio LL, et al. The Evolution of Polymorphic Hybrid Incompatibilities in House Mice. Genetics. 2018;209:845\u0026ndash;59.\u003c/li\u003e\n\u003cli\u003eLarson EL, Keeble S, Vanderpool D, Dean MD, Good JM. The composite regulatory basis of the large X-effect in mouse speciation. Mol Biol Evol. 2016;msw243.\u003c/li\u003e\n\u003cli\u003eSchield DR, Scordato ESC, Smith CCR, Carter JK, Cherkaoui SI, Gombobaatar S, et al. Sex‐linked genetic diversity and differentiation in a globally distributed avian species complex. Mol Ecol. 2021;30:2313\u0026ndash;32.\u003c/li\u003e\n\u003cli\u003eS\u0026oslash;rensen EF, Harris RA, Zhang L, Raveendran M, Kuderna LFK, Walker JA, et al. Genome-wide coancestry reveals details of ancient and recent male-driven reticulation in baboons. Science. 2023;380:eabn8153.\u003c/li\u003e\n\u003cli\u003eChiou KL, Bergey CM, Burrell AS, Disotell TR, Rogers J, Jolly CJ, et al. Genome‐wide ancestry and introgression in a Zambian baboon hybrid zone. Mol Ecol. 2021;\u003c/li\u003e\n\u003cli\u003eVilgalys TP, Fogel AS, Anderson JA, Mututua RS, Warutere JK, Siodi IL, et al. Selection against admixture and gene regulatory divergence in a long-term primate field study. Science. 2022;377:635\u0026ndash;41.\u003c/li\u003e\n\u003cli\u003eRogers J, Raveendran M, Harris RA, Mailund T, Lepp\u0026auml;l\u0026auml; K, Athanasiadis G, et al. The comparative genomics and complex population history of Papio baboons. Sci Adv. 2019;5:eaau6947.\u003c/li\u003e\n\u003cli\u003eKopp GH, Sithaldeen R, Trede F, Grathwol F, Roos C, Zinner D. A Comprehensive Overview of Baboon Phylogenetic History. Genes. 2023;14:614.\u003c/li\u003e\n\u003cli\u003eJolly CJ. Philopatry at the frontier: A demographically driven scenario for the evolution of multilevel societies in baboons (Papio). J Hum Evol. 2020;146:102819.\u003c/li\u003e\n\u003cli\u003eZinner D, Wertheimer J, Liedigk R, Groeneveld LF, Roos C. Baboon phylogeny as inferred from complete mitochondrial genomes. Am J Phys Anthropol. 2013;150:133\u0026ndash;40.\u003c/li\u003e\n\u003cli\u003eSkov L, Maci\u0026agrave; MC, Lucotte E, Cavassim MIA, Castellano D, Schierup MH, et al. Extraordinary selection on the human X chromosome associated with archaic admixture. Biorxiv. 2022;2022.09.19.508556.\u003c/li\u003e\n\u003cli\u003eLucotte EA, Skov L, Jensen JM, Maci\u0026agrave; MC, Munch K, Schierup MH. Dynamic Copy Number Evolution of X- and Y-Linked Ampliconic Genes in Human Populations. Genetics. 2018;209:907\u0026ndash;20.\u003c/li\u003e\n\u003cli\u003eVicoso B, Charlesworth B. Effective Population Size and the Faster-X Effect: An Extended Model. Evolution. 2009;63:2413\u0026ndash;26.\u003c/li\u003e\n\u003cli\u003eMeisel RP, Connallon T. The faster-X effect: integrating theory and data. Trends Genet. 2013;29:537\u0026ndash;44.\u003c/li\u003e\n\u003cli\u003ePool JE, Nielsen R. Inference of Historical Changes in Migration Rate From the Lengths of Migrant Tracts. Genetics. 2009;181:711\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eKeinan A, Reich D. Can a Sex-Biased Human Demography Account for the Reduced Effective Population Size of Chromosome X in Non-Africans? Mol Biol Evol. 2010;27:2312\u0026ndash;21.\u003c/li\u003e\n\u003cli\u003eLarson EL, Keeble S, Vanderpool D, Dean MD, Good JM. The composite regulatory basis of the large X-effect in mouse speciation. Molecular biology and evolution. 2016;msw243.\u003c/li\u003e\n\u003cli\u003eRathje C, Johnson E, Drage D, Patinioti C, Silvestri G, Affara N, et al. Differential Sperm Motility Mediates the Sex Ratio Drive Shaping Mouse Sex Chromosome Evolution. Curr Biol. 2019;29:3692-3698.e4.\u003c/li\u003e\n\u003cli\u003eHughes JF, Skaletsky H, Pyntikova T, Koutseva N, Raudsepp T, Brown LG, et al. Sequence analysis in Bos taurus reveals pervasiveness of X\u0026ndash;Y arms races in mammalian lineages. Genome Res. 2020;\u003c/li\u003e\n\u003cli\u003eSkov L, Maci\u0026agrave; MC, Lucotte EA, Cavassim MIA, Castellano D, Schierup MH, et al. Extraordinary selection on the human X chromosome associated with archaic admixture. Cell Genom. 2023;100274.\u003c/li\u003e\n\u003cli\u003eDutheil JY, Munch K, Nam K, Mailund T, Schierup MH. Strong Selective Sweeps on the X Chromosome in the Human-Chimpanzee Ancestor Explain Its Low Divergence. PLOS Genetics. 2015;11:e1005451.\u003c/li\u003e\n\u003cli\u003eNam K, Munch K, Hobolth A, Dutheil J, Veeramah KR, Woerner AE, et al. Extreme selective sweeps independently targeted the X chromosomes of the great apes. Proceedings of the National Academy of Sciences. 2015;112:6413\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eNam K, Munch K, Mailund T, Nater A, Greminger M, Kr\u0026uuml;tzen M, et al. Evidence that the rate of strong selective sweeps increases with population size in the great apes. Proceedings of the National Academy of Sciences. 2017;114:201605660.\u003c/li\u003e\n\u003cli\u003eVeller C, Edelman NB, Muralidhar P, Nowak MA. Recombination and selection against introgressed DNA. Evolution. 2023;77:1131\u0026ndash;44.\u003c/li\u003e\n\u003cli\u003eMatheson J, Masel J. Background Selection From Unlinked Sites Causes Nonindependent Evolution of Deleterious Mutations. Genome Biol Evol. 2024;16:evae050.\u003c/li\u003e\n\u003cli\u003eLiu KJ, Steinberg E, Yozzo A, Song Y, Kohn MH, Nakhleh L. Interspecific introgressive origin of genomic diversity in the house mouse. Proc Natl Acad Sci. 2015;112:196\u0026ndash;201.\u003c/li\u003e\n\u003cli\u003eVilgalys TP, Fogel AS, Mututua RS, Warutere JK, Siodi L, Anderson JA, et al. Selection against admixture and gene regulatory divergence in a long-term primate field study. Biorxiv. 2021;2021.08.19.456711.\u003c/li\u003e\n\u003cli\u003eHellenthal G, Busby GBJ, Band G, Wilson JF, Capelli C, Falush D, et al. A Genetic Atlas of Human Admixture History. Science. 2014;343:747\u0026ndash;51.\u003c/li\u003e\n\u003cli\u003eWangkumhang P, Greenfield M, Hellenthal G. An efficient method to identify, date and describe admixture events using haplotype information. Biorxiv. 2021;2021.08.12.455263.\u003c/li\u003e\n\u003cli\u003eMaples BK, Gravel S, Kenny EE, Bustamante CD. RFMix: A Discriminative Modeling Approach for Rapid and Robust Local-Ancestry Inference. Am J Hum Genetics. 2013;93:278\u0026ndash;88.\u003c/li\u003e\n\u003cli\u003eMassarat AR, Lamkin M, Reeve C, Williams AL, D\u0026rsquo;Antonio M, Gymrek M. Haptools: a toolkit for admixture and haplotype analysis. Bioinformatics. 2023;39:btad104.\u003c/li\u003e\n\u003cli\u003eSpence JP, Song YS. Inference and analysis of population-specific fine-scale recombination maps across 26 diverse human populations. Sci Adv. 2019;5:eaaw9206.\u003c/li\u003e\n\u003cli\u003eKamm JA, Spence JP, Chan J, Song YS. Two-Locus Likelihoods Under Variable Population Size and Fine-Scale Recombination Rate Estimation. Genetics. 2016;203:1381\u0026ndash;99.\u003c/li\u003e\n\u003cli\u003eRonquist F, Kudlicka J, Senderov V, Borgstr\u0026ouml;m J, Lartillot N, Lund\u0026eacute;n D, et al. Universal probabilistic programming offers a powerful approach to statistical phylogenetics. Commun Biology. 2021;4:244.\u003c/li\u003e\n\u003cli\u003eTerhorst J, Kamm JA, Song YS. Robust and scalable inference of population history from hundreds of unphased whole genomes. Nat Genet. 2017;49:303\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003ePool JE, Nielsen R. POPULATION SIZE CHANGES RESHAPE GENOMIC PATTERNS OF DIVERSITY. Evolution. 2007;61:3001\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eCharlesworth B, Jensen JD. Effects of Selection at Linked Sites on Patterns of Genetic Variability. Annu Rev Ecol, Evol, Syst. 2021;52:177\u0026ndash;97.\u003c/li\u003e\n\u003cli\u003eEllegren H, Galtier N. Determinants of genetic diversity. Nat Rev Genet. 2016;17:422\u0026ndash;33.\u003c/li\u003e\n\u003cli\u003eBurri R, Nater A, Kawakami T, Mugal CF, Olason PI, Smeds L, et al. Linked selection and recombination rate variation drive the evolution of the genomic landscape of differentiation across the speciation continuum of Ficedula flycatchers. Genome Res. 2015;25:1656\u0026ndash;65.\u003c/li\u003e\n\u003cli\u003eNordborg M, Charlesworth B, Charlesworth D. The effect of recombination on background selection*. Genet Res. 1996;67:159\u0026ndash;74.\u003c/li\u003e\n\u003cli\u003eWu FL, Strand AI, Cox LA, Ober C, Wall JD, Moorjani P, et al. A comparison of humans and baboons suggests germline mutation rates do not track cell divisions. Plos Biol. 2020;18:e3000838.\u003c/li\u003e\n\u003cli\u003eBergeron LA, Besenbacher S, Zheng J, Li P, Bertelsen MF, Quintard B, et al. Evolution of the germline mutation rate across vertebrates. Nature. 2023;615:285\u0026ndash;91.\u003c/li\u003e\n\u003cli\u003eBergeron LA, Besenbacher S, Bakker J, Zheng J, Li P, Pacheco G, et al. The germline mutational process in rhesus macaque and its implications for phylogenetic dating. GigaScience. 2021;10:giab029.\u003c/li\u003e\n\u003cli\u003ePool JE, Nielsen R. POPULATION SIZE CHANGES RESHAPE GENOMIC PATTERNS OF DIVERSITY. Evolution. 2007;61:3001\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eAlberts SC, Watts HE, Altmann J. Queuing and queue-jumping: long-term patterns of reproductive skew in male savannah baboons, Papio cynocephalus. Anim Behav. 2003;65:821\u0026ndash;40.\u003c/li\u003e\n\u003cli\u003eAlberts SC, Buchan JC, Altmann J. Sexual selection in wild baboons: from mating opportunities to paternity success. Anim Behav. 2006;72:1177\u0026ndash;96.\u003c/li\u003e\n\u003cli\u003eAlberts SC. Social influences on survival and reproduction: Insights from a long‐term study of wild baboons. J Anim Ecol. 2019;88:47\u0026ndash;66.\u003c/li\u003e\n\u003cli\u003eSt\u0026auml;dele V, Roberts ER, Barrett BJ, Strum SC, Vigilant L, Silk JB. Male\u0026ndash;female relationships in olive baboons (Papio anubis): Parenting or mating effort? J Hum Evol. 2019;127:81\u0026ndash;92.\u003c/li\u003e\n\u003cli\u003eKopp GH, Silva MJF da, Fischer J, Brito JC, Regnaut S, Roos C, et al. The Influence of Social Systems on Patterns of Mitochondrial DNA Variation in Baboons. Int J Primatol. 2014;35:210\u0026ndash;25.\u003c/li\u003e\n\u003cli\u003eHubisz MJ, Williams AL, Siepel A. Mapping gene flow between ancient hominins through demography-aware inference of the ancestral recombination graph. Plos Genet. 2020;16:e1008895.\u003c/li\u003e\n\u003cli\u003eJiang J, Yu J, Li J, Li P, Fan Z, Niu L, et al. Mitochondrial Genome and Nuclear Markers Provide New Insight into the Evolutionary History of Macaques. PLoS ONE. 2016;11:e0154665.\u003c/li\u003e\n\u003cli\u003eMcVicker G, Gordon D, Davis C, Green P. Widespread Genomic Signatures of Natural Selection in Hominid Evolution. Plos Genet. 2009;5:e1000471.\u003c/li\u003e\n\u003cli\u003eLoh P-R, Lipson M, Patterson N, Moorjani P, Pickrell JK, Reich D, et al. Inferring Admixture Histories of Human Populations Using Linkage Disequilibrium. Genetics. 2013;193:1233\u0026ndash;54.\u003c/li\u003e\n\u003cli\u003eBaran Y, Pasaniuc B, Sankararaman S, Torgerson DG, Gignoux C, Eng C, et al. Fast and accurate inference of local ancestry in Latino populations. Bioinformatics. 2012;28:1359\u0026ndash;67.\u003c/li\u003e\n\u003cli\u003eWall JD, Robinson JA, Cox LA. High-Resolution Estimates of Crossover and Noncrossover Recombination from a Captive Baboon Colony. Genome Biol Evol. 2022;14:evac040.\u003c/li\u003e\n\u003cli\u003eBreusch TS, Pagan AR. A Simple Test for Heteroscedasticity and Random Coefficient Variation. Econometrica. 1979;47:1287.\u003c/li\u003e\n\u003cli\u003eKuderna LFK, Gao H, Janiak MC, Kuhlwilm M, Orkin JD, Bataillon T, et al. A global catalog of whole-genome diversity from 233 primate species. Science. 2023;380:906\u0026ndash;13.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"genome-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gbio","sideBox":"Learn more about [Genome Biology](https://genomebiology.biomedcentral.com/)","snPcode":"13059","submissionUrl":"https://submission.springernature.com/new-submission/13059/3","title":"Genome Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7456420/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7456420/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eThe six admixing baboon species offer a natural experiment to study negative selection on admixture and the nature of genetic incompatibility. In Tanzania, a secondary contact between olive and yellow baboons allows admixture despite 1.3\u0026nbsp;million years of divergence. An independent secondary contact occurred in Ethiopia, upon which olive baboons invaded and displaced an ancient Hamadryas-like population separated by 0.6\u0026nbsp;million years, mirroring the displacement of Neanderthals by modern humans.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eWe analyze 156 high-coverage genomes sampled from seven olive and four yellow baboon populations in East Africa. Analyzing local ancestry across the whole genome, we find evidence of negative selection on minor parent ancestry in both Tanzanian yellow and olive baboon populations that reaches far beyond the hybrid zone. Across populations, we find that selection on minor parent ancestry is stronger on the X chromosome than on autosomes, most extremely in one yellow baboon population, which shows a seven-fold difference. The proportion of minor parent ancestry (MPA) is substantially higher on the X chromosome in Ethiopian olive and yellow baboon populations, which both displaced the populations now representing their minor parent ancestry, owing mainly to a few genomic regions with MPA at very high frequencies. We hypothesize that strong negative selection on MPA allowed these X chromosome regions to retain the original ancestry, as this was slowly displaced across the remaining genome.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eOur findings provide deeper insights into admixture dynamics in primates, highlighting the persistence of selection against admixture across various levels of admixture, and underscoring the need to include chromosome X in admixture analyses.\u003c/p\u003e","manuscriptTitle":"Negative selection on baboon admixture is strongest on chromosome X","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-15 05:47:41","doi":"10.21203/rs.3.rs-7456420/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-01T15:32:23+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-28T20:50:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"3955099379895890349213493539708481636","date":"2026-04-10T14:45:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"226231699563742270891420474316526682304","date":"2025-12-14T17:14:58+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-13T09:16:38+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"249446190916473148973385219942306110892","date":"2025-10-06T15:34:43+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-01T15:25:06+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-26T11:45:46+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-26T07:49:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"Genome Biology","date":"2025-08-25T18:43:10+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"genome-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gbio","sideBox":"Learn more about [Genome Biology](https://genomebiology.biomedcentral.com/)","snPcode":"13059","submissionUrl":"https://submission.springernature.com/new-submission/13059/3","title":"Genome Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"39e1f5fb-1b8d-466e-b5d7-b92d3603ee45","owner":[],"postedDate":"October 15th, 2025","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-01T15:32:23+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-01T15:39:42+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-15 05:47:41","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7456420","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7456420","identity":"rs-7456420","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}