Hybridization breaks species barriers in long-term coevolution of a cyanobacterial population Authors: Gabriel Birzu, Harihara Subrahmaniam Muralidharan, Danielle Goudeau, Rex R Malmstrom, Daniel S Fisher, Devaki Bhaya, Detlef Weigel, Gabriel Birzu, Harihara Subrahmaniam Muralidharan, Danielle Goudeau, Rex R Malmstrom, Daniel S Fisher, Devaki Bhaya Published: 2025 ABSTRACT Bacterial species often undergo rampant recombination yet maintain cohesive genomic identity. Ecological differences can generate recombination barriers between species and sustain genomic clusters in the short term. But can these forces prevent genomic mixing during long-term coevolution? Cyanobacteria in Yellowstone hot springs comprise several diverse species that have coevolved for hundreds of thousands of years, providing a rare natural experiment. By analyzing more than 300 single-cell genomes, we show that despite each species forming a distinct genomic cluster, much of the diversity within species is the result of hybridization driven by selection, which has mixed their ancestral genotypes. This widespread mixing is contrary to the prevailing view that ecological barriers can maintain cohesive bacterial species and highlights the importance of hybridization as a source of genomic diversity. Introduction The origin, persistence, and extinction of species in obligate sexual populations is a fundamental problem in evolutionary theory (Coyne and Orr, 2004). But in bacteria, long thought to be primarily asexual, our understanding is still very limited (Fraser et al., 2009; Rocha, 2018). While it has long been known that bacteria can acquire non-essential genes from distantly related strains (Rivera et al., 1998), in recent years, studies across a wide range of bacteria have found that homologous recombination of core genes is also ubiquitous (Didelot and Maiden, 2010; Yahara et al., 2016; Rosen et al., 2015; Garud et al., 2019; Sakoparnig et al., 2021). How does the interplay between selection acting both asexually and on transferred elements, and barriers to recombination from geographical separation, mechanistic effects, and epistatic interactions lead to formation and maintenance of cohesive species? One explanation is that ecological specialization provides barriers to recombination (Kashtan et al., 2014; Arevalo et al., 2019). This view is supported by evidence that selective sweeps of individual genes through parts of a population can lead to distinct ecological subtypes, or ecotypes, even in highly recombining bacterial populations (Croucher et al., 2011; Shapiro et al., 2012; Bao et al., 2016). But it also raises an important open question: if species coexist and recombine with each other, do genetic differences continue to increase over evolutionary time, or does recombination gradually break down barriers between species? Answering this question is challenging because data on the underlying evolutionary dynamics is lacking. The time scales of direct observations and laboratory experiments are too short, while those corresponding to phylogenetic relationships between extant species are too long. And since spatial dynamics over these intermediate time scales are unknown in most bacteria (Hanson et al., 2012; Garud et al., 2019), it is difficult to control for changes in local community structure during evolution. Therefore, microbial communities that coevolved for long time periods can act as natural experiments for studying bacterial speciation. The cyanobacterial populations found in hot springs from Yellowstone National Park form a natural incubator in which one such natural experiment is ongoing. These communities have two key features that allow us to identify and quantify the effects of coevolution on intermediate time scales. First, like other thermophilic communities, these populations are geographically isolated (Papke et al., 2003; Whitaker et al., 2003). To date, the two most abundant species, Synechococcus A and B, have not been found outside of North America, suggesting that their global mixing time is much longer than for most other bacteria (Papke et al., 2003). Comparison of two isolate genomes from these species, OS-A and OS-B’, suggest a divergence time considerably longer than the formation of the Yellowstone caldera around 640,000 years ago (Rosen et al., 2018). This would imply their ancestors colonized the caldera independently. Second, these populations have considerable sub-species diversity, suggesting ongoing evolutionary processes generate and maintain it. Analysis of variants of 16S rRNA and marker genes was initially interpreted as evidence of asexual ecotypes occupying distinct niches (Ward et al., 2006; Ward et al., 1998). However, deep amplicon data from a diverse, but largely OS-B’-like sample revealed extensive recombination (Rosen et al., 2015). Genetic exchange with other genomic clusters—which we refer to as hybridization—was also evident, particularly with relatives of OS-A. To date, evidence of hybridization in bacteria has mainly been from comparisons between reference genomes from large genomic databases (Smillie et al., 2011; Diop et al., 2022), with only a handful of studies investigating hybridization in natural environments (Tyson et al., 2004; Lo et al., 2007; Sheppard et al., 2008). The primary focus of the current paper is to determine the impact of hybridization on the long-term coevolution of a natural microbial community. Characterizing hybridization requires long-distance linkage between variants that cannot be obtained from traditional metagenomics. Strain isolates are often not representative of the population diversity (Browne et al., 2016; Zheng et al., 2022), while metagenome assembly destroys linkage. Recruiting metagenomic reads to reference genomes can, in some cases, be used to infer linkage (Garud et al., 2019), but the high diversity of Synechococcus prevents us from using such approaches. To circumvent these problems, we used single-cell genomics (Rinke et al., 2013; Zheng et al., 2022) to obtain over 300 genomes from the Mushroom and Octopus Springs, which contain two of the most well-studied communities from the Yellowstone caldera (Ward et al., 2012). This is to our knowledge the first large sample of genomes from hot spring cyanobacterial populations without the compositional bias due to strain isolation. Results Population comprises three distinct highly recombining genomic clusters Previous analysis of 16S amplicons identified two main clusters similar to OS-A and OS-B’, but also several others at frequencies below 1% (Rosen et al., 2018). But whether clusters within 16S amplicons correspond to clusters at the whole-genome level in such a rapidly recombining population is not clear (Rosen et al., 2015; Miller and Carvey, 2019). Therefore, we first compared each single-amplified genome (SAG) to Synechococcus OS-A and OS-B’. Almost all SAGs (330 out of 331) comprised two clusters with an average nucleotide divergence of ∼2% from one of the reference genomes and ∼15% from the other. Note that the divergence between clusters is well above the 5% threshold that is commonly used to distinguish different species (Nayfach et al., 2016; Jain et al., 2018; Garud et al., 2019). We labeled the cluster closer to OS-A by α (128 SAGs) and the other by β (202 SAGs), to distinguish them from the reference genomes. Surprisingly, one SAG (approximately 70% complete) was ∼15% diverged from both OS-A and OS-B’ and formed a separate cluster which we labeled γ (Figure 1A). Clustering methods based on different metrics gave identical results (Appendix 2—figures 2 and 3). Synechococcus population comprises three genomic clusters. (A) The mean divergence across homologous genes from the two reference genomes is shown for every genome in the data. Each dot represents a single genome and is colored according to its assigned species. (B) The relative abundances of all three species across samples from different temperatures within Mushroom Spring. Abundances were estimated based on read recruitment from the metagenome to core genes from the γ\begin{document}$ \gamma$\end{document} genome, as well as representative genomes from the α and β populations. Dashed lines show the minimum detection frequencies from the single-cell (black) and metagenome (gray) data, respectively (see Appendix 3). Shaded gray region shows temperature range of the eight single-cell samples. To investigate the extent that these clusters characterize the distribution of genomes throughout the two springs, we used single-cell genomes as references and recruited reads from a large collection of metagenome samples to determine the abundance of α, β, and, particularly, γ across different environments (see Appendix 3). We found that 95% of reads were part of a main cloud of sequences with ≤5% nucleotide divergence from one of the three main clusters. We used the fraction of reads within each main cloud to infer the abundances of each cluster in the metagenomic samples. The abundance of γ was highly correlated with α (Appendix 3—figures 3 and 4) and increased rapidly with temperature in the Mushroom Spring samples (Figure 1B). One interpretation of this pattern is that α, β, and γ represent distinct species adapted to different temperatures. Alternatively, genotypes within the same cluster could be adapted to different temperatures or other environmental conditions. The relative abundances of the clusters would then reflect the complex eco-evolutionary processes that maintain the diversity in the population. Previous analysis using a much more limited dataset proposed that β formed a quasisexual population, in which rapid recombination led to random assortment of alleles within the cluster (Rosen et al., 2015). Motivated by these results, we investigated the impact of recombination on the diversity within the two main clusters. We quantified recombination rates in α and β using a standard measure of linkage disequilibrium, σd2\begin{document}$ \sigma_{d}^{2}$\end{document}. Given a pair of sites with alleles a/A\begin{document}$ \mathtt{a/A}$\end{document} and b/B\begin{document}$ \mathtt{b/B}$\end{document}, σd2\begin{document}$ \sigma_{d}^{2}$\end{document} is defined as(1)σd2=⟨(fab−fafb⟩)⟨fa(1−fa)fb(1−fb)⟩,\begin{document}$$\displaystyle \sigma_{d}^{2}=\frac{\left\langle(f_{\mathtt{ab}}-f_{\mathtt{a}}f_{\mathtt{b}}\right\rangle)}{\left\langle f_{\mathtt{a}}(1-f_{\mathtt{a}})f_{\mathtt{b}}(1-f_{\mathtt{b}})\right\rangle},$$\end{document} where ⟨⋅⟩\begin{document}$ \left\langle\cdot\right\rangle$\end{document} is the average across pairs of sites. In asexual populations with conventional demographic drift, σd2=5/11\begin{document}$ \sigma_{d}^{2}=5/11$\end{document} independently of the location of the sites. Recombination events affecting one but not both sites can unlink mutations, leading to a decrease in σd2\begin{document}$ \sigma_{d}^{2}$\end{document}. Because sites further apart have more opportunities to become unlinked, σd2\begin{document}$ \sigma_{d}^{2}$\end{document} decreases with the distance between sites. In β, the linkage disequilibrium σd2\begin{document}$ \sigma_{d}^{2}$\end{document} decreased by around a factor of 10 over distances of ∼1 kbp (Figure 2A). This agreed quantitatively with earlier results based on deep amplicon sequencing from Rosen et al., 2015, which showed a decrease by a factor of ∼1 over distances of up to 300 bp, using a different but related measure of linkage, conventionally denoted as ⟨r2⟩\begin{document}$ \left\langle r^{2}\right\rangle$\end{document}. Over genome-wide length scales, linkage was small (σd2=0.06\begin{document}$ \sigma_{d}^{2}=0.06$\end{document}) and close to a fully unlinked control (σd2=0.04\begin{document}$ \sigma_{d}^{2}=0.04$\end{document}). We found similar results for α\begin{document}$ \alpha$\end{document}, but with a quantitatively slower decrease in linkage by a factor of ∼8 over ∼1 kbp. Nevertheless, genome-wide linkage values were still close to the fully unlinked control (σd2=0.14\begin{document}$ \sigma_{d}^{2}=0.14$\end{document} for the data vs σd2=0.12\begin{document}$ \sigma_{d}^{2}=0.12$\end{document} for the control). As a consequence of rapid recombination, we found that diversity in marker genes, such as 16S rRNA, was poor predictors of whole-genome diversity (Figure 2B). Further details and analyses can be found in Appendices 4 and 8. Appendix 8 also provides estimates of the recombination parameters for each species and compares these to estimates obtained from previously proposed methods (Didelot and Wilson, 2015; Lin and Kussell, 2019). Altogether, these results are consistent with each cluster representing a quasisexual population and imply that phylogenies inferred from whole genomes do not reflect the evolutionary history of the population. Genetic linkage analysis shows extensive recombination within α and β. (A) Linkage disequilibrium σd2\begin{document}$ \sigma_{d}^{2}$\end{document} as a function of separation between SNPs x\begin{document}$ x$\end{document} averaged across all core genes from α (orange circles) and β (blue squares). Genome-wide estimates from SNPs in different genes are shown to the right of the black dashed line using the same symbols. Fully unlinked controls for the genome-wide linkage are shown by crosses of corresponding colors. Dashed lines show fit to linkage decay curves in an infinite sample using Equation 6 for α (orange) and β (blue) separately. (B) Scatter plot of 16S and genome-wide divergences between all pairs of genomes within the same species. Solid lines show linear regressions indicating very weak Pearson correlation coefficient, R\begin{document}$ R$\end{document}, within each species. This demonstrates that 16S divergence is a poor proxy of whole-genome divergence. While linkage decay in α and β was similar, we found diversity patterns at the gene level were qualitatively different (Figure 3A). The typical diversity at synonymous sites in β was πS≈0.04\begin{document}$ \pi_{S}\approx 0.04$\end{document}, with ∼2× variation across genes (Figure 3A), consistent with previous results (Rosen et al., 2018). In contrast, πS\begin{document}$ \pi_{S}$\end{document} in α varied wildly by over two orders of magnitude (Figure 3A), ranging from πS≈0.2\begin{document}$ \pi_{S}\approx 0.2$\end{document} in some genes to others that did not have a single mutation. Note that because here we only used synonymous sites, these variations cannot be explained by differences in the strength of purifying selection or the accumulation of non-synonymous mutations during rapid adaptation (Neher, 2022). Importantly, the low-diversity segments of the α genome spanned multiple samples and springs and are thus unlikely to be the result of a very recent clonal expansion as is often seen in human commensal and pathogenic bacteria (Didelot and Wilson, 2015; Garud et al., 2019; Sakoparnig et al., 2021; Weimann et al., 2024). α and β have distinct patterns of within-species diversity. (A) The synonymous diversity πS\begin{document}$ \pi_{S}$\end{document} within α (orange) and β (blue) at all core genes ordered according to their respective reference genomes. The solid line shows the value of πS\begin{document}$ \pi_{S}$\end{document}, with lighter colors showing one standard deviation across different pairs around the mean. Both lines were smoothed by averaging over a sliding window of five genes with a step size of two genes. (B) Distribution of the fraction of polymorphic fourfold degenerate (4D) sites for each α core gene (blue histogram) is shown together with Poisson fit based on the number of genes without any polymorphic sites (black line and dots). The inset shows zoomed-in region of low-diversity genes (<5% polymorphic 4D sites) together with the same fit. (C) The same as (B) using data from β. The Poisson distribution is fitted using the mean of the distribution, which provides a better fit than the number of genes without polymorphic sites (data not shown). (D) Histogram of synonymous diversities πS\begin{document}$ \pi_{S}$\end{document} within genes for β (blue), α low-diversity backbone (orange), and α high-diversity (hybrid) genes (red). Note the overlap between the diversity in the β core genome and the hybrid α genes. Beyond the low-diversity genes in α\begin{document}$ \alpha$\end{document}, we found genomes in both species containing segments very similar or identical to homologous regions from the other species. We performed extensive checks to verify that these segments were not assembly artifacts or cross-contamination between reaction wells during whole-genome amplification (Appendix 10). Such segments could represent accessory genes transferred between species, as has been widely reported in both Synechococcus and other species (Bhaya et al., 2007; Smillie et al., 2011). Alternatively, they could be the result of homologous recombination with other species within the core genome. We refer to the later process as hybridization throughout. The rest of our analysis aims to quantify the extent of hybridization and its impact on the evolution of the population. We began by identifying the core genes shared by α and β. Gene-level analysis reveals distinct patterns of hybridization between species We defined core genes by constructing groups of orthologous genes (orthogroups) and identifying those that were likely present in all genomes. Specifically, we clustered gene sequences in several steps using standard methods based on similarity (Appendix 1). We used phylogenetic distances to identify and separate distant paralogs within the same orthogroup. To assign clusters corresponding to each species, we analyzed the abundance distributions of orthogroups in α and β separately. Because of the incomplete nature of our genomes (average coverage ≈0.3, with a range of ≈0.05–0.95; see Appendix 1), the assignment has to be done statistically. Distributions for both species were bimodal, with a peak near zero corresponding to flexible genes and another at high abundances corresponding to core genes. We used a binomial fit to determine single-copy genes that were likely present in all genomes and found 1825 core α and 1737 core β orthogroups. Of these, 1448 orthogroups were shared by both α and β and make up the Synechococcus core genes. We focus the rest of our analysis on these core genes. We first performed a quantitative comparison of the synonymous diversity between α and β across the core genes. We hypothesized that the low-diversity segments of the α genome were the result of mutation accumulation since their common ancestor. If mutation rates across genes are similar, we expect the fraction of polymorphic sites fp\begin{document}$ f_{p}$\end{document} for each gene to follow a Poisson distribution. Empirically, we found that for fp<0.05\begin{document}$ f_{p} < 0.05$\end{document}, the data closely matched the Poisson expectation with a single fit parameter (Figure 3B). This group, which we call the ancestral backbone ofα\begin{document}$ \alpha$\end{document}, contained around half of the core genes (675 out of 1448). Note that this result is distinct from the distribution of divergences between species, which was shown previously to also follow a Poisson distribution (Rosen et al., 2018). The remaining half (773 out of 1448) had up to ten times higher fraction of segregating sites and were inconsistent with the backbone polymorphism rate. As our subsequent analysis shows, the high diversity in these genes is the result of hybridization (see Figures 4—7 and subsequent discussion). We therefore refer to this group as the hybrid α genes. In β, the distribution of fp\begin{document}$ f_{p}$\end{document} had a single peak, but a simple Poisson process was a poor fit to the data (Figure 3C). This suggests a substantial fraction of the diversity is not simply due to mutations but is introduced through recombination with other strains or species. Consistent with this interpretation, we found the distribution of synonymous diversities in the hybrid α overlapped that of β, and both were two orders of magnitude higher than in the α backbone (Figure 3D). These results suggest that hybridization had a large impact on the genetic diversity of both α and β. We therefore developed a method to identify hybrid segments directly and test this hypothesis. Alignments reveal distinct patterns of hybridization across different genes. Three core gene alignments, each chosen to illustrate a distinct pattern of hybridization, are shown. Rows represent different cells and columns different positions along the gene. One sequence was chosen randomly as a reference (top row), with nucleotides shown in different colors. The species assigned to each cell are indicated by the colored rectangles on the left, using the same color scheme as the species clusters (α shown in orange, β in blue, and γ in green). Rows were sorted hierarchically based on pairwise distances. Nucleotides are shown in gray if they are identical to the reference genome and in a different color if different. Species clusters were assigned as described in the main text and Appendix 1. (A) Segment of an alignment containing three distinct species clusters of a non-hybrid gene. (B) Segment of an alignment containing two distinct clusters (α and β) and six different simple hybrids. The direction of the transfer was inferred as described in the main text and is indicated on the left. Transfers from α into β and β into α are highlighted in orange and blue, respectively. Note that the presence of more than one type of transfer is rare and was chosen here for illustration. (C) Segment of an alignment without distinct species clusters chosen to illustrate mosaic hybrid genes. Regions with clear hybrid blocks are highlighted in purple. Note the extensive hybridization on short length scales, despite the fact that sequences from the same species cluster with each other based on pairwise distances. Transfers of whole-gene alleles between species are common. The frequency of hybrid gene alleles along the α (A) and β (B) genomes. Different colors show the donor species, with unknown species labeled X\begin{document}$ X$\end{document} and shown in magenta. The orange lines on top show mosaic loci without distinct gene clusters. Note that the x-axis in (B) is shared with (C). (C) The mean synonymous divergence between α and β at homologous loci, ordered according to OS-B’. Smoothed line over a sliding window of five genes with a step size of two genes is shown in black. Maximum and minimum divergences at each locus are shown in red and blue, respectively. (Insets, middle) A zoomed-in region from a typical genomic region (left) and a divergence trough (right), with dots representing individual genes. (Insets, bottom) Alignment segments of individual genes, subsampled to 40 sequences for clarity. The first sequence is arbitrarily chosen as the reference with each nucleotide represented by a different color. Other sites are colored gray if same as reference or using nucleotide colors if mutated. Left-colored rectangles show the species of each genome. Diversity within hybrid genes and divergence troughs reveals mixture of soft and hard sweeps. (A) The distribution of synonymous diversities in α genomes in the divergence troughs (purple histogram) is compared to a random sample of genes from the α backbone of the same size (orange histogram). The orange line shows the bootstrapped distribution of the α backbone. (B) Same as (A), but comparing divergence troughs in β genomes with the β core genome. (C) Each circle shows the synonymous diversity of a non-singleton simple hybrid locus (see Figure 5AB) in α. The donor species are shown in different colors as labeled on the x-axis. For transfers between α and β, the diversity within β is shown as squares of the same color as a control, with sequences from the same locus connected by a line. The orange horizontal lines show the synonymous diversity in the low-diversity α backbone. Transfers above this line likely contain multiple independent events. Hybridization between species leads to mixing of short DNA segments within single genes. (A, top) Alignment segment of α genomes from a representative locus with four SNP blocks. The inferred ancestral α haplotype (labeled α1\begin{document}$ \alpha_{1}$\end{document}) is highlighted in orange and the hybrid haplotype (all β here; labeled α2\begin{document}$ \alpha_{2}$\end{document}) is shown in blue. Note the SNPs in different blocks in different sets of cells. (Middle) Illustration of SNP block analysis pipeline. (Right) The consensus alignment of the two α haplotypes, β, and γ\begin{document}$ \gamma$\end{document} is constructed. (Left) Pairwise divergences between consensus sequences are calculated. Note that α2\begin{document}$ \alpha_{2}$\end{document} is identical to β here. (B) The joint distribution of divergences of α SNP block haplotypes from the other species for all SNP blocks. Each point represents a haplotype with coordinates showing the divergence from β (x-axis) and γ\begin{document}$ \gamma$\end{document} (y-axis). The orange and blue dots show the two haplotypes from the example in (A). Marginal distributions are shown on the top and right. Note the high density of haplotypes near the axes, indicating transfer from one of the other known species. (C) Same as panel (B) for β block haplotypes. Note the much larger number of haplotypes far from the axes compared to (B), indicating transfers of unknown origin. (D) Histogram of linkage coefficients between pairs of block haplotypes at different loci for α (orange histogram) and β\begin{document}$ \beta$\end{document} (blue histogram) compared to the expectation under a fully unlinked model (solid orange and blue lines, respectively). We used hierarchical clustering to identify and characterize hybridization patterns between species. If hybridization is common, as in our case, inferences based on phylogenetic trees would not provide reliable estimates of pairwise divergences. We therefore used pairwise nucleotide distances directly. The distribution of pairwise distances was bimodal, with a clear gap around d=0.075\begin{document}$ d=0.075$\end{document}, consistent with our comparisons to the reference genomes (Figure 1). Based on this observation, we defined species clusters for each gene using a simple cutoff at dc=0.075\begin{document}$ d_{c}=0.075$\end{document} (Appendix 1—figure 2) together with additional constraints to ensure that clusters were well separated and consistent with asexual divergence from a common ancestor (Appendix 1). Detailed analysis of species clusters revealed evidence of hybridization in roughly half of the genome. We found two distinct patterns of hybridization. Around 75% of core genes (1078 out of 1448) had two to three clusters of sizes roughly proportional to abundances of α, β, and γ in the samples (Appendix 5—figure 1). The majority of these genes (844 out of 1078) had clusters, each of which contained sequences from a single species (Figure 4A), as expected if they were asexually diverged. We call these non-hybrid loci. A significant minority (234 out of 1078) contained a mixture of sequences from different species (Figure 4B), representing recent transfers of whole genes between species. We call these simple hybrid loci. The remaining 25% of core genes (370 out of 1448) formed either a single cluster or had one very large cluster containing most sequences (Figure 4C). As we show below, these large clusters result from complex patterns of hybridization within single genes. We thus call these mosaic hybrid loci. In around a quarter of these loci, we found evidence of complete or near-complete gene sweeps, but in most, the divergence between α and was typical. Almost half (297 out of 773) of high-diversity α genes were mosaic hybrids. To better understand the dynamics of hybridization in the population, we analyzed both hybridization patterns in detail, starting with the simple hybrid loci. Simple hybrid genes are common and reveal distinct patterns of hybridization across species We used a simple set of heuristic criteria to assign the transfer direction of simple hybrids. If species evolved by asexual divergence from a common ancestor, we expect each orthogroup to contain two to three clusters of sizes roughly proportional to the species abundances. For example, based on the average coverage and the proportion of β in our samples (Figure 1A), we expect the largest cluster to contain ∼80 sequences from β genomes. In some cases (45 out of 1063), we also find a minority of sequences from α\begin{document}$ \alpha$\end{document} genomes in this β cluster, which we infer to be the result of a hybridization event from β into α (Figure 4B). An alternative explanation would be a transfer from a third species that underwent a full selective sweep through β and only a partial sweep through α. This scenario would imply that hybridization is even more common than we infer, so our method is conservative. We infer transfers from α into β in a similar way (Figure 4B). When the third cluster is present and contains γ, we assign all other sequences in the cluster to hybridization with γ. Because we only have a single γ sequence, these clusters are difficult to classify. We cannot distinguish between transfer from and to γ, and it is possible that some simple hybrid loci involving γ may actually be mosaic hybrids. When the third cluster is present but does not contain γ or in cases where we find a fourth cluster, we assign those sequences to hybridization with an unknown species X\begin{document}$ X$\end{document}. Using this procedure, we systematically determine the frequencies of different hybridization events in α and β. We found simple hybrids were present in 5−10% of core genes in both species (Figure 5A, B). Specifically, we found 185 simple hybrid loci in α and 68 in β. Note that we conservatively excluded contigs that were fully hybrid from this analysis, so our numbers likely underestimate hybridization rates in the population (Appendix 5). In both species, simple hybrids were scattered along the genome and contained alleles from different donor species (shown in different colors in Figure 5A, B), implying multiple hybridization events occurred independently, as expected when recombined segments are much shorter than the genome. Around 80% of α hybrid alleles were present in multiple cells, while in β the fraction was lower, at around 50%. The frequencies of simple hybrids were highly variable, ranging from singletons to frequencies close to 100%, consistent with a continual hybridization scenario. Surprisingly, around half of hybrids in β and a fifth of those in α were from unknown donors. However, a systematic search over the 34 metagenomic samples did not find evidence of other Synechococcus species in either spring (Appendix 3—figures 5–7). These sequences could be the result of hybridization with species that were formerly present in the caldera and are now either extinct or at very low abundances. Alternatively, the hybrid sequences could have resulted from hybridization with, as yet, undiscovered species from other springs. But regardless of the exact scenario, overall, these results suggest the current population is descended from multiple cyanobacterial species that hybridized with each other. Soft and hard selective gene sweeps drive hybridization of whole genes Is the genomic mixing between species the result of neutral processes or does positive selection for beneficial hybrids play an important role? Distinguishing between these alternatives is difficult in general, so we instead focused on cases where beneficial hybrid genes swept through the entire host species, that is, full gene sweeps. Full gene sweeps would greatly reduce the synonymous divergence between α and β, making them much easier to detect. Visually, we observed the synonymous divergence between α and β contained several divergence troughs, in which the average diversity was well below typical values. We identified the genes contained in these troughs by imposing a simple cutoff dtrough=0.2\begin{document}$ d_{trough}=0.2$\end{document} on the synonymous divergence between α and β, which we chose close to the inflection point of the synonymous divergence distribution across genes (data not shown). A systematic search revealed 91 very low divergence loci, spread across more than half a dozen divergence troughs (Figure 5C). Comparing alignments of typical genes and divergence through genes revealed a dramatic difference in fixed substitutions between species, which can only be explained by a recent selective sweep. The largest divergence trough contained 27 genes, representing around 25% of divergence trough genes, most of which were part of a nitrogen-fixation pathway. In addition, the nitrogen-fixation pathway is one of the largest genomic segments in which the gene order between α and β is conserved (Bhaya et al., 2007), suggesting all genes were transferred simultaneously. The most likely explanation is that at least one of the ancestors acquired the ability to fix nitrogen after it colonized the Yellowstone caldera and that this gave it a substantial selective advantage. The genetic diversity within the pathway genes gave further clues about the hybridization process. The diversity within the trough was similar in both species (πS=0.052\begin{document}$ \pi_{S}=0.052$\end{document} in α and πS=0.039\begin{document}$ \pi_{S}=0.039$\end{document} in β) and was close to the average diversity within β (πS=0.042\begin{document}$ \pi_{S}=0.042$\end{document}), suggesting the β ancestor already had the ability to fix nitrogen when it colonized the caldera. However, the trough diversity was more than an order of magnitude larger than the diversity within the α backbone (Figure 6A, B). This large discrepancy strongly suggests α acquired the pathway from β through multiple independent transfers after they colonized the Yellowstone caldera. We performed a similar analysis of diversity within simple hybrid genes to determine the direction and timing of transfers. We focused our analysis on α, where we had more data. Presence of hybrid genes at high frequencies suggests the hybridization process may be driven by selection. Two distinct patterns of diversity could result. If a single allele was transferred into α and then underwent a selective sweep, we would see very little diversity among the hybrid alleles. This is known as a hard sweep. Alternatively, if multiple hybrid alleles were transferred into α\begin{document}$ \alpha$\end{document} while undergoing a sweep, the diversity would be comparable to that in the host species. This is the pattern we observed in the divergence troughs and is known as a soft sweep. We found evidence that both hard and soft sweeps contribute to hybridization in the population (Figure 6C). Around 15% of α simple hybrids (31 out of 153 genes) had no diversity at all, regardless of the donor species, consistent with hard sweeps driving their spread. In the remaining 85% (122 out of 153 genes), diversity within α was 1–2 orders of magnitude higher than the α backbone diversity. As a result, the diversity could not be due to mutation accumulation since the common α ancestor, but must have originated from multiple independent transfers—that is, a soft sweep. We further verified the presence of soft sweeps within α by comparing the diversity of individual genes. In most cases, the diversity within α at these loci was similar to β, with a handful of cases of α having significantly less, consistent with a subset of β alleles sweeping through the population simultaneously (Figure 6C). Recombination within hybrid genes leads to extensive genomic mixing on short genomic length scales Selective sweeps of individual genes could also lead to mixing of other genomic segments from the donor species into the host population through genetic hitchhiking. If the diversity within the two hybridizing species is low, this process would result in blocks of tightly linked SNPs consisting of primarily two haplotypes, one from the host and another from the donor (Figure 7A). Anecdotally, we frequently observed such patterns in mosaic hybrid genes (Figures 4C and 7A). We developed a systematic method to search for SNP blocks based on clustering the SNP linkage matrix. Briefly, we search for consecutive SNPs that have perfect linkage (see Appendix 6). We used α hybrid genes to calibrate our method and chose very conservative parameters to avoid false positives. Consistent with this, only around 17% of SNPs from the α hybrid genes belonged to blocks. Despite the very conservative thresholds used in our algorithm, we found over one thousand SNP blocks in each species, scattered across hundreds of genes, including in highly conserved ribosomal genes and genes in which distinct species clusters could be assigned (Figure 7BC; Appendix 6—figures 1 and 2). Most blocks were short, with typical lengths around 70 bp in α and around 30 bp in β (Appendix 6—figure 1). The divergences between SNP block haplotypes were similar to that between species, consistent with hybridization (Appendix 6—figure 3). To confirm their hybrid origin, we mapped the haplotypes from each block to the two other species in our sample. We found 66% (406 out of 605) of SNP blocks within α and 15% of those within β (128 out of 776) contained a haplotype identical to consensus of other species (Figure 7B). Note that we only considered blocks in which all three species were present for this comparison. In addition, we found a comparable number of blocks that were statistically similar but did not match the other species α, β, and γ\begin{document}$ \gamma$\end{document} (Figure 7B, C). The most likely explanation is that these blocks resulted from hybridization with other species, similar to the whole-gene hybrids found previously (Figure 5). How much linkage is maintained between blocks? Given the relatively high divergences between species, one might expect epistatic interactions between blocks to constrain the genotypic structure of the population. Conversely, if recombination is dominant, each genome would be a mosaic of segments from the two ancestral species. To answer this question, we calculated the linkage coefficients between haplotypes for each pair of blocks and compared their distribution to a fully unlinked control (see Appendix 6 for details). Within the β species, the two distributions were almost identical, implying that the genomes are consistent with a random mixture of block haplotypes, as was previously proposed (Rosen et al., 2015). Within α\begin{document}$ \alpha$\end{document} the linkage coefficients were statistically larger than those predicted from the unlinked control (Figure 7D), but the overall value was still low, with ⟨r2⟩≈0.12\begin{document}$ \left\langle r^{2}\right\rangle\approx 0.12$\end{document} for data compared to ⟨r2⟩≈0.05\begin{document}$ \left\langle r^{2}\right\rangle\approx 0.05$\end{document} for unlinked control. These results are consistent with the slower rate of linkage decrease within α\begin{document}$ \alpha$\end{document} compared to β discussed previously. We also found similar results for the linkage between hybrid gene alleles (Appendix 4). Thus, rather than finding clonal subpopulations, every one of the ∼300 cells in our sample has a unique combination of SNP block haplotypes. Discussion By analyzing over 300 Synechococcus single-cell genomes, we show that hybridization plays a major role in shaping their genomic diversity. Our results suggest a simple scenario for the evolutionary history of the population (Figure 8; see also Appendix 7). Multiple strains were geographically separated and evolved independently into distinct species from an ancestral population. They then colonized the Yellowstone caldera after its formation around 640,000 years ago and have since been gradually hybridizing. The presence of genes with multiple-sequence clusters and previous analysis of 16S sequences (Rosen et al., 2018) suggests at least four independent colonizers, but the current population is dominated by the descendants of only three: the ancestors of α, β, and γ. The β population has been hybridizing with others and recombining with itself long enough that its genomes are a mosaic of different ancestral segments, consistent with a quasisexual population. But α still exhibits a large fraction of low-diversity genomic segments, interlaced with clearly hybridized segments. Molecular clock estimates using the low-diversity regions of the α genomes suggest it emerged around 104−5\begin{document}$ 10^{4-5}$\end{document} years ago and on these time scales the population is well-mixed across springs (Appendix 9—figure 1). Using the hybridized genomic segments of α as a control, we also estimated the contribution of de novo mutations to the β diversity. This analysis suggests that the ancestor of β colonized the Yellowstone caldera soon after its formation (Appendix 7). Hybridization between previously isolated lineages is the main source of diversity within the population. (Main panel) An illustration of the evolutionary scenario consistent with the main results of this paper. A single common ancestor (gray) for the population (t0\begin{document}$ t_{0}$\end{document}) diverged into several species (four shown here) following spatial isolation (t1\begin{document}$ t_{1}$\end{document}). The species colonized the Yellowstone caldera, represented by the yellow rectangle, at different times (t2−t3\begin{document}$ t_{2}-t_3$\end{document}) and began hybridizing. This process continued up to the present (t4\begin{document}$ t_{4}$\end{document}). (Top inset) Illustration of the mechanism for generating unlinked SNP blocks through recombination between a hybrid segment (blue) containing a beneficial region (red star) and the rest of the host species (orange). Note the four distinct haplotypes highlighted at the end. (Bottom inset) Zoom in of the ancestry of a particular genome (middle) compared with the donor (orange, top) and host (green, bottom) ancestral genomes. After the initial hybridization, recombination with other host genomes (red arrows) creates a mosaic structure across a range of genomic length scales. Diversity within species is often modeled using neutral processes (Didelot and Wilson, 2015; Rocha, 2018; Chen et al., 2022). But in this Synechococcus population, there is overwhelming evidence for the crucial role played by selection. The divergence troughs resulted from selective sweeps through the whole population that likely overcame mechanistic barriers to recombination, genetic incompatibilities, and ecological differences, evident in the apparent preferences to different temperatures of the three species. These results are consistent with previous analyses showing that recombination events between pathogenic Streptococcus species were primarily driven by selection (Lefébure and Stanhope, 2007; Shapiro et al., 2009). Moreover, the thousands of blocks composed of tightly linked SNPs we observe are difficult to explain through neutral drift, due to the long time it would take for them to reach observable frequencies. Instead, genetic hitchhiking during selective sweeps is by far the most likely explanation. Selective sweeps of single genes through subpopulations have been argued to lead to the formation of separate ecotypes, recombination barriers, and eventual speciation (Croucher et al., 2011; Shapiro et al., 2012). But we find multiple cases of genes from one species partially sweeping through another, sometimes via several independent transfer events. Thus, gene sweeps do not necessarily lead to ecological specialization, but can do the opposite, acting to homogenize the population. Overall, the population shows abundant evidence of hybridization on a wide range of genomic scales leading to both homogenization between species, through the erosion of genomic clusters, and diversification within species. Our results suggest several directions for future work. A significant fraction of SNP blocks in α\begin{document}$ \alpha$\end{document} and the majority of those in β were transferred from species not present in our samples. But further work is needed to determine the precise number of hybridizing species and whether some may still be present at high abundances in other springs. Metagenome samples from a wider range of springs would provide valuable information for addressing this question (see Appendix 3). Another important question is how do variants reach observable frequencies in the population. Our analysis revealed direct evidence for genetic sweeps at several hundred loci within divergence troughs and simple hybrid loci. But these loci are only a lower bound on the effect of the selection within the population. A more quantitative analysis of SNP statistics would provide important insights into evolutionary dynamics at these intermediate time scales, which could be informative for understanding bacterial evolution more broadly. Whether the erosion of genomic clusters we observe would continue indefinitely remains an open question. The hybridization process appears to be very slow: we estimate that the genomic mixing we observe occurred over at least 104 years and likely started soon after α became abundant (see Appendix 7). It is therefore possible that the current population is an evolutionary transient that would eventually lead to a single hybrid population. This scenario would be contrary to the prevailing view that microbial evolution leads to increased specialization (Shapiro et al., 2012; Jain et al., 2018). Distinguishing between these two scenarios would require a detailed understanding of how the diversity changes during the hybridization process. For example, a better theoretical understanding of how SNP correlations within and between SNP blocks change after an initial transfer could generate testable predictions about single- and multi-site frequency spectra that could be checked in the data. Some statistical features will likely be more sensitive to selective pressures that act at the whole-genome level and could help maintain clusters indefinitely. Similar approaches could shed light on the evolutionary history of other highly recombining bacteria, such as H. pylori or SAR11 (Vergin et al., 2007; Kennemann et al., 2011). Thus, how extensive hybridization affects, and is affected by, ecology and how the interplay between selection at both gene and genome scales shapes the diversity of long-term coexisting populations are intriguing questions for future work. The importance of hybridization in eukaryotic evolution is increasingly being recognized, but there have been few investigations in prokaryotes (Sankararaman et al., 2014; Moran et al., 2021). Metagenomic studies on communities in acid mine drainages revealed several hybrid bacterial and archaeal populations whose genomes were mosaics of hundreds of kbp segments from two different ancestors (Tyson et al., 2004; Lo et al., 2007). Those results are consistent with very recent hybridization, after the start of mining operations (Denef and Banfield, 2012). Our study shows the effects of this process over much longer time scales. At the level of individual genes, evidence for hybridization has been observed among commensal and pathogenic species of Campylobacter (Sheppard et al., 2008; Mourkas et al., 2022). But unlike the thermophilic Synechococcus, such host-associated bacteria are globally dispersed and have likely sampled many different environments, on a wide range of time scales, during their evolutionary history. It is thus difficult to identify which evolutionary forces drive their hybridization. For the Yellowstone Synechococcus, a fortuitous combination of spatial separation and later mixing on the same time scales on which speciation and then hybridization can extensively shape—but not completely overwrite—their diversity has enabled us to infer a great deal about the evolutionary history and the effects of full and partial sweeps. This provides a window into processes that shape species-level bacterial diversity. But for most bacterial populations, the effects of these processes are obscured by the broad range—and little knowledge—of the time scales on which different strains have co-occurred and are able to exchange DNA. Investigating whether hybridization plays a major role in other communities will thus require developing a predictive theory that can distinguish between possible evolutionary scenarios. This study provides important insights into bacterial genome evolution by analyzing single-cell genome sequences of cyanobacteria from Yellowstone hot springs. Using compelling evidence, the authors demonstrate that both homologous recombination within species and frequent hybridization across species are major drivers of genome diversification. Despite the challenges that are inherent to sparse and fragmented single-cell data, the analyses are thorough, carefully controlled, and supported by multiple complementary approaches, making the conclusions highly robust. This work represents a significant advance in our understanding of microbial evolution in natural environments. Summary: What are the overarching principles by which prokaryotic genomes evolve? This fundamental question motivates the investigations in this excellent piece of work. While it is still very common in this field to simply assume that prokaryotic genome evolution can be described by a standard model from mathematical population genetics, and fit the genomic data to such a model, a smaller group of researchers rightly insists that we should not have such preconceived ideas and instead try to carefully look at what the genomic data tell us about how prokaryotic genomes evolve. This is the approach taken by the authors of this work. Lacking a tight theoretical framework, the challenge of such approaches is to device analysis methods that are robust to all our uncertainties about what the underlying evolutionary dynamics might be. The authors here focus on a collection of ~300 single-cell genomes from a relatively well-isolated habitat with a relatively simple species composition, i.e. cyanobacteria living in hot springs in Yellowstone National Park. They convincingly demonstrate that the relative simplicity of this habitat increases our ability to interpret what the genomic data tells us about the evolutionary dynamics. Using a very thorough and multi-faceted analysis of these data, the authors convincingly show that there are three main species of Synechococcus cyanobacteria living in this habitat, and that apart from very frequent recombination within each species (which is in line with insights from other recent studies) there is also a remarkably frequent occurrence of hybridization events between the different species, and with as of yet unindentified other genomes. Moreover, these hybridization events drive much of the diversity within each species. The authors also show convincing evidence that many of these hybridization events are not neutral but are driven by natural selection. Strengths: The great strength of this paper is that, by not making any preconceived assumptions about what the evolutionary dynamics is expected to look like, but instead devicing careful analysis methods to tease apart what the data tells us about what has happened in the evolution in these genomes, highly novel and unexpected results are obtained, i.e. the major role of hybridization across the 3 main species living in this habitat. The analysis is very thorough and reading the detailed descriptions in the appendices it is clear that these authors took a lot of care in using these methods and avoiding the pitfalls that unfortunately affect many other studies in this research area. The picture of the evolutionary dynamics of these three Synechococcus species that emerges from this analysis is quite novel and surprising. I think this study is a major stepping stone toward development of more realistic quantitative theories of genome evolution in prokaryotes. The analysis methods that the authors employ are also partially quite novel and will no doubt by very valuable for analysis of many other datasets. Weaknesses: The main text is tight and concise, but this sort of hides the very large amount of careful complementary analyses that went into the conclusions presented in the main text. The appendices are quite well written but they are substantial, so that really understanding the paper is not an easy read. However, I do not really think the authors can be faulted for this. The topic is complex and a lot of care is required to make sure conclusions are valid. A very interesting observation is that a lot of hybridization events (i.e. about half) originate from species other than the alpha, beta, and gamma Synechococcus species from which the genomes that are analyzed here derive. For this to occur, these other species must presumably also be living in the same habitat and must be relatively abundant. But if they are, why are they not being captured by the sampling? I did not see a clear explanation for this very common occurrence of hybridization events from outside of these Synechococcus species. The authors raise the possibility that these other species used to live in these hot springs but are now extinct or that the occur in other pools. I guess this is possible but I still find it puzzling and wonder if these donors could have been filtered out at some step of the experimental and/or analysis procedures. Summary. Birzu et al. describe two sympatric hotspring cyanobacterial species ("alpha" and "beta") and infer recombination across the genome, including inter-species recombination events (hybridization) based on single-cell genome sequencing. The evidence for hybridization is strong and the authors took care to control for artefacts such as contamination during sequencing library preparation. Despite hybridization, the species remain genetically distinct from each other. The authors also present evidence for selective sweeps of genes across both species - a phenomenon which is widely observed for antibiotic resistance genes in pathogens, but rarely documented in environmental bacteria. Strengths. This manuscript describes some of the most thorough and convincing evidence to date of recombination happening within and between co-habitating bacteria in nature. Their single-cell sequencing approach allows them to sample the genetic diversity from two dominant species. Although single-cell genome sequences are incomplete, they contain much more information about genetic linkage than typical short-read shotgun metagenomes, enabling a reliable analysis of recombination. The authors also go to great lengths to quality-filter the single-cell sequencing data and to exclude contamination and read mismapping as major drivers of the signal of recombination. This is a fascinating dataset with intricate analyses showing the great extent of between-species hybridization that is possible in nature. Weaknesses. This revised version is much improved, with a much clearer flow and organisation within both the main text and supplement. The remaining weaknesses that I note below are certainly not critical, but are simply useful context for the reader to keep in mind. My main concern is that the evidence for selection on the hybridized genes is incomplete and statements about the 'overwhelming evidence for the crucial role played by selection' (lines 334-5) are a bit overstated. What fraction of the hybridization events were driven by positive selection? The breakdown of hard (15%) vs soft (85%) sweeps is given, out of 153 (as sidenote, it is not clear if this is 153 genes or events, troughs, etc.). But how many of the hybridization events (or genes) have evidence for a selective sweep relative to those that do not? I recognize that this may be a hard question to answer, because it may be statistically easier to identify a hybridization event that rises to high frequency due to positive selection from a neutral event that remains rare. Even a rough estimate would be useful; would it be something like 153 out of the number of core genes tested (~700)? Regardless, I think that Figure 6 (A and B) could benefit from comparison to a neutral model, including hybridization but no selection to see if a similar pattern (notably, higher synonymous diversity in alpha troughs compared to the backbone) could arise due to hybridization alone without selection. An implicit assumption in microbiology is often that cross-species recombination events are driven by selection. The authors recognize that "diversity troughs resulted from selective sweeps [...] likely overcame mechanistic barriers to recombination, genetic incompatibilities, and ecological differences" (lines 335-7) and thus would not be retained unless they had some strong adaptive value to offset these costs. There are surprisingly few tests of the hypothesis that cross-species recombination events tend to be driven by selection. An analysis of Streptococcus spp. genomes showed that between-species recombination events tended to be accompanied by positive selection, whereas most within-species events were not (Shapiro et al. Trends in Microbiology 2009; reanalysis of data from Lefebure & Stanhope, Genome Biology 2007). There are probably other examples out there, but the authors could highlight that they provide rare data to support a common expectation. The following is the authors’ response to the original reviews Public Reviews: Reviewer #1 (Public Review): Summary: What are the overarching principles by which prokaryotic genomes evolve? This fundamental question motivates the investigations in this excellent piece of work. While it is still very common in this field to simply assume that prokaryotic genome evolution can be described by a standard model from mathematical population genetics, and fit the genomic data to such a model, a smaller group of researchers rightly insists that we should not have such preconceived ideas and instead try to carefully look at what the genomic data tell us about how prokaryotic genomes evolve. This is the approach taken by the authors of this work. Lacking a tight theoretical framework, the challenge of such approaches is to devise analysis methods that are robust to all our uncertainties about what the underlying evolutionary dynamics might be. The authors here focus on a collection of ~300 single-cell genomes from a relatively well-isolated habitat with relatively simple species composition, i.e. cyanobacteria living in hotsprings in Yellowstone National Park, and convincingly demonstrate that the relative simplicity of this habitat increases our ability to interpret what the genomic data tells us about the evolutionary dynamics. Using a very thorough and multi-faceted analysis of these data, the authors convincingly show that there are three main species of Synechococcus cyanobacteria living in this habitat, and that apart from very frequent recombination within each species (which is in line with insights from other recent studies) there is also a remarkably frequent occurrence of hybridization events between the different species, and with as of yet unidentified other genomes. Moreover, these hybridization events drive much of the diversity within each species. The authors also show convincing evidence that these hybridization events are not neutral but are driven by selected by natural selection. Strengths: The great strength of this paper is that, by not making any preconceived assumptions about what the evolutionary dynamics is expected to look like, but instead devising careful analysis methods to tease apart what the data tells us about what has happened in the evolution in these genomes, highly novel and unexpected results are obtained, i.e. the major role of hybridization across the 3 main species living in this habitat. The analysis is very thorough and reading the detailed supplementary material it is clear that these authors took a lot of care in devising these methods and avoiding the pitfalls that unfortunately affect many other studies in this research area. The picture of the evolutionary dynamics of these three Synechococcus species that emerge from this analysis is highly novel and surprising. I think this study is a major stepping stone toward the development of more realistic quantitative theories of genome evolution in prokaryotes. The analysis methods that the authors employ are also partially novel and will no doubt be very valuable for analysis of many other datasets. We thank the reviewer for their appreciation of our work. Weaknesses: I feel the main weakness of this paper is that the presentation is structured such that it is extremely difficult to read. I feel readers have essentially no chance to understand the main text without first fully reading the 50-page supplement with methods and 31 supplementary materials. I think this will unfortunately strongly narrow the audience for this paper and below in the recommendations for the authors I make some suggestions as to how this might be improved. A very interesting observation is that a lot of hybridization events (i.e. about half) originate from species other than the alpha, beta, and gamma Synechococcus species from which the genomes that are analyzed here derive. For this to occur, these other species must presumably also be living in the same habitat and must be relatively abundant. But if they are, why are they not being captured by the sampling? I did not see a clear explanation for this very common occurrence of hybridization events from outside of these Synechococcus species. The authors raise the possibility that these other species used to live in these hot springs but are now extinct. I'm not sure how plausible this is and wonder if there would be some way to find support for this in the data (e.g that one does not observe recent events of import from one of these unknown other species). This was one major finding that I believe went without a clear interpretation. We agree with the reviewer that the extent of hybridization with other species is surprising. While we do feel that our metagenome data provide convincing evidence that “X” species are not present in MS or OS, we cannot currently rule out the presence of X in other springs. In the revision we explicitly mention the alternative hypothesis (Lines 239-242). The core entities in the paper are groups of orthologous genes that show clear evidence of hybridization. It is thus very frustating that exactly the methods for identifying and classifying these hybridization events were really difficult to understand (sections I and V of the supplement). Even after several readings, I was unsure of exactly how orthogroups were classified, i.e. what the difference between M and X clusters is, what a `simple hybrid' corresponds to (as opposed to complex hybrids?), what precisely the definitions of singlet and non-singlet hybrids are, etcetera. It also seems that some numbers reported in the main text do not match what is shown in the supplement. For example, the main text talks about "around 80 genes with more than three clusters (SM, Sec. V; fig. S17).", but there is no group with around 80 genes shown in Fig S17! And similarly, it says "We found several dozen (100 in α and 84 in β) simple hybrid loci" and I also cannot match those numbers to what is shown in the supplement. I am convinced that what the authors did probably made sense. But as a reader, it is frustrating that when one tries to understand the results in detail, it is very difficult to understand what exactly is going on. I mention this example in detail because the hybrid classification is the core of this paper, but I had similar problems in other sections. We thank the reviewer for pointing out these issues with our original presentation. In the revision, we have redone most of the analysis to simplify the methods and check the consistency of the results. We did not find any qualitative differences in our results after reanalysis, but some of the numbers for different hybridization patterns have changed. The most notable difference is an increase in the number of alpha-gamma simple hybrids and a corresponding decrease in mixed-species clusters (now labeled mosaic hybrids). These transfers are difficult to assign because we only have access to a single gamma genome. We have added a short explanation of this point in Lines 219-222. To improve the presentation, we significantly expanded the “Results” section to better explain our analysis and the different steps we take. We included two additional figures (Figs. 3 and 4) that illustrate the different types of hybrids and the heterogeneity in the diversity of alpha which is discussed in the main text and is important for interpreting our results. We also included two additional figures (Figs. 2 and 6) that were previously in the Appendix but were mentioned in the main text. We believe these changes should address most of the issues raised by the reviewer and hopefully make the manuscript easier to read. Although I generally was quite convinced by the methods and it was clear that the authors were doing a very thorough job, there were some instances where I did not understand the analysis. For example, the way orthogroups were built is very much along the lines used by many in the field (i.e. orthoMCL on the graph of pairwise matchings, building phylogenies of connected components of the graph, splitting the phylogenies along long branches). But then to subdivide orthogroups into clusters of different species, the authors did not use the phylogenetic tree already built but instead used an ad hoc pairwise hierarchical average linkage clustering algorithm. The reviewer is correct that there is an unexplained discrepancy between the clustering methods we used at different steps in our pipeline. We followed previous work by using phylogenetic distances for the initial clustering of orthogroups. On these scales we expect hybridization to play a minor role and phylogenetic distances to correlate reasonably well with evolutionary divergence. However, because of the extensive hybridization we observed, the use of phylogenetic models for species clustering is more difficult to justify. We therefore chose to simply use pairwise nucleotide distances, which make fewer assumptions about the underlying evolutionary processes and should be more robust. We have briefly explained our reasoning and the details of our clustering method in the revision (Lines 182-190). Reviewer #2 (Public Review): Summary: Birzu et al. describe two sympatric hotspring cyanobacterial species ("alpha" and "beta") and infer recombination across the genome, including inter-species recombination events (hybridization) based on single-cell genome sequencing. The evidence for hybridization is strong and the authors took care to control for artefacts such as contamination during sequencing library preparation. Despite hybridization, the species remain genetically distinct from each other. The authors also present evidence for selective sweeps of genes across both species - a phenomenon which is widely observed for antibiotic resistance genes in pathogens, but rarely documented in environmental bacteria. Strengths: This manuscript describes some of the most thorough and convincing evidence to date of recombination happening within and between cohabitating bacteria in nature. Their single-cell sequencing approach allows them to sample the genetic diversity from two dominant species. Although single-cell genome sequences are incomplete, they contain much more information about genetic linkage than typical short-read shotgun metagenomes, enabling a reliable analysis of recombination. The authors also go to great lengths to quality-filter the single-cell sequencing data and to exclude contamination and read mismapping as major drivers of the signal of recombination. We thank the reviewer for their appreciation of our work. Weaknesses: Despite the very thorough and extensive analyses, many of the methods are bespoke and rely on reasonable but often arbitrary cutoffs (e.g. for defining gene sequence clusters etc.). Much of this is warranted, given the unique challenges of working with single-cell genome sequences, which are often quite fragmented and incomplete (30-70% of the genome covered). I think the challenges of working with this single-cell data should be addressed up-front in the main text, which would help justify the choices made for the analysis. We have significantly expanded the “Results” section to better justify and explain the choices we made during our analysis. We hope these changes address the reviewer’s concerns and make the manuscript more accessible to readers. The conclusions could also be strengthened by an analysis restricted to only a subset of the highest quality (>70% complete) genomes. Even if this results in a much smaller sample size, it could enable more standard phylogenetic methods to be applied, which could give meaningful support to the conclusions even if applied to just ~10 genomes or so from each species. By building phylogenetic trees, recombination events could be supported using bootstraps, which would add confidence to the gene sequence clustering-based analyses which rely on arbitrary cutoffs without explicit measures of support. It seems to us that the reviewer’s suggestion presupposes that the recombination events we find can be described as discrete events on an asexual phylogeny, similar to how rare mutations are treated in standard phylogenetic inference. Popular tools, such as ClonalFrame and its offshoots, have attempted to identify individual recombination events starting from these assumptions. But the main conclusion of both our linkage and SNP block analysis is that the ClonalFrame assumptions do not hold for our data. Under a clonal frame, the SNP blocks we observe should be perfectly linked, similar to mutations on an asexual tree. But our results in Fig. 7D show the opposite. Part of the issue may have been that in our original presentation, we only briefly discuss the results of our linkage analysis and refer readers to the Appendix for more details. To fix this issue we have added an extra figure (Fig. 2), showing rapid linkage decrease in both species and that at long distances the linkage values are essentially identical to the unlinked case, similar to sexual populations. We hope that this change will help clarify this point. The manuscript closes without a cartoon (Figure 4) which outlines the broad evolutionary scenario supported by the data and analysis. I agree with the overall picture, but I do think that some of the temporal ordering of events, especially the timing of recombination events could be better supported by data. In particular, is there evidence that inter-species recombination events are increasing or decreasing over time? Are they currently at steady-state? This would help clarify whether a newly arrived species into the caldera experiences an initial burst of accepting DNA from already-present species (perhaps involving locally adaptive alleles), or whether recombination events are relatively constant over time. The reviewer raises some very interesting questions about the dynamics of recombination in the population, which we hope to pursue in future work. We have added this as an open question in the Discussion (Lines 365-382). These questions could be answered by counting recombination events that occur deeper or more recently in a phylogenetic tree. The reviewer here seems to presuppose that recombination is rare enough that a phylogenetic tree can reliably be inferred, which is contrary to our linkage analysis (see the response to an earlier comment). Perhaps the reviewer missed this point in our original manuscript since it was discussed primarily in the Appendix. See also our response to a previous comment by the reviewer. The cartoon also shows a 'purple' species that is initially present, then donates some DNA to the 'blue' species before going extinct. In this model, 'purple' DNA should also be donated to the more recently arrived 'orange' species, in proportion to its frequency in the 'blue' genome. This is a relatively subtle detail, but it could be tested in the real data, and this may actually help discern the order of the inferred recombination events. We have included an extra figure in the main text (Fig. 6) that addresses the question of timing of events. A quantitative test of our cartoon model along the lines the reviewer suggested would certainly be worthwhile and we hope to do that in future work. The abstract also makes a bold claim that is not well-supported by the data: "This widespread mixing is contrary to the prevailing view that ecological barriers can maintain cohesive bacterial species..." In fact, the two species are cohesive in the sense that they are identifiable based on clustering of genome-wide genetic diversity (as shown in Fig 1A). I agree that the mixing is 'widespread' in the sense that it occurs across the genome (as shown in Figure 2A) but it is clearly not sufficient to erode species boundaries. So I believe the data is consistent with a Biological Species Concept (sensu Bobay & Ochman, Genome Biology & Evolution 2017) that remains 'fuzzy' - such that there are still inter-species recombination events, just not sufficient to erode the cohesion of genomic clusters. Therefore, I think the data supports the emerging picture of most bacteria abiding by some version of a BSC, and is not particularly 'contrary' to the prevailing view. We have revised the phrase mentioned by the reviewer to “prevent genetic mixture between bacterial species,” which more accurately represents our conclusions. The final Results paragraph begins by posing a question about epistatic interactions, but fails to provide a definitive answer to the extent of epistasis in these genomes. Quantifying epistatic effects in bacterial genomes is certainly of interest, but might be beyond the scope of this paper. This could be a Discussion point rather than an underdeveloped section of the Results. We agree with the reviewer that an exhaustive analysis of epistasis in the population is beyond the scope of the manuscript. Our original intention was to answer whether SNP blocks we discovered showed evidence of strong linkage, as might be expected if only a small number of strains are present in the population. In light of the previous comments by the reviewer regarding the consistency with the clonal frame hypothesis, we believe this is especially relevant for our results. Moreover, the results we found‑especially for the beta population‑were quite conclusive: SNP block linkages in beta are indistinguishable from an unlinked model. To avoid misdirecting the reader about the significance of our results, we have revised the relevant paragraph (Lines 316-319). Recommendations For The Authors: Reviewer #1 (Recommendations For The Authors): Although I am entirely convinced of the validity of the results, methodology, and interpretations presented in this work, I must say I found the paper very hard to read. And I think I am really quite familiar with these kinds of approaches. I fear that for people other than experts on these kinds of comparative genomic analyses, this paper will be almost impossible to read. With the aim of expanding the audience for this compelling work, I think the authors might want to consider ways to improve the presentation. At the end of a long project, the obtained results typically form a web of mutual interconnections and dependencies and one of the key challenges in presenting the results in a paper is having to untangle this web of connected results and analysis into a linear ordered narrative so that, at any point in the narrative, understanding the next point only depends on previous points in the narrative. I frankly feel that this paper fails at this. The paper reads to me as if one author put together the supplement by essentially writing a report of all the analyses that were done together with supplementary figures summarizing all those analyses, and that another author then wrote the main text by using the materials in the supplement almost in the way a cook uses ingredients for a dish. Almost every other sentence in the main text refers to results in the (31!) supplementary figures and can only be understood by reading the appropriate corresponding sections in the supplementary materials. I found it essentially impossible to read the main text without having first read the entire 50-page supplement. I think the paper could be hugely improved by trying to restructure the presentation so as to make it more linear. The main text can be expanded to include a summary of the crucial methods and analysis results from the supplement needed to understand the narrative in the main text. For example, as it currently stands it is really challenging to understand what is shown in figures 2 and 3 of the main text without having to first read a very substantial part of the supplement. Figure 3, even after having read the relevant sections in the supplement, took me quite a while to understand and almost felt like a puzzle to decypher. Rethinking which parts of the supplement are really necessary would also help. Finally, it would also help if the terminology was kept as simple, transparent, and consistent as possible. I understand that my suggestion to thoroughly reorganize the presentation may feel like a big hassle, but I am afraid that in its current form, these important results are essentially rendered inaccessible to all but a small group of experts in this area. This paper deserves a wider readership. We thank the reviewer for these valuable suggestions. In the revision, we have significantly expanded and restructured the “Results” section to make the presentation more linear, as the reviewer suggested (see our reply to the public comment by the reviewer for details). We hope these changes will make the manuscript easier to read. Reviewer #2 (Recommendations For The Authors): I found this paper challenging to follow since the main text was so condensed and the supplementary material so extensive. Given that eLife does not impose strong limits on the length of the main text, I suggest moving some key sections from the supplement into the main text to make it easier for the reader to follow rather than flipping back and forth. Adding to the confusion, supplementary figures were referenced out of order in the main text (e.g. S23 is referenced before S1). Please check the numbering and ensure figures are mentioned in the main text in the correct order. We thank the reviewer for their feedback on the presentation of the results. In response to similar comments from Reviewer #1, we have significantly expanded and restructured the “Results” section to make it easier to read (see also our responses to Reviewer #1). Page 2: The term 'coevolution' is typically reserved for two species that mutually impose selective pressures on one another (e.g. predator-prey interactions; see Janzen, Evolution 1980). In the context of these two cyanobacterial species, it's not clear that this is the case so I would simply refer to them 'cohabitating' or being sympatric in the same environment. It is true that the term "coevolution” has become associated with predator-prey interactions, as the reviewer said. However, we feel that in our case “coevolution” fairly accurately describes the continual hybridization over long time scales we observe. We have therefore chosen to keep the term. Page 3: The authors mention that the gamma SAG is ~70% complete, which turns out to be quite high. It would be useful to mention early in the Results the mean/median completeness across SAGs, and how this leads to some challenges in analysing the data. Some of the material from the Supplement could be moved into the Results here. We have added a short note on the completeness in the Results (Lines 153-154). We have also added an extra figure in Appendix 1 with the completeness of all the SAGs for interested readers. I was left puzzled by the sentence: "Alternatively, high rates of recombination could generate different genotypes within each genome cluster that are adapted to different temperatures, with the relative frequencies of each cluster being only a correlated and not a causal driver of temperature adaptation." This is suggesting that individual genes or alleles, rather than entire genomes, could be adapted to temperature. But figure 1B seems to imply that the entire genome is adapted to different temperatures. Anyway, this does not seem to be a key point and could probably be removed (or clarified if the authors deem this an important point, which I failed to understand). We have revised this section to clarify the alternative hypothesis mentioned by the reviewer (Lines 100-103). Page 4. 'Several dozen' hybrid genes were found, but please also specify how many genes were tested. In general, it would be good to briefly outline the sample size (SAGs or genes) considered for each analysis. We have added the total numbers of genes we analyzed at each step of our analysis. 'Mosaic hybrid loci' are mentioned alongside the issue of poor alignment. Presumably, the mosaic hybrid loci are first filtered to remove the poor alignments? This should be specified, and please mention how many loci are retained before/after this filter. We thank the reviewer for highlighting this important point. In the revision, we have implemented a more aggressive filtering of genes with poor alignments. We have added an extra paragraph to Appendix 1 (step 5 in the pipeline analysis) briefly explaining the issue. Page 5. "By contrast, the diversity of mosaic loci was typical of other loci within beta, suggesting most of the beta genome has undergone hybridization." Please point to the data (figure) to support this statement. We have restructured our discussion of the different hybrid loci so this comment is no longer relevant. In case the reviewer is interested, the synonymous diversity within beta was 0.047, while in mosaic hybrids it was 0.064. Page 6. "The largest diversity trough contained 28 genes." Since this trough is discussed in detail and seems to be of interest, it would be nice to illustrate it, perhaps as an inset in Figure 2 or as a separate figure. If I understood correctly, this trough includes genes (in a nitrogen-fixation pathway) that are present in all genomes, but are exchanged by homologous recombination. So I don't think it's correct to say that the "ancestors acquired the ability to fix nitrogen." Rather, the different alleles of these same genes were present in the ancestor. So perhaps there was a selective sweep involving alleles in this region that provided adaptation to local nitrogen sources or concentrations, but not a gain of new genes. Perhaps I misunderstood, in which case clarification would be appreciated. The reviewer raises an interesting possibility. We agree that it is in principle possible that the ancestor contained the nitrogen fixation genes and the selective sweep simply replaced the ancestral alleles. In this particular case, there is additional evidence that the entire pathway was acquired around roughly the same time from gene order. The gene order between alpha and beta is almost entirely different, with only a few segments containing more than 2-3 genes in the same order, as shown by Bhaya et al. 2007 and confirmed by additional unpublished analysis of the SAGs. One of the few exceptions is the nitrogen fixation pathway, which has essentially the same gene order over more than 20 kbp. Thus, if the ancestor of both alpha and beta contained the nitrogen-fixation pathway, we would expect these genes to be scatter across the genome. We have revised the sentences in question to clarify this point (Lines 260-271). Page 6. Last paragraph on epistasis references Fig 3C, but I believe it should be Fig 3D. Fixed. Page 7. Figure 3 legend. "Note that alpha-2 is identical to gamma here." I believe it should be beta, not gamma. The reviewer is correct. We have fixed this error. Page 8. What is the evidence for "at least six independent colonizers"? I could not find the data supporting this claim. The statement mentioned by the reviewer was based on the maximum number of species clusters we identified in different core genes. However, during the revision, we found that only a handful of genes contained five or more clusters. We did find several tens of genes with four clusters. In addition, Rosen et al. (2018) also found additional 16S clusters at low frequency in the same springs. Based on these results we conservatively estimate that at least four independent strains colonized the caldera, but the number could be much greater. We have revised the text in question accordingly (Lines 336-339) and added Fig. 2 in Appendix 1 to support the conclusion. Page 9. Line 200: "acting to homogenize the population." It should be specified that the population is only homogenized at these introgressed loci, not genome-wide. Otherwise, the genome-wide species clusters seen in Fig 1 would not be maintained. It is true that the selective sweeps that lead to diversity throughs only homogenize the introgressed loci. But other hybrid segments could also rise to high frequency in the population during the sweep through hitchhiking. The fact that we observe SNP blocks generated through secondary recombination events of introgressed segments throughout the genome supports this view. While we do not fully understand the dynamics of this process currently, we do feel that the current evidence supports the statement that mixing is occurring throughout the genome and not just at a few loci so we have kept the original statement. The final sentence (lines 221-222) is vague and uninformative. On the one hand, "investigating whether hybridization plays a major role" is what the current manuscript has already done - depending on what is meant by 'major' (how much of the genome? Or whether there are ecological implications?). It is also not clear what is meant by a predictive theory and 'possible evolutionary scenarios. This should be elaborated upon, otherwise, it is not clear what the authors mean. Otherwise, this sentence could be cut. We thank the reviewer for their feedback. One possible source of confusion could be that in this sentence we were referring to detecting hybridization in other communities. We have changed “these communities” to “other communities” to make this clearer. Supplement. Broadly speaking, I appreciate the thorough and careful analysis of the single cell data. On the other hand, it is hard to evaluate whether these custom analyses are doing what is intended in many cases. Would it be possible to consider an analysis using more established methods, e.g. taking a subset of genomes with 'good' completeness and using Panaroo to find the core and accessory genome, then ClonalFrameML or Gubbins to infer a phylogeny and recombination events? Such analyses could probably be applied to a subset of the sample with relatively complete genomes. I don't want to suggest an overly time-consuming analysis, but the authors could consider what would be feasible. We have added a comparison between our analysis and that from two other methods, including ClonalFrameML mentioned by the author. One important point that we feel might have been lost in the first version is that our linkage results imply that recombination is not rare such that it can be mapped onto an asexual tree as assumed by ClonalFrameML. Note that this is not simply due to technical limitations due to incomplete coverage and is instead a consequence of the evolutionary dynamics of the population. Consistent with this, we found several inconsistencies in how recombination events were assigned by ClonalFrameML. We have summarized these conclusions in Appendix 7 of the revised manuscript. Page 8. Line 190. What is meant by 'minimal compositional bias'? We mean that the sample is not biased towards strains that grow in the lab. We have revised the sentence to clarify. Page 25. Figure S14 is not referenced in the text. We have added part of this figure to the main text since it illustrates one of our main results, namely that sites at long genomic distances are essentially unlinked. Page 26. The 'unlinked controls' (line 530) are very useful, but it would be even more informative to see if these controls also show the same decline in linkage with distance in the genome as observed in the real data. In particular, it would be good to know if the observed rapid decline in linkage with distance in the low-diversity regions is also observed in controls. Currently, it is unclear if this observation might be due to higher uncertainty in inferring linkage in low-diversity regions, which by definition have less polymorphism to include in the linkage calculation. We thank the reviewer for the suggestion. After further consideration, we have decided to remove the subsection on linkage decrease in the low-diversity regions. We feel such detailed quantitative analysis would be better suited for a more technical paper, which we hope to do at a later time. Page 26. There are some sections with missing identifiers (Sec ??). Fixed. Page 27. The information about the typical breadth of SAG coverage (~30%) would be better to include earlier in the Supplement, and also mentioned in the main text so the reader can more easily understand the nature of the dataset. We have added an extra figure with the SAG coverages to Appendix 1. Page 29. Any sensitivity analysis around the S = 0.9 value? Even if arbitrary, could the authors provide justification why they think this value is reasonable? We have significantly revised this section in response to earlier comments by one of the reviewers. We hope that this would clarify the details of our methods to interested readers. To answer the reviewer’s specific question, we chose this heuristic after examining the fraction of cells of each species in different species clusters. For the clusters assigned to alpha and beta, we found a sharp peak near one and that a cutoff of 0.9 captured most clusters while still being high enough to inconsistent with a mixed cluster. Page 30. I could not see where Fig. S17 was mentioned in the text. Also, how are 'simple hybrid genes' defined? We have removed this figure in the revision. The definition of the different types of hybrid genes have been added to the main text in response to a comment from the other reviewer. Page 36. It is hard to see that divergence is 'high' relative to what reference. Would it be possible to include the expected value (from ref. 12) in the plot, or at least explicitly mentioned in the text? We have added the mean synonymous and non-synonymous divergences between alpha and beta to the figures for reference. Page 38. Line 770 "would be comparable to that of beta." This is not necessarily the case since beta could have a different time to its most recent common ancestor. It could have a different time to the last bottleneck or selective sweep, etc. We thank the reviewer for pointing out this misleading statement. Our point here was that in the first scenario the TMRCA of alpha and beta would be similar since the diversity in the high-diversity alpha genes is similar to beta. We have clarified this statement in the revision. Page 39. Line 793. The use of the term 'genomic backbone' implies the presence of a clonal frame, which is not what the data seems to support. Perhaps another term such as 'genetic diversity' would more appropriately capture the intended meaning here. We agree with the reviewer that the low-diversity regions may not be asexual. We used “genomic backbone” to distinguish from the “clonal frame,” which is usually used to mean that the backbone is asexual. We have added a note in the revision to clarify this point. Page 39. Lines 802-805. I found this explanation hard to follow. Could the logic be clarified? We simply meant that although the beta distribution is unimodal, it is not consistent with a simple Poisson distribution, unlike in alpha. We have added an extra sentence to clarify this.