Recombination Hotspots Research Articles

Fast and accurate deep learning scans for signatures of natural selection in genomes using FASTER-NN

Deep learning classification models based on Convolutional Neural Networks (CNNs) are increasingly used in population genetic inference for detecting signatures of natural selection. Prevailing detection methods treat the design of the classifier as a discrete phase, assuming that high classification accuracy is the sole prerequisite for precise detection. This frequently steers method development toward classification-driven optimizations that can inadvertently impede detection. We present FASTER-NN, a CNN classifier designed specifically for the precise detection of natural selection. It has higher sensitivity than state-of-the-art CNN classifiers while only processing allele frequencies and genomic positions through dilated convolutions to maximize data reuse. As a result, execution time is invariant to the sample size and the chromosome length, creating a highly suitable solution for large-scale, whole-genome scans. Furthermore, FASTER-NN can accurately identify selective sweeps in recombination hotspots, which is a highly challenging detection problem with very limited theoretical treatment to date.

Molecular Epidemiology and Genetic Evolution of Porcine Reproductive and Respiratory Syndrome Virus in Northern China During 2021-2023.

Porcine reproductive and respiratory syndrome virus (PRRSV), an important pathogen affecting the pig industry, is an RNA virus with high genetic diversity. In this study, 12,299 clinical samples were collected from northern China during 2021-2023 to investigate the molecular epidemiological characteristics and genetic evolution of PRRSV. All samples were screened using qRT-PCR and further analyzed through ORF5 gene and whole-genome sequencing. The results showed that the positive rate of PRRSV in northern China was 18.42%, and positivity rates were relatively high in spring. The phylogenetic analysis of the ORF5 gene indicated that the 174 gene sequences were classified as PRRSV-2, predominantly found in Lineage 1.8 (L1.8), Lineage 1.5 (L1.5), and Lineage 8 (L8). L1.8 and L1.5 showed considerable polymorphism at decoy and neutralizing epitopes. Mutations of specific amino acids were present in L1.8 and L1.5 at T- and B-cell epitopes. Moreover, the 27 whole-genome sequences were analyzed. As indicated, 24 of them were exposed to gene recombination, and L1.8 provided the backbone for recombination events. The predominant recombination modes were L1.8 + L8.7 + L1.5/L3, with L1.5 and L3.5 generally yielding GP2~GP6 structural proteins. Recombination hotspots were primarily located within the ranges of 780~2200 (Nsp1~Nsp2), 5400~6200 (Nsp3~Nsp4), 7800~9000 (Nsp9), and 12,200~14,800 (ORF2~ORF6). This study enriches the epidemiological data of PRRSV in northern China, thereby providing theoretical references for the prevention and control of PRRSV in northern China.

The APOL1 p.N264K variant is co-inherited with the G2 kidney disease risk variant through a proximity recombination event.

Black Americans are three to four times more likely to develop nondiabetic kidney disease than other populations. Exclusively found in people of recent African (AFR) ancestry, risk variants in Apolipoprotein L1 (APOL1) termed G1 and G2 contribute significantly to this increased susceptibility. Our group and others showed that a missense variant in APOL1, rs73885316 (p.N264K, "M1"), is remarkably protective against APOL1 kidney disease when co-inherited with the G2 risk allele. Since the distance between the M1 and G2 variants is only 367 base pairs, we initially suspected that two independent mutation events occurred to create non-risk M1-G0 and M1-G2 haplotypes. Here, we examined APOL1 haplotypes in individuals of AFR ancestry from the 1000 Genomes Project, the Nephrotic Syndrome Study Network (NEPTUNE), and an ancient individual from the Allen Ancient Genome Diversity Project to determine how the M1-G2 haplotype arose. We demonstrate that M1 most likely first appeared on a non-risk G0 haplotype, and that a subsequent recombination event bypassed strong recombination hotspots flanking APOL1 and occurred between p.N388Y389del on a G2 haplotype and M1 on a G0 haplotype to create the M1-G2 haplotype. Observing a recombination event within a small region between clinically relevant loci emphasizes the importance of studying the entire haplotype repertoire of a disease gene and the impact of haplotype backgrounds in disease susceptibility.

Multiple losses of aKRAB from PRDM9 coincide with a teleost-specific intron size distribution

BackgroundPrimary transcripts are largely comprised of intronic sequences that are excised and discarded shortly after synthesis. In vertebrates, the shape of the intron size distribution is largely constant; however, most teleost fish have a diverged log-bimodal ‘teleost distribution’ (TD) that is seen only in teleosts. How the TD evolved and to what extent this was affected by adaptative or non-adaptive mechanisms is unknown.ResultsHere, we show that the TD has evolved independently at least six times and that its appearance is linked to the loss of the aKRAB domain from PRDM9. We determined intron size distributions and identified PRDM9 orthologues from annotated genomes in addition to scanning 1193 teleost assemblies for the aKRAB domain. We show that a diverged form of PRDM9 (β\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\beta$$\\end{document}) is predominant in teleosts whereas the α\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\alpha$$\\end{document} version is absent from most species. Only a subset of PRDM9-α\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\alpha$$\\end{document} proteins contain aKRAB, and hence, it is present only in a small number of teleost lineages. Almost all lineages lacking aKRAB (but no species with) had TDs.ConclusionsIn mammals, PRDM9 defines the sites of meiotic recombination through a mechanism that increases structural variance and depends on aKRAB. The loss of aKRAB is likely to have shifted the locations of both recombination and structural variance hotspots. Our observations suggest that the TD evolved as a side-effect of these changes and link recombination to the evolution of intron size illustrating how genome architectures can evolve in the absence of selection.

Genome-wide atlas of loci involved in chromosomal recombination in common wheat

Chromosomal recombination is a pivotal biological mechanism for generating novel genetic diversity, essential for plant breeding and genetic research endeavors. This study investigated the genetic loci involved in chromosomal recombination through analyzing five published recombinant inbred line (RIL) populations and four double haploid (DH) populations. Great phenotypic variations in recombination frequency were observed between populations and chromosomes. A total of 29 QTL were mapped, which were predominantly located on the B genome. Notably, one QTL on chromosome 6AL was identified from two RIL populations and one QTL on chromosome 3B was identified in both RIL and DH populations. Additionally, a map delineating recombination hotspot regions was developed, and these regions were observed on all chromosomes except for 6B. Recombination hotspot regions tended to locate on chromosomes 1D, 3A, 3B, 6A, and 7D compared to chromosomes 1B, 5B, and 6B. In addition, most hotspot regions were located at chromosome termini, with some clustering in specific regions. Besides genetic factors, the study also explored the impact of chip size and population type on the number of identifiable chromosomal recombination events. Overall, this work improves our understanding of the molecular mechanisms for the chromosomal recombination rate and may contribute to the optimization of breeding strategies in wheat.

Data preprocessing methods for selective sweep detection using convolutional neural networks

The identification of positive selection has been framed as a classification task, with Convolutional Neural Networks (CNNs) already outperforming summary statistics and likelihood-based approaches in accuracy. Despite the prevalence of CNN-based methods that manipulate the pixels of images representing raw genomic data as a preprocessing step to improve classification accuracy, the efficacy of these pixel-rearrangement techniques remains inadequately examined, particularly in the presence of confounding factors like population bottlenecks, migration and recombination hotspots. We introduce a set of pixel rearrangement algorithms aimed at enhancing CNN classification accuracy in detecting selective sweeps. These algorithms are employed to assess the performance of four CNN models for selective sweep detection. Our findings illustrate that the judicious application of rearrangement algorithms notably enhances the overall classification accuracy of a CNN across various datasets simulating confounding factors. We observed that sorting the columns of the genomic matrices has higher on CNN performance than rearranging the sequences. To some extent, these rearrangement algorithms are more robust to misspecified demographic models compared with the utilization of the default preprocessing algorithm as suggested by the respective authors of each CNN architecture. We provide the data rearrangement algorithms as a distinct package available for download at: https://github.com/Zhaohq96/Genetic-data-rearrangement.

Characterization and distribution of de novo mutations in the zebra finch

Germline de novo mutations (DNMs) provide the raw material for evolution. The DNM rate varies considerably between species, sexes and chromosomes. Here, we identify DNMs in the zebra finch (Taeniopygia guttata) across 16 parent-offspring trios using two genome assemblies of different quality. Using an independent genotyping assay, we validate 82% of the 150 candidate DNMs. DNM rates are consistent between both assemblies, with estimates of 6.14 × 10–9 and 6.36 × 10–9 per site per generation. We observe a strong paternal bias in DNM rates (male-to-female ratio ɑ ≈ 4), but this bias is in transition mutations only, leading to a transition-to-transversion ratio of 3.18 and 3.57. Finally, we find that DNMs tend to be randomly distributed across chromosomes, not associated with recombination hotspots or genic regions. However, the sex chromosome chrZ shows a roughly fourfold increased DNM rate compared to autosomes, which is more than the expected increase due to chrZ spending two-thirds of its time in males. Overall, our results further enhance our understanding of DNMs in passerine songbirds.

Squash vein yellowing virus from California emerged in the Middle East via intragenic and intergeneric recombination events in the hypervariable potyvirus P1 and ipomovirus P1a genes.

We present the complete sequence of the genomic RNA of an isolate of squash vein yellowing virus (Ipomovirus cucurbitavenaflavi) from California (SqVYV-CA) and show it is a recombinant virus with a highly divergent 5' UTR and proximal P1a gene. The evolution of SqVYV-CA involved an intrageneric event between unknown potyviruses, related to isolates of papaya ringspot virus (Potyvirus papayanuli) from the Old World, and an intergeneric event between this recombinant potyvirus (minor parent) and an isolate of SqVYV from Israel (SqVYV-IL) (major parent). These events occurred in mixed infections and in the potyvirus P1 and ipomovirus P1a recombination hotspots and resulted in SqVYV-CA having a potyvirus 5' UTR and chimeric P1-P1a gene/protein and the remainder of the genome from SqVYV-IL. The SqVYV-CA chimeric P1-P1a gene is under positive selection, and the protein is intrinsically disordered and may localize to the nucleus via nuclear localization signals in the P1 part. The C-terminal SqVYV-IL P1a part also diverged but retained the conserved serine protease motif. Furthermore, substantial divergence in SqVYV isolates from the Middle East was associated with genetic drift and a long evolutionary history in this region. The finding that the host range and symptomatology in cucurbits of SqVYV-CA is similar to those of SqVYV from Florida and SqVYV-IL, indicated that the recombinant part of the genome had no obvious effect on the virus-host interaction. A divergent part of the P1 sequence of the SqVYV-CA P1-P1a gene was used to develop a primer pair and RT-PCR test for specific detection of SqVYV-CA. This test was used to detect spread of SqVYV-CA to a new production area of California in 2021 and 2022. Together, these results demonstrate (i) a high level of genetic diversity exists among isolates of SqVYV and involved intra- and intergeneric recombination and genetic drift (mutation), (ii) evidence that SqVYV originated in the Middle East and that there were independent introductions into the New World and (iii) the remarkable genetic flexibility of the 5' proximal genes of these viruses.

Feline bocaviruses found in Thailand have undergone genetic recombination for their evolutions

Feline bocaviruses (FBoVs) have been discovered for a decade and are often detected in feces, possibly associated with diarrhea in cats. Studies on FBoV evolution remain limited and have mainly focused on prevalence and genetic characterization. Although genetic recombination serves as a potential mechanism in bocavirus evolution, research on this process for FBoVs has been scarce. In this study, we characterized 19 complete coding sequences of FBoVs obtained from Thai cats, revealing that FBoV-1, -2, and -3 were endemic in Thailand. Genetic characterizations showed that most Thai FBoVs were closely related to previously detected strains in Thailand and China. Recombination analyses indicated intragenic, intraspecies recombination in all FBoV species, with recombination breakpoints commonly found in the NP1 and VP1/2 genes, highlighting these genes may be hotspots for FBoV recombination. However, no interspecies recombination was detected. Selective pressure analysis of various FBoV genes revealed that these viruses underwent purifying selection. Although the VP1/2 gene of all FBoV species was under the strongest negative selection pressure, positive selection sites were only found in FBoV-1 and FBoV-3. This study is the first to identify natural recombination in FBoV-2 and FBoV-3 and provides evidence that genetic recombination is a potential driver of FBoV evolutions. Additionally, this study offers up-to-date information on the genetic characteristics, evolutionary dynamics, and selective pressure status of FBoVs, which should be continuously monitored.

Deep longitudinal lower respiratory tract microbiome profiling reveals genome-resolved functional and evolutionary dynamics in critical illness

The lower respiratory tract (LRT) microbiome impacts human health, especially among critically ill patients. However, comprehensive characterizations of the LRT microbiome remain challenging due to low microbial mass and host contamination. We develop a chelex100-based low-biomass microbial-enrichment method (CMEM) that enables deep metagenomic profiling of LRT samples to recover near-complete microbial genomes. We apply the method to 453 longitudinal LRT samples from 157 intensive care unit (ICU) patients in three geographically distant hospitals. We recover 120 high-quality metagenome-assembled genomes (MAGs) and associated plasmids without culturing. We detect divergent longitudinal microbiome dynamics and hospital-specific dominant opportunistic pathogens and resistomes in pneumonia patients. Diagnosed pneumonia and the ICU stay duration were associated with the abundance of specific antibiotic-resistance genes (ARGs). Moreover, CMEM can serve as a robust tool for genome-resolved analyses. MAG-based analyses reveal strain-specific resistome and virulome among opportunistic pathogen strains. Evolutionary analyses discover increased mobilome in prevailing opportunistic pathogens, highly conserved plasmids, and new recombination hotspots associated with conjugative elements and prophages. Integrative analysis with epidemiological data reveals frequent putative inter-patient strain transmissions in ICUs. In summary, we present a genome-resolved functional, transmission, and evolutionary landscape of the LRT microbiota in critically ill patients.

Multiple independent losses of crossover interference during yeast evolutionary history.

Meiotic recombination is essential for the accurate chromosome segregation and the generation of genetic diversity through crossover and gene conversion events. Although this process has been studied extensively in a few selected model species, understanding how its properties vary across species remains limited. For instance, the ancestral ZMM pathway that generates interference-dependent crossovers has undergone multiple losses throughout evolution, suggesting variations in the regulation of crossover formation. In this context, we first characterized the meiotic recombination landscape and properties of the Kluyveromyces lactis budding yeast. We then conducted a comprehensive analysis of 29,151 recombination events (19, 212 COs and 9, 939 NCOs) spanning 577 meioses in the five budding yeast species Saccharomyces cerevisiae, Saccharomyces paradoxus, Lachancea kluyveri, Lachancea waltii and K. lactis. Eventually, we found that the Saccharomyces yeasts displayed higher recombination rates compared to the non-Saccharomyces yeasts. In addition, bona fide crossover interference and associated crossover homeostasis were detected in the Saccharomyces species only, adding L. kluyveri and K. lactis to the list of budding yeast species that lost crossover interference. Finally, recombination hotspots, although highly conserved within the Saccharomyces yeasts are not conserved beyond the Saccharomyces genus. Overall, these results highlight great variability in the recombination landscape and properties through budding yeasts evolution.

Identification and functional analysis of recent IS transposition events in rhizobia

Rhizobia are alpha- and beta- Proteobacteria that, through the establishment of symbiotic interactions with leguminous plants, are able to fix atmospheric nitrogen as ammonium. The successful establishment of a symbiotic interaction is highly dependent on the availability of nitrogen sources in the soil, and on the specific rhizobia strain. Insertion sequences (ISs) are simple transposable genetic elements that can move to different locations within the host genome and are known to play an important evolutionary role, contributing to genome plasticity by acting as recombination hot-spots, and disrupting coding and regulatory sequences. Disruption of coding sequences may have occurred either in a common ancestor of the species or more recently. By means of ISComapare, we identified Differentially Located ISs (DLISs) in nearly related rhizobial strains of the genera Bradyrhizobium, Mesorhizobium, Rhizobium and Sinorhizobium. Our results revealed that recent IS transposition could have a role in adaptation by enabling the activation and inactivation of genes that could dynamically affect the competition and survival of rhizobia in the rhizosphere.Graphical

Haplotype-Aware Detection of SERPINA1 Variants by Nanopore Sequencing

Alpha-1 antitrypsin (AAT) is an acute-phase reactant with immunomodulatory properties that mainly inhibits neutrophil elastase. Low serum levels cause AAT deficiency (AATD), an underdiagnosed condition that predisposes to pulmonary and hepatic diseases. The SERPINA1 gene, which encodes AAT, contains more than 500 variants. PI*Z and PI*S alleles are the most diagnosed causes of AATD, but the role of the SERPINA1 haplotypes in AAT function remains unknown. SERPINA1 gene was PCR amplified from 94 asthma patients, using primers with tails for indexing. Sequencing libraries were loaded into a MinION-Mk1C, and MinKNOW was used for basecalling and demultiplexing. Nanofilt and Minimap2 were used for filtering and mapping/alignment. Variant calling/phasing were performed with PEPPER-Margin-DeepVariant. SERPINA1 gene was 100% covered for all samples, with a minimum sequencing depth of 500X. 75 single nucleotide variants and 4 indels were detected, with 45 and 2 of them highly polymorphic (MAF>0.1), respectively. Nine of the SNVs showed differences in allele frequencies when compared with the overall Spanish population. More than 90% of heterozygous SNVs were phased, yielding 91 and 58 different haplotypes for each SERPINA1 amplified region. Haplotype-based Linkage Disequilibrium (LD) analysis suggests that a recombination hotspot could generate variation in the SERPINA1 gene. The proposed workflow enables haplotype-aware genotyping of the SERPINA1 gene by nanopore sequencing, which will allow the development of novel AATD diagnostic strategies.

Novel crossover and recombination hotspots massively spread across primate genomes

BackgroundThe recombination landscape and subsequent natural selection have vast consequences forevolution and speciation. However, most of the crossover and recombination hotspots are yet to be discovered. We previously reported the relevance of C and G trinucleotide two-repeat units (CG-TTUs) in crossovers and recombination.MethodsOn a genome-wide scale, here we mapped all combinations of A and T trinucleotide two-repeat units (AT-TTUs) in human, consisting of AATAAT, ATAATA, ATTATT, TTATTA, TATTAT, and TAATAA. We also compared a number of the colonies formed by the AT-TTUs (distance between consecutive AT-TTUs < 500 bp) in several other primates and mouse.ResultsWe found that the majority of the AT-TTUs (> 96%) resided in approximately 1.4 million colonies, spread throughout the human genome. In comparison to the CG-TTU colonies, the AT-TTU colonies were significantly more abundant and larger in size. Pure units and overlapping units of the pure units were readily detectable in the same colonies, signifying that the units were the sites of unequal crossover. We discovered dynamic sharedness of several of the colonies across the primate species studied, which mainly reached maximum complexity and size in human.ConclusionsWe report novel crossover and recombination hotspots of the finest molecular resolution, massively spread and shared across the genomes of human and several other primates. With respect to crossovers and recombination, these genomes are far more dynamic than previously envisioned.

Genomic Underpinnings of Cytoplasmic Incompatibility: CIF Gene-Neighborhood Diversification Through Extensive Lateral Transfers and Recombination in Wolbachia.

Cytoplasmic incompatibility (CI), a non-Mendelian genetic phenomenon, involves the manipulation of host reproduction by Wolbachia, a maternally transmitted alphaproteobacterium. The underlying mechanism is centered around the CI Factor (CIF) system governed by two genes, cifA and cifB, where cifB induces embryonic lethality, and cifA counteracts it. Recent investigations have unveiled intriguing facets of this system, including diverse cifB variants, prophage association in specific strains, copy number variation, and rapid component divergence, hinting at a complex evolutionary history. We utilized comparative genomics to systematically classify CIF systems, analyze their locus structure and domain architectures, and reconstruct their diversification and evolutionary trajectories. Our new classification identifies ten distinct CIF types, featuring not just versions present in Wolbachia, but also other intracellular bacteria, and eukaryotic hosts. Significantly, our analysis of CIF loci reveals remarkable variability in gene composition and organization, encompassing an array of diverse endonucleases, variable toxin domains, deubiquitinating peptidases (DUBs), prophages, and transposons. We present compelling evidence that the components within the loci have been diversifying their sequences and domain architectures through extensive, independent lateral transfers and interlocus recombination involving gene conversion. The association with diverse transposons and prophages, coupled with selective pressures from host immunity, likely underpins the emergence of CIF loci as recombination hotspots. Our investigation also posits the origin of CifB-REase domains from mobile elements akin to CR (Crinkler-RHS-type) effectors and Tribolium Medea1 factor, which is linked to another non-Mendelian genetic phenomenon. This comprehensive genomic analysis offers novel insights into the molecular evolution and genomic foundations of Wolbachia-mediated host reproductive control.

Deciphering the evolutionary landscape of severe fever with thrombocytopenia syndrome virus across East Asia.

Severe fever with thrombocytopenia syndrome virus (SFTSV) poses a significant public health challenge in East Asia, necessitating a deeper understanding of its evolutionary dynamics to effectively manage its spread and pathogenicity. This study provides a comprehensive analysis of the genetic diversity, recombination patterns, and selection pressures across the SFTSV genome, utilizing an extensive dataset of 2041 sequences from various hosts and regions up to November 2023. Employing maximum likelihood and Bayesian evolutionary analysis by sampling trees (BEAST), we elucidated the phylogenetic relationships among nine distinct SFTSV genotypes (A, B1, B2, B3, B4, C, D, E, and F), revealing intricate patterns of viral evolution and genotype distribution across China, South Korea, and Japan. Furthermore, our analysis identified 34 potential reassortments, underscoring a dynamic genetic interplay among SFTSV strains. Genetic recombination was observed most frequently in the large segment and least in the small segment, with notable recombination hotspots characterized by stem-loop hairpin structures, indicative of a structural propensity for genetic recombination. Additionally, selection pressure analysis on critical viral genes indicated a predominant trend of negative selection, with specific sites within the RNA-dependent RNA polymerase and glycoprotein genes showing positive selection. These sites suggest evolutionary adaptations to host immune responses and environmental pressures. This study sheds light on the intricate evolutionary mechanisms shaping SFTSV, offering insights into its adaptive strategies and potential implications for vaccine development and therapeutic interventions.

Recombination of Porcine Reproductive and Respiratory Syndrome Virus: Features, Possible Mechanisms, and Future Directions.

Recombination is a pervasive phenomenon in RNA viruses and an important strategy for accelerating the evolution of RNA virus populations. Recombination in the porcine reproductive and respiratory syndrome virus (PRRSV) was first reported in 1999, and many case reports have been published in recent years. In this review, all the existing reports on PRRSV recombination events were collected, and the genotypes, parental strains, and locations of the recombination breakpoints have been summarized and analyzed. The results showed that the recombination pattern constantly changes; whether inter- or intra-lineage recombination, the recombination hotspots vary in different recombination patterns. The virulence of recombinant PRRSVs was higher than that of the parental strains, and the emergence of virulence reversion was caused by recombination after using MLV vaccines. This could be attributed to the enhanced adaptability of recombinant PRRSV for entry and replication, facilitating their rapid propagation. The aim of this paper was to identify common features of recombinant PRRSV strains, reduce the recombination risk, and provide a foundation for future research into the mechanism of PRRSV recombination.

Homologous Recombination Shapes the Architecture and Evolution of Bacterial Genomes.

Homologous recombination is a key evolutionary force that varies considerably across bacterial species. However, how the landscape of homologous recombination varies across genes and within individual genomes has only been studied in a few species. Here, we used Approximate Bayesian Computation to estimate the recombination rate along the genomes of 145 bacterial species. Our results show that homologous recombination varies greatly along bacterial genomes and shapes many aspects of genome architecture and evolution. The genomic landscape of recombination presents several key signatures: rates are highest near the origin of replication in most species, patterns of recombination generally appear symmetrical in both replichores (i.e. replicational halves of circular chromosomes) and most species have genomic hotpots of recombination. Furthermore, many closely related species share conserved landscapes of recombination across orthologs indicating that recombination landscapes are conserved over significant evolutionary distances. We show evidence that recombination drives the evolution of GC-content through increasing the effectiveness of selection and not through biased gene conversion, thereby contributing to an ongoing debate. Finally, we demonstrate that the rate of recombination varies across gene function and that many hotspots of recombination are associated with adaptive and mobile regions often encoding genes involved in pathogenicity.

The Prediction of Recombination Hotspot Based on Automated Machine Learning

Meiotic recombination plays a pivotal role in genetic evolution. Genetic variation induced by recombination is a crucial factor in generating biodiversity and a driving force for evolution. At present, the development of recombination hotspot prediction methods has encountered challenges related to insufficient feature extraction and limited generalization capabilities. This paper focused on the research of recombination hotspot prediction methods. We explored deep learning-based recombination hotspot prediction and scrutinized the shortcomings of prevalent models in addressing the challenge of recombination hotspot prediction. To addressing these deficiencies, an automated machine learning approach was utilized to construct recombination hotspot prediction model. The model combined sequence information with physicochemical properties by employing TF-IDF-Kmer and DNA composition components to acquire more effective feature data. Experimental results validate the effectiveness of the feature extraction method and automated machine learning technology used in this study. The final model was validated on three distinct datasets and yielded accuracy rates of 97.14%, 79.71%, and 98.73%, surpassing the current leading models by 2%, 2.56%, and 4%, respectively. In addition, we incorporated tools such as SHAP and AutoGluon to analyze the interpretability of black-box models, delved into the impact of individual features on the results, and investigated the reasons behind misclassification of samples. Finally, an application of recombination hotspot prediction was established to facilitate easy access to necessary information and tools for researchers. The research outcomes of this paper underscore the enormous potential of automated machine learning methods in gene sequence prediction.

Corrigendum: The evolutionary turnover of recombination hot spots contributes to speciation in mice.

Corrigendum: The evolutionary turnover of recombination hot spots contributes to speciation in mice.