Machine-learning predicts genomic determinants of meiosis-driven structural variation in a eukaryotic pathogen

Recombination and transposition drive genomic structural variation potentially impacting life history traits in a host-generalist fungal plant pathogen

BACKGROUND

An understanding of plant pathogen evolution is important for sustainable management of crop diseases. Plant pathogen populations must maintain adequate heritable phenotypic variability to survive. Polymorphisms ≥ 50 bp, known as structural variants (SVs), could contribute strongly to this variability by disrupting gene activities. SV acquisition is largely driven by mobile genetic elements called transposons, though a less appreciated source of SVs is erroneous meiotic double-strand break repair. The relative impacts of transposons and recombination on SV diversity and the overall contribution of SVs to phenotypic variability is elusive, especially in host generalists.

RESULTS

We use 25 high-quality genomes to create a graphical pan-genome of the globally distributed host-generalist crop pathogen Sclerotinia sclerotiorum. Outcrossing and recombination rates in this self-fertile species have been debated. Using bisulfite sequencing and short-read data from 190 strains, we show that S. sclerotiorum has many hallmarks of eukaryotic meiosis, including recombination hot and cold spots, centromeric and genic recombination suppression, and rapid linkage disequilibrium decay. Using a new statistic that captures average pairwise structural variation, we show that recombination and transposons make distinct contributions to SV diversity. Furthermore, despite only 5% of genes being dispensable, SVs often had a stronger impact than other variants across 14 life history traits measured in 103 distinct strains.

CONCLUSIONS

Transposons and recombination make distinct contributions to SV diversity in S. sclerotiorum. Despite limited gene content diversity, SVs may strongly impact phenotypic variability. This sheds light on the genomic forces shaping adaptive flexibility in host generalists.

Development of an early prediction model for vomiting during hemodialysis using LASSO regression and Boruta feature selection

Hemodialysis patients (HD) frequently experience nausea and vomiting as side effects, which can make the procedure uncomfortable for them and cause it to end too soon. There are no known predictors of vomiting. We aim to create a nomogram that can anticipate vomiting in hemodialysis patients. We conducted a retrospective screening of patients with end-stage renal disease (ESRD) who received regular hemodialysis at the First People's Hospital of Nantong from January 1, 2023, to October 31, 2024. The outcome of the nomogram indicated vomiting, which was evaluated using the Korttila scale. The least absolute shrinkage selection operator (LASSO) method and Boruta feature selection were employed for the optimal prediction of predictors. Multiple logistic regression was employed to construct predictive models presented as nomograms. The efficacy of nomograms was evaluated using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA). The model underwent internal validation by assessing the validation cohort's performance. The study included 281 patients. Ninety-two patients, representing 32.74%, exhibited symptoms of vomiting. Participants were randomly assigned to training (n = 196) and validation (n = 85) groups. The nomogram incorporated predictors such as sex, height, heart rate, spKt.V, lymphocytes, and lactate dehydrogenase. The ROC curves for both the training and verification groups demonstrate strong recognition capability, while the calibration curves indicate that the correction outcomes for both groups are highly satisfactory. This nomogram assists clinicians in identifying high-risk populations and supports the formulation of effective preventive strategies.

Copy number variation introduced by a massive mobile element facilitates global thermal adaptation in a fungal wheat pathogen

Copy number variation (CNV) can drive rapid evolution in changing environments. In microbial pathogens, such adaptation is a key factor underpinning epidemics and colonization of new niches. However, the genomic determinants of such adaptation remain poorly understood. Here, we systematically investigate CNVs in a large genome sequencing dataset spanning a worldwide collection of 1104 genomes from the major wheat pathogen Zymoseptoria tritici. We found overall strong purifying selection acting on most CNVs. Genomic defense mechanisms likely accelerated gene loss over episodes of continental colonization. Local adaptation along climatic gradients was likely facilitated by CNVs affecting secondary metabolite production and gene loss in general. One of the strongest loci for climatic adaptation is a highly conserved gene of the NAD-dependent Sirtuin family. The Sirtuin CNV locus localizes to an ~68-kb Starship mobile element unique to the species carrying genes highly expressed during plant infection. The element has likely lost the ability to transpose, demonstrating how the ongoing domestication of cargo-carrying selfish elements can contribute to selectable variation within populations. Our work highlights how standing variation in gene copy numbers at the global scale can be a major factor driving climatic and metabolic adaptation in microbial species.

The landscape and predicted roles of structural variants in Fusarium graminearum genomes

Structural rearrangements, such as inversions, translocations, duplications, and large insertions and deletions, are large-scale genomic variants that can play an important role in shaping phenotypic variation and in genome adaptation and evolution. We used chromosomal-level assemblies from eight Fusarium graminearum isolates to study structural variants and their role in fungal evolution. We generated the assemblies of four of these genomes after Oxford Nanopore sequencing. A total of 87 inversions, 159 translocations, 245 duplications, 58,489 insertions, and 34,102 deletions were detected. Regions of high recombination rate are associated with structural rearrangements, and a significant proportion of inversions, translocations, and duplications overlap with the repeat content of the genome, suggesting recombination and repeat elements are major factors in the origin of structural rearrangements in F. graminearum. Large insertions and deletions introduce presence-absence polymorphisms for many genes, including secondary metabolite biosynthesis cluster genes and predicted effectors genes. Translocation events were found to be shuffling predicted effector-rich regions of the genomes and are likely contributing to the gain and loss of effectors facilitated by recombination. Breakpoints of some structural rearrangements fall within coding sequences and are likely altering the protein products. Structural rearrangements in F. graminearum thus have an important role to play in shaping pathogen-host interactions and broader evolution through genome reorganization, the introduction of presence-absence polymorphisms, and changing protein products and gene regulation.

Distinct genomic contexts predict gene presence-absence variation in different pathotypes of Magnaporthe oryzae

Fungi use the accessory gene content of their pangenomes to adapt to their environments. While gene presence-absence variation contributes to shaping accessory gene reservoirs, the genomic contexts that shape these events remain unclear. Since pangenome studies are typically species-wide and do not analyze different populations separately, it is yet to be uncovered whether presence-absence variation patterns and mechanisms are consistent across populations. Fungal plant pathogens are useful models for studying presence-absence variation because they rely on it to adapt to their hosts, and members of a species often infect distinct hosts. We analyzed gene presence-absence variation in the blast fungus, Magnaporthe oryzae (syn. Pyricularia oryzae), and found that presence-absence variation genes involved in host-pathogen and microbe-microbe interactions may drive the adaptation of the fungus to its environment. We then analyzed genomic and epigenomic features of presence-absence variation and observed that proximity to transposable elements, gene GC content, gene length, expression level in the host, and histone H3K27me3 marks were different between presence-absence variation genes and conserved genes. We used these features to construct a model that was able to predict whether a gene is likely to experience presence-absence variation with high precision (86.06%) and recall (92.88%) in M. oryzae. Finally, we found that presence-absence variation genes in the rice and wheat pathotypes of M. oryzae differed in their number and their genomic context. Our results suggest that genomic and epigenomic features of gene presence-absence variation can be used to better understand and predict fungal pangenome evolution. We also show that substantial intra-species variation can exist in these features.

Recent reactivation of a pathogenicity-associated transposable element is associated with major chromosomal rearrangements in a fungal wheat pathogen

Transposable elements (TEs) are key drivers of genomic variation contributing to recent adaptation in most species. Yet, the evolutionary origins and insertion dynamics within species remain poorly understood. We recapitulate the spread of the pathogenicity-associated Styx element across five species that last diverged ∼11 000 years ago. We show that the element likely originated in the Zymoseptoria fungal pathogen genus and underwent multiple independent reactivation events. Using a global 900-genome panel of the wheat pathogen Zymoseptoria tritici, we assess Styx copy number variation and identify renewed transposition activity in Oceania and South America. We show that the element can mobilize to create additional Styx copies in a four-generation pedigree. Importantly, we find that new copies of the element are not affected by genomic defenses suggesting minimal control against the element. Styx copies are preferentially located in recombination breakpoints and likely triggered multiple types of large chromosomal rearrangements. Taken together, we establish the origin, diversification and reactivation of a highly active TE with likely major consequences for chromosomal integrity and the expression of disease.

Combined reference-free and multi-reference based GWAS uncover cryptic variation underlying rapid adaptation in a fungal plant pathogen

Microbial pathogens often harbor substantial functional diversity driven by structural genetic variation. Rapid adaptation from such standing variation threatens global food security and human health. Genome-wide association studies (GWAS) provide a powerful approach to identify genetic variants underlying recent pathogen adaptation. However, the reliance on single reference genomes and single nucleotide polymorphisms (SNPs) obscures the true extent of adaptive genetic variation. Here, we show quantitatively how a combination of multiple reference genomes and reference-free approaches captures substantially more relevant genetic variation compared to single reference mapping. We performed reference-genome based association mapping across 19 reference-quality genomes covering the diversity of the species. We contrasted the results with a reference-free (i.e., k-mer) approach using raw whole-genome sequencing data in a panel of 145 strains collected across the global distribution range of the fungal wheat pathogen Zymoseptoria tritici. We mapped the genetic architecture of 49 life history traits including virulence, reproduction and growth in multiple stressful environments. The inclusion of additional reference genome SNP datasets provides a nearly linear increase in additional loci mapped through GWAS. Variants detected through the k-mer approach explained a higher proportion of phenotypic variation than a reference genome-based approach and revealed functionally confirmed loci that classic GWAS approaches failed to map. The power of GWAS in microbial pathogens can be significantly enhanced by comprehensively capturing structural genetic variation. Our approach is generalizable to a large number of species and will uncover novel mechanisms driving rapid adaptation of pathogens.

Reliable genotyping of recombinant genomes using a robust hidden Markov model

Meiotic recombination is an essential mechanism during sexual reproduction and includes the exchange of chromosome segments between homologous chromosomes. New allelic combinations are transmitted to the new generation, introducing novel genetic variation in the offspring genomes. With the improvement of high-throughput whole-genome sequencing technologies, large numbers of recombinant individuals can now be sequenced with low sequencing depth at low costs, necessitating computational methods for reconstructing their haplotypes. The main challenge is the uncertainty in haplotype calling that arises from the low information content of a single genomic position. Straightforward sliding window-based approaches are difficult to tune and fail to place recombination breakpoints precisely. Hidden Markov model (HMM)-based approaches, on the other hand, tend to over-segment the genome. Here, we present RTIGER, an HMM-based model that exploits in a mathematically precise way the fact that true chromosome segments typically have a certain minimum length. We further separate the task of identifying the correct haplotype sequence from the accurate placement of haplotype borders, thereby maximizing the accuracy of border positions. By comparing segmentations based on simulated data with known underlying haplotypes, we highlight the reasons for RTIGER outperforming traditional segmentation approaches. We then analyze the meiotic recombination pattern of segregants of 2 Arabidopsis (Arabidopsis thaliana) accessions and a previously described hyper-recombining mutant. RTIGER is available as an R package with an efficient Julia implementation of the core algorithm.

Development and assessment of novel machine learning models to predict the probability of postoperative nausea and vomiting for patient-controlled analgesia

Postoperative nausea and vomiting (PONV) can lead to various postoperative complications. The risk assessment model of PONV is helpful in guiding treatment and reducing the incidence of PONV, whereas the published models of PONV do not have a high accuracy rate. This study aimed to collect data from patients in Sichuan Provincial People's Hospital to develop models for predicting PONV based on machine learning algorithms, and to evaluate the predictive performance of the models using the area under the receiver characteristic curve (AUC), accuracy, precision, recall rate, F1 value and area under the precision-recall curve (AUPRC). The AUC (0.947) of our best machine learning model was significantly higher than that of the past models. The best of these models was used for external validation on patients from Chengdu First People's Hospital, and the AUC was 0.821. The contributions of variables were also interpreted using SHapley Additive ExPlanation (SHAP). A history of motion sickness and/or PONV, sex, weight, history of surgery, infusion volume, intraoperative urine volume, age, BMI, height, and PCA_3.0 were the top ten most important variables for the model. The machine learning models of PONV provided a good preoperative prediction of PONV for intravenous patient-controlled analgesia.

Distinct genomic contexts predict gene presence-absence variation in different pathotypes of a fungal plant pathogen

Background

Fungi use the accessory segments of their pan-genomes to adapt to their environments. While gene presence-absence variation (PAV) contributes to shaping these accessory gene reservoirs, whether these events happen in specific genomic contexts remains unclear. Additionally, since pan-genome studies often group together all members of the same species, it is uncertain whether genomic or epigenomic features shaping pan-genome evolution are consistent across populations within the same species. Fungal plant pathogens are useful models for answering these questions because members of the same species often infect distinct hosts, and they frequently rely on gene PAV to adapt to these hosts.

Results

We analyzed gene PAV in the rice and wheat blast fungus, Magnaporthe oryzae, and found that PAV of disease-causing effectors, antibiotic production, and non-self-recognition genes may drive the adaptation of the fungus to its environment. We then analyzed genomic and epigenomic features and data from available datasets for patterns that might help explain these PAV events. We observed that proximity to transposable elements (TEs), gene GC content, gene length, expression level in the host, and histone H3K27me3 marks were different between PAV genes and conserved genes, among other features. We used these features to construct a random forest classifier that was able to predict whether a gene is likely to experience PAV with high precision (86.06%) and recall (92.88%) in rice-infecting M. oryzae. Finally, we found that PAV in wheat- and rice-infecting pathotypes of M. oryzae differed in their number and their genomic context.

Conclusions

Our results suggest that genomic and epigenomic features of gene PAV can be used to better understand and even predict fungal pan-genome evolution. We also show that substantial intra-species variation can exist in these features.

Fungal Strains with Identical Genomes Were Found at a Distance of 2000 Kilometers after 40 Years

Heredity and variation are inherent characteristics of species and are mainly reflected in the stability and variation of the genome; the former is relative, while the latter is continuous. However, whether life has both stable genomes and extremely diverse genomes at the same time is unknown. In this study, we isolated Sclerotinia sclerotiorum strains from sclerotium samples in Quincy, Washington State, USA, and found that four single-sclerotium-isolation strains (PB4, PB273, PB615, and PB623) had almost identical genomes to the reference strain 1980 isolated in the west of Nebraska 40 years ago. The genome of strain PB4 sequenced by the next-generation sequencing (NGS) and Pacific Biosciences (PacBio) sequencing carried only 135 single nucleotide polymorphisms (SNPs) and 18 structural variations (SVs) compared with the genome of strain 1980 and 48 SNPs were distributed on Contig_20. Based on data generated by NGS, three other strains, PB273, PB615, and PB623, had 256, 275, and 262 SNPs, respectively, against strain 1980, which were much less than in strain PB4 (532 SNPs) and none of them occurred on Contig_20, suggesting much closer genomes to strain 1980 than to strain PB4. All other strains from America and China are rich in SNPs with a range of 34,391-77,618 when compared with strain 1980. We also found that there were 39-79 SNPs between strain PB4 and its sexual offspring, 53.1% of which also occurred on Contig_20. Our discoveries show that there are two types of genomes in S. sclerotiorum, one is very stable and the other tends to change constantly. Investigating the mechanism of such genome stability will enhance our understanding of heredity and variation.

A machine learning-based risk warning platform for potentially inappropriate prescriptions for elderly patients with cardiovascular disease

Potentially inappropriate prescribing (PIP), including potentially inappropriate medications (PIMs) and potential prescribing omissions (PPOs), is a major risk factor for adverse drug reactions (ADRs). Establishing a risk warning model for PIP to screen high-risk patients and implementing targeted interventions would significantly reduce the occurrence of PIP and adverse drug events. Elderly patients with cardiovascular disease hospitalized at the Sichuan Provincial People’s Hospital were included in the study. Information about PIP, PIM, and PPO was obtained by reviewing patient prescriptions according to the STOPP/START criteria (2nd edition). Data were divided into a training set and test set at a ratio of 8:2. Five sampling methods, three feature screening methods, and eighteen machine learning algorithms were used to handle data and establish risk warning models. A 10-fold cross-validation method was employed for internal validation in the training set, and the bootstrap method was used for external validation in the test set. The performances were assessed by area under the receiver operating characteristic curve (AUC), and the risk warning platform was developed based on the best models. The contributions of features were interpreted using SHapley Additive ExPlanation (SHAP). A total of 404 patients were included in the study (318 [78.7%] with PIP; 112 [27.7%] with PIM; and 273 [67.6%] with PPO). After data sampling and feature selection, 15 datasets were obtained and 270 risk warning models were built based on them to predict PIP, PPO, and PIM, respectively. External validation showed that the AUCs of the best model for PIP, PPO, and PIM were 0.8341, 0.7007, and 0.7061, respectively. The results suggested that angina, number of medications, number of diseases, and age were the key factors in the PIP risk warning model. The risk warning platform was established to predict PIP, PIM, and PPO, which has acceptable accuracy, prediction performance, and potential clinical application perspective.

Telomere-to-Telomere Genome Sequences across a Single Genus Reveal Highly Variable Chromosome Rearrangement Rates but Absolute Stasis of Chromosome Number

Genome rearrangements in filamentous fungi are prevalent but little is known about the modalities of their evolution, in part because few complete genomes are available within a single genus. To address this, we have generated and compared 15 complete telomere-to-telomere genomes across the phylogeny of a single genus of filamentous fungi, Epichloë. We find that the striking distinction between gene-rich and repeat-rich regions previously reported for isolated species is ubiquitous across the Epichloë genus. We built a species phylogeny from single-copy gene orthologs to provide a comparative framing to study chromosome composition and structural change through evolutionary time. All Epichloë genomes have exactly seven nuclear chromosomes, but despite this conserved ploidy, analyses reveal low synteny and substantial rearrangement of gene content across the genus. These rearrangements are highly lineage-dependent, with most occurring over short evolutionary distances, with long periods of structural stasis. Quantification of chromosomal rearrangements shows they are uncorrelated with numbers of substitutions and evolutionary distances, suggesting that different modes of evolution are acting to create nucleotide and chromosome-scale changes.

Giant Starship Elements Mobilize Accessory Genes in Fungal Genomes

Accessory genes are variably present among members of a species and are a reservoir of adaptive functions. In bacteria, differences in gene distributions among individuals largely result from mobile elements that acquire and disperse accessory genes as cargo. In contrast, the impact of cargo-carrying elements on eukaryotic evolution remains largely unknown. Here, we show that variation in genome content within multiple fungal species is facilitated by Starships, a newly discovered group of massive mobile elements that are 110 kb long on average, share conserved components, and carry diverse arrays of accessory genes. We identified hundreds of Starship-like regions across every major class of filamentous Ascomycetes, including 28 distinct Starships that range from 27 to 393 kb and last shared a common ancestor ca. 400 Ma. Using new long-read assemblies of the plant pathogen Macrophomina phaseolina, we characterize four additional Starships whose activities contribute to standing variation in genome structure and content. One of these elements, Voyager, inserts into 5S rDNA and contains a candidate virulence factor whose increasing copy number has contrasting associations with pathogenic and saprophytic growth, suggesting Voyager's activity underlies an ecological trade-off. We propose that Starships are eukaryotic analogs of bacterial integrative and conjugative elements based on parallels between their conserved components and may therefore represent the first dedicated agents of active gene transfer in eukaryotes. Our results suggest that Starships have shaped the content and structure of fungal genomes for millions of years and reveal a new concerted route for evolution throughout an entire eukaryotic phylum.

Using structural variants to understand the ecological and evolutionary dynamics of fungal plant pathogens

Deletions, duplications, insertions, inversions and translocations are commonly referred to as structural variants (SVs). Fungal plant pathogens have compact genomes, facilitating the generation of accurate maps of SVs for these species in recent studies. Structural variants have been found to constitute a significant proportion of the standing genetic variation in fungal plant pathogen populations, potentially leading to the generation of accessory genes, regions or chromosomes enriched in pathogenicity factors. Structural variants are involved in the rapid adaptation and ecological traits of pathogens, including host specialization and mating. Long-read sequencing techniques coupled with theoretical and experimental approaches have considerable potential for elucidating the phenotypic effects of SVs and deciphering the evolutionary and genomic mechanisms underlying the formation of SVs in fungal plant pathogens.

A machine learning-based risk warning platform for potentially inappropriate prescriptions for elderly patients with cardiovascular disease

Potentially inappropriate prescribing (PIP), including potentially inappropriate medications (PIMs) and potential prescribing omissions (PPOs), is a major risk factor for adverse drug reactions (ADRs). Establishing a risk warning model for PIP to screen high-risk patients and implementing targeted interventions would significantly reduce the occurrence of PIP and adverse drug events. Elderly patients with cardiovascular disease hospitalized at the Sichuan Provincial People's Hospital were included in the study. Information about PIP, PIM, and PPO was obtained by reviewing patient prescriptions according to the STOPP/START criteria (2nd edition). Data were divided into a training set and test set at a ratio of 8:2. Five sampling methods, three feature screening methods, and eighteen machine learning algorithms were used to handle data and establish risk warning models. A 10-fold cross-validation method was employed for internal validation in the training set, and the bootstrap method was used for external validation in the test set. The performances were assessed by area under the receiver operating characteristic curve (AUC), and the risk warning platform was developed based on the best models. The contributions of features were interpreted using SHapley Additive ExPlanation (SHAP). A total of 404 patients were included in the study (318 [78.7%] with PIP; 112 [27.7%] with PIM; and 273 [67.6%] with PPO). After data sampling and feature selection, 15 datasets were obtained and 270 risk warning models were built based on them to predict PIP, PPO, and PIM, respectively. External validation showed that the AUCs of the best model for PIP, PPO, and PIM were 0.8341, 0.7007, and 0.7061, respectively. The results suggested that angina, number of medications, number of diseases, and age were the key factors in the PIP risk warning model. The risk warning platform was established to predict PIP, PIM, and PPO, which has acceptable accuracy, prediction performance, and potential clinical application perspective.

Domestication shapes recombination patterns in tomato

Meiotic recombination is a biological process of key importance in breeding, to generate genetic diversity and develop novel or agronomically relevant haplotypes. In crop tomato, recombination is curtailed as manifested by linkage disequilibrium decay over a longer distance and reduced diversity compared with wild relatives. Here, we compared domesticated and wild populations of tomato and found an overall conserved recombination landscape, with local changes in effective recombination rate in specific genomic regions. We also studied the dynamics of recombination hotspots resulting from domestication and found that loss of such hotspots is associated with selective sweeps, most notably in the pericentromeric heterochromatin. We detected footprints of genetic changes and structural variants, among them associated with transposable elements, linked with hotspot divergence during domestication, likely causing fine-scale alterations to recombination patterns and resulting in linkage drag.

A population-level invasion by transposable elements triggers genome expansion in a fungal pathogen

Genome evolution is driven by the activity of transposable elements (TEs). The spread of TEs can have deleterious effects including the destabilization of genome integrity and expansions. However, the precise triggers of genome expansions remain poorly understood because genome size evolution is typically investigated only among deeply divergent lineages. Here, we use a large population genomics dataset of 284 individuals from populations across the globe of Zymoseptoria tritici, a major fungal wheat pathogen. We built a robust map of genome-wide TE insertions and deletions to track a total of 2456 polymorphic loci within the species. We show that purifying selection substantially depressed TE frequencies in most populations, but some rare TEs have recently risen in frequency and likely confer benefits. We found that specific TE families have undergone a substantial genome-wide expansion from the pathogen's center of origin to more recently founded populations. The most dramatic increase in TE insertions occurred between a pair of North American populations collected in the same field at an interval of 25 years. We find that both genome-wide counts of TE insertions and genome size have increased with colonization bottlenecks. Hence, the demographic history likely played a major role in shaping genome evolution within the species. We show that both the activation of specific TEs and relaxed purifying selection underpin this incipient expansion of the genome. Our study establishes a model to recapitulate TE-driven genome evolution over deeper evolutionary timescales.

A devil's bargain with transposable elements in plant pathogens

Transposable elements (TEs) spread in genomes through self-copying mechanisms and are a major cause of genome expansions. Plant pathogens have finely tuned the expression of virulence factors to rely on epigenetic control targeted at nearby TEs. Stress experienced during the plant infection process leads to derepression of TEs and concurrently allows the expression of virulence factors. We argue that the derepression of TEs elements causes an evolutionary conflict by favoring TEs that can be reactivated. Active TEs and recent genome size expansions indicate that plant pathogens could face long-term consequences from the short-term benefit of fine-tuning the infection process. Hence, encoding key virulence factors close to TEs under epigenetic control constitutes a devil's bargain for pathogens.