Renaud Vitalis’s research while affiliated with Université de Montpellier and other places

Introduced over seventy years ago, F-statistics have been and remain central to population and evolutionary genetics. Among them, F ST is one of the most commonly used descriptive statistics in empirical studies, notably to characterize the structure of genetic polymorphisms within and between populations, to shed light on the evolutionary history of populations, or to identify marker loci under differential selection for adaptive traits. However, the use of F ST in simplified population models can overlook important hierarchical structures, such as geographic or temporal subdivisions, potentially leading to misleading interpretations and increasing false positives in genome scans for adaptive differentiation. Hierarchical F-statistics have been introduced to account for multiple pre-defined levels of population structure. Several estimators have also been proposed, including robust ones implemented in the popular R package hierfstat . Nevertheless, these were primarily designed for individual genotyping data and can be computationally intensive for large genomic datasets. In this study, we extend previous work by developing unbiased method-of-moments estimators for hierarchical F-statistics tailored for Pool-Seq data, a cost-effective alternative to individual genome sequencing. These Pool--Seq estimators have been developed in an Anova framework, using definitions based on identity-in-state probabilities. The new estimators have been implemented in an updated version of the R package poolfstat , together with estimators for sample allele count data derived from individual genotyping data. We validate and compare the performance of these estimators through extensive simulations under a hierarchical island model. Finally, we apply these estimators to real Pool-Seq data from Drosophila melanogaster populations, demonstrating their usefulness in revealing population structure and identifying loci with high differentiation within or between groups of subpopulations and associated with spatial or temporal genetic variation.

Large scale genomic resources can place genetic variation into an ecologically informed context. To advance our understanding of the population genetics of the fruit fly Drosophila melanogaster , we present an expanded release of the community-generated population genomics resource Drosophila Evolution over Space and Time (DEST 2.0; https://dest.bio/). This release includes 530 high-quality pooled libraries from flies collected across six continents over more than a decade (2009-2021), most at multiple time points per year; 211 of these libraries are sequenced and shared here for the first time. We used this enhanced resource to elucidate several aspects of the species' demographic history and identify novel signs of adaptation across spatial and temporal dimensions. We showed that patterns of secondary contact, originally characterized in North America, are replicated in South America and Australia. We also found that the spatial genetic structure of populations is stable over time, but that drift due to seasonal contractions of population size causes populations to diverge over time. We identified signals of adaptation that vary between continents in genomic regions associated with xenobiotic resistance, consistent with independent adaptation to common pesticides. Moreover, by analyzing samples collected during spring and fall across Europe, we provide new evidence for seasonal adaptation related to loci associated with pathogen response. Furthermore, we have also released an updated version of the DEST genome browser. This is a useful tool for studying spatio-temporal patterns of genetic variation in this classic model system.

The reconstruction of geographic and demographic scenarios of dissemination for invasive pathogens of crops is a key step toward improving the management of emerging infectious diseases. Nowadays, the reconstruction of biological invasions typically uses the information of both genetic and historical information to test for different hypotheses of colonization. The Approximate Bayesian Computation framework and its recent Random Forest development (ABC-RF) have been successfully used in evolutionary biology to decipher multiple histories of biological invasions. Yet, for some organisms, typically plant pathogens, historical data may not be reliable notably because of the difficulty to identify the organism and the delay between the introduction and the first mention. We investigated the history of the invasion of Africa by the fungal pathogen of banana Pseudocercospora fijiensis, by testing the historical hypothesis against other plausible hypotheses. We analyzed the genetic structure of eight populations from six eastern and western African countries, using 20 microsatellite markers and tested competing scenarios of population foundation using the ABC-RF methodology. We do find evidence for an invasion front consistent with the historical hypothesis, but also for the existence of another front never mentioned in historical records. We question the historical introduction point of the disease on the continent. Crucially, our results illustrate that even if ABC-RF inferences may sometimes fail to infer a single, well-supported scenario of invasion, they can be helpful in rejecting unlikely scenarios, which can prove much useful to shed light on disease dissemination routes.

Multipartite viruses have a segmented genome, with each segment encapsidated separately. In all multipartite virus species for which the question has been addressed, the distinct segments reproducibly accumulate at a specific and host-dependent relative frequency, defined as the 'genome formula'. Here, we test the hypothesis that the multipartite genome organization facilitates the regulation of gene expression via changes of the genome formula and thus via gene copy number variations. In a first experiment, the faba bean necrotic stunt virus (FBNSV), whose genome is composed of eight DNA segments each encoding a single gene, was inoculated into faba bean or alfalfa host plants, and the relative concentrations of the DNA segments and their corresponding messenger RNAs (mRNAs) were monitored. In each of the two host species, our analysis consistently showed that the genome formula variations modulate gene expression, the concentration of each genome segment linearly and positively correlating to that of its cognate mRNA but not of the others. In a second experiment, twenty parallel FBNSV lines were transferred from faba bean to alfalfa plants. Upon host switching, the transcription rate of some genome segments changes, but the genome formula is modified in a way that compensates for these changes and maintains a similar ratio between the various viral mRNAs. Interestingly, a deep-sequencing analysis of these twenty FBNSV lineages demonstrated that the host-related genome formula shift operates independently of DNA-segment sequence mutation. Together, our results indicate that nanoviruses are plastic genetic systems, able to transiently adjust gene expression at the population level in changing environments, by modulating the copy number but not the sequence of each of their genes.

Abstract Resurrection studies are a useful tool to measure how phenotypic traits have changed in populations through time. If these trait modifications correlate with the environmental changes that occurred during the time period, it suggests that the phenotypic changes could be a response to selection. Selfing, through its reduction of effective size, could challenge the ability of a population to adapt to environmental changes. Here, we used a resurrection study to test for adaptation in a selfing population of Medicago truncatula, by comparing the genetic composition and flowering times across 22 generations. We found evidence for evolution toward earlier flowering times by about two days and a peculiar genetic structure, typical of highly selfing populations, where some multilocus genotypes (MLGs) are persistent through time. We used the change in frequency of the MLGs through time as a multilocus fitness measure and built a selection gradient that suggests evolution toward earlier flowering times. Yet, a simulation model revealed that the observed change in flowering time could be explained by drift alone, provided the effective size of the population is small enough (

Tracking genetic changes of populations through time allows a more direct study of the evolutionary processes acting on the population than a single contemporary sample. Several statistical methods have been developed to characterize the demography and selection from temporal population genetic data. However, these methods are usually developed under the assumption of outcrossing reproduction and might not be applicable when there is substantial selfing in the population. Here, we focus on a method to detect loci under selection based on a genome scan of temporal differentiation, adapting it to the particularities of selfing populations. Selfing reduces the effective recombination rate and can extend hitch-hiking effects to the whole genome, erasing any local signal of selection on a genome scan. Therefore, selfing is expected to reduce the power of the test. By means of simulations, we evaluate the performance of the method under scenarios of adaptation from new mutations or standing variation at different rates of selfing. We find that the detection of loci under selection in predominantly selfing populations remains challenging even with the adapted method. Still, selective sweeps from standing variation on predominantly selfing populations can leave some signal of selection around the selected site thanks to historical recombination before the sweep. Under this scenario, ancestral advantageous alleles at low frequency leave the strongest local signal, while new advantageous mutations leave no local footprint of the sweep.

By capturing various patterns of the structuring of genetic variation across populations, f ‐statistics have proved highly effective for the inference of demographic history. Such statistics are defined as covariance of SNP allele frequency differences among sets of populations without requiring haplotype information and are hence particularly relevant for the analysis of pooled sequencing (Pool‐Seq) data. We here propose a reinterpretation of the F (and D) parameters in terms of probability of gene identity and derive from this unified definition unbiased estimators for both Pool‐Seq data and standard allele count data obtained from individual genotypes. We implemented these estimators in a new version of the R package poolfstat, which now includes a wide range of inference methods: (i) three‐population test of admixture; (ii) four‐population test of treeness; (iii) F4‐ratio estimation of admixture rates; and (iv) fitting, visualization and (semi‐automatic) construction of admixture graphs. A comprehensive evaluation of the methods implemented in poolfstat on both simulated Pool‐Seq (with various sequencing coverages and error rates) and allele count data confirmed the accuracy of these approaches, even for the most cost‐effective Pool‐Seq design involving relatively low sequencing coverages. We further analyzed a real Pool‐Seq data made of 14 populations of the invasive species Drosophila suzukii which allowed refining both the demographic history of native populations and the invasion routes followed by this emblematic pest. Our new package poolfstat provides the community with a user‐friendly and efficient all‐in‐one tool to unravel complex population genetic histories from large‐size Pool‐Seq or allele count SNP data.

By capturing various patterns of the structuring of genetic variation across populations, f -statistics have proved highly effective for the inference of demographic history. Such statistics are defined as covariance of SNP allele frequency differences among sets of populations without requiring haplotype information and are hence particularly relevant for the analysis of pooled sequencing (Pool-Seq) data. We here propose a reinterpretation of the F (and D ) parameters in terms of probability of gene identity and derive from this unified definition unbiased estimators for both Pool-Seq and standard allele count data obtained from individual genotypes. We implemented these estimators in a new version of the R package poolfstat , which now includes a wide range of inference methods: (i) three-population test of admixture; (ii) four-population test of treeness; (iii) F 4 -ratio estimation of admixture rates; and (iv) fitting, visualization and (semi-automatic) construction of admixture graphs. A comprehensive evaluation of the methods implemented in poolfstat on both simulated Pool-Seq (with various sequencing coverages and error rates) and allele count data confirmed the accuracy of these approaches, even for the most cost-effective Pool-Seq design involving low sequencing coverages. We further analyzed a real Pool-Seq data made of 14 populations of the invasive species Drosophila suzukii which allowed refining both the demographic history of native populations and the invasion routes followed by this emblematic pest. Our new package poolfstat provides the community with a user-friendly and efficient all-in-one tool to unravel complex population genetic histories from large-size Pool-Seq or allele count SNP data.

Simulation‐based methods such as Approximate Bayesian Computation (ABC) are well‐adapted to the analysis of complex scenarios of populations and species genetic history. In this context, supervised machine learning (SML) methods provide attractive statistical solutions to conduct efficient inferences about scenario choice and parameter estimation. The Random Forest methodology (RF) is a powerful ensemble of SML algorithms used for classification or regression problems. RF allows conducting inferences at a low computational cost, without preliminary selection of the relevant components of the ABC summary statistics, and bypassing the derivation of ABC tolerance levels. We have implemented a set of RF algorithms to process inferences using simulated datasets generated from an extended version of the population genetic simulator implemented in DIYABC v2.1.0. The resulting computer package, named DIYABC Random Forest v1.0, integrates two functionalities into a user‐friendly interface: the simulation under custom evolutionary scenarios of different types of molecular data (microsatellites, DNA sequences or SNPs) and RF treatments including statistical tools to evaluate the power and accuracy of inferences. We illustrate the functionalities of DIYABC Random Forest v1.0 for both scenario choice and parameter estimation through the analysis of pseudo‐observed and real datasets corresponding to pool‐sequencing and individual‐sequencing SNP datasets. Because of the properties inherent to the implemented RF methods and the large feature vector (including various summary statistics and their linear combinations) available for SNP data, DIYABC Random Forest v1.0 can efficiently contribute to the analysis of large SNP datasets to make inferences about complex population genetic histories.

Resurrection studies are a useful tool to measure how phenotypic traits have changed in populations and they allow testing whether these traits modifications are a response to selection caused by an environmental change. Selfing, through its reduction of effective size, could challenge the ability of a population to adapt to environmental changes. Here, we used a resurrection study to test for adaptation in a selfing population of Medicago truncatula , by comparing the genetic composition and flowering across 22 generations. We found evidence for evolution towards earlier flowering times by about two days and a peculiar genetic structure, typical for highly selfing population, where some multilocus genotypes (MLGs) are persistent through time. We used the change in frequency of the MLGs through time as a multilocus fitness measure and built a selection gradient that suggests evolution towards earlier flowering times. Yet, a simulation model revealed that the observed change in flowering time could be explained by drift alone, provided the effective size of the population is small enough (<150). These analyses suffer from the difficulty to estimate the effective size in a highly selfing population, where effective recombination is severely reduced.