Search

International audience; The academic journal publishing model is deeply unethical: today, a few major, for-profit conglomerates control more than 50% of all articles in the natural sciences and social sciences, driving subscription and open-access publishing fees above levels that can be sustainably maintained by publicly funded universities, libraries, and research institutions worldwide. About a third of the costs paid for publishing papers is profit for these dominant publishers' shareholders, and about half of them covers costs to keep the system running, including lobbying, marketing fees, and paywalls. The paywalls in turn restrict access of scientific outputs, preventing them from being freely shared with the public and other researchers. Thus, money that the public is told goes into science is actually being funneled away from it, or used to limit access to it. Alternatives to this model exist and have increased in popularity in recent years, including diamond open-access journals and community-driven recommendation models. These are free of charge for authors and minimize costs for institutions and agencies, while making peer-reviewed scientific results publicly accessible. However, for-profit publishing agents have made change difficult, by co-opting open-access schemes and creating journal-driven incentives that prevent an effective collective transition away from profiteering. Here, we give a brief overview of the current state of the academic publishing system, including its most important systemic problems. We then describe alternative systems. We explain the reasons why the move toward them can be perceived as costly to individual researchers, and we demystify common roadblocks to change. Finally, in view of the above, we provide a set of guidelines and recommendations that academics at all levels can implement, in order to enable a more rapid and effective transition toward ethical publishing.

National audience; Le concept de résilience, utilisé par de nombreuses disciplines, renvoie au sens commun de la capacité d’un système à surmonter des perturbations pour retrouver un fonctionnement souhaité. Cette capacité s’apprécie au regard de la capacité de réponse et de la sévérité de la perturbation subie. La résilience renvoie autant à une capacité d’adaptation que de transformation. Elle induit, dans le temps, une capacité d’apprentissage, tout particulièrement des acteurs quand il s’agit d’un système géré par des humains.

International audience

International audience; Recombination is expected to affect functional sequence evolution in several ways. On the one hand, recombination is thought to improve the efficiency of multilocus selection by dissipating linkage disequilibrium. On the other hand, natural selection can be counteracted by recombination-associated transmission distorters such as GC-biased gene conversion (gBGC), which tends to promote G and C alleles irrespective of their fitness effect in high-recombining regions. It has been suggested that gBGC might impact coding sequence evolution in vertebrates, and particularly the ratio of nonsynonymous to synonymous substitution rates (dN/dS). However, distinctive gBGC patterns have been reported in mammals and birds, maybe reflecting the documented contrasts in evolutionary dynamics of recombination rate between these two taxa. Here, we explore how recombination and gBGC affect coding sequence evolution in mammals and birds by analyzing proteome-wide data in six species of Galloanserae (fowls) and six species of catarrhine primates. We estimated the dN/dS ratio and rates of adaptive and nonadaptive evolution in bins of genes of increasing recombination rate, separately analyzing AT ! GC, GC ! AT, and G $ C/A $ T mutations. We show that in both taxa, recombination and gBGC entail a decrease in dN/dS. Our analysis indicates that recombination enhances the efficiency of purifying selection by lowering Hill-Robertson effects, whereas gBGC leads to an overestimation of the adaptive rate of AT ! GC mutations. Finally, we report a mutagenic effect of recombination, which is independent of gBGC.

This preprint has been reviewed and recommended by Peer Community In Evolutionary Biology (http://dx.doi.org/10.24072/pci.evolbiol.100031).Conservation policy in the giant Galápagos tortoise, an iconic endangered animal, has been assisted by genetic markers for ∼15 years: a dozen loci have been used to delineate thirteen (sub)species, between which hybridization is prevented. Here, comparative reanalysis of a previously published NGS data set reveals a conflict with traditional markers. Genetic diversity and population substructure in the giant Galápagos tortoise are found to be particularly low, questioning the genetic relevance of current conservation practices. Further examination of giant Galapagos tortoise population genomics is critically needed.

Summary In a context where commuter rail systems are more and more saturated, there is a strong interest in metering benefits for the quality of service provided by the introduction of Automatic Train Operation (ATO) or Automatic Train Supervision (ATS) systems on an existing saturated commuter line. Our original methodology and tool can perform this evaluation, based on a stochastic approach. This methodology was applied on an existing commuter with the stochastic analysis of 5 scenarios. The overall objective of the study is to discriminate the benefits of an ATS from that of an ATO alone. The study shows that the benefits brought by an ATS are comparable with the benefits brought by an upgrade of the signaling system., Dans un contexte de saturation des réseaux de transport ferroviaires urbains, la mesure des gains de qualité de service procurés par la mise en oeuvre de pilotage automatique (ATO) ou de régulation automatique (ATS) présente un grand intérêt. Notre outil, utilisant sur une méthodologie innovante basé sur une approche stochastique permet de réaliser cette évaluation. Cette méthodologie a été appliqué sur une ligne existante de trains de banlieue avec la comparaison de 5 scénarios. L'objectif principale de l'étude est de distinguer les bénéfices d'un ATS par rapport à un ATO seul. L'étude montre que les gains apportés par la régulation automatique sont du même ordre que ceux permis par une modernisation du système de signalisation.

Summary In a context where commuter rail systems are more and more saturated, there is a strong interest in metering benefits for the quality of service provided by the introduction of Automatic Train Operation (ATO) or Automatic Train Supervision (ATS) systems on an existing saturated commuter line. Our original methodology and tool can perform this evaluation, based on a stochastic approach. This methodology was applied on an existing commuter with the stochastic analysis of 5 scenarios. The overall objective of the study is to discriminate the benefits of an ATS from that of an ATO alone. The study shows that the benefits brought by an ATS are comparable with the benefits brought by an upgrade of the signaling system., Dans un contexte de saturation des réseaux de transport ferroviaires urbains, la mesure des gains de qualité de service procurés par la mise en oeuvre de pilotage automatique (ATO) ou de régulation automatique (ATS) présente un grand intérêt. Notre outil, utilisant sur une méthodologie innovante basé sur une approche stochastique permet de réaliser cette évaluation. Cette méthodologie a été appliqué sur une ligne existante de trains de banlieue avec la comparaison de 5 scénarios. L'objectif principale de l'étude est de distinguer les bénéfices d'un ATS par rapport à un ATO seul. L'étude montre que les gains apportés par la régulation automatique sont du même ordre que ceux permis par une modernisation du système de signalisation.

International audience

Letter; International audience

Background We report on the probable horizontal transfer of a mitochondrial gene, cytb, between species of Neotropical bruchid beetles, in a zone where these species are sympatric. The bruchid beetles Acanthoscelides obtectus, A. obvelatus, A. argillaceus and Zabrotes subfasciatus develop on various bean species in Mexico. Whereas A. obtectus and A. obvelatus develop on Phaseolus vulgaris in the Mexican Altiplano, A. argillaceus feeds on P. lunatus in the Pacific coast. The generalist Z. subfasciatus feeds on both bean species, and is sympatric with A. obtectus and A. obvelatus in the Mexican Altiplano, and with A. argillaceus in the Pacific coast. In order to assess the phylogenetic position of these four species, we amplified and sequenced one nuclear (28S rRNA) and two mitochondrial (cytb, COI) genes. Results Whereas species were well segregated in topologies obtained for COI and 28S rRNA, an unexpected pattern was obtained in the cytb phylogenetic tree. In this tree, individuals from A. obtectus and A. obvelatus, as well as Z. subfasciatus individuals from the Mexican Altiplano, clustered together in a unique little variable monophyletic unit. In contrast, A. argillaceus and Z. subfasciatus individuals from the Pacific coast clustered in two separated clades, identically to the pattern obtained for COI and 28S rRNA. An additional analysis showed that Z. subfasciatus individuals from the Mexican Altiplano also possessed the cytb gene present in individuals of this species from the Pacific coast. Zabrotes subfasciatus individuals from the Mexican Altiplano thus demonstrated two cytb genes, an "original" one and an "infectious" one, showing 25% of nucleotide divergence. The "infectious" cytb gene seems to be under purifying selection and to be expressed in mitochondria. Conclusion The high degree of incongruence of the cytb tree with patterns for other genes is discussed in the light of three hypotheses: experimental contamination, hybridization, and pseudogenisation. However, none of these seem able to explain the patterns observed. A fourth hypothesis, involving recent horizontal gene transfer (HGT) between A. obtectus and A. obvelatus, and from one of these species to Z. subfasciatus in the Mexican Altiplano, seems the only plausible explanation. The HGT between our study species seems to have occurred recently, and only in a zone where the three beetles are sympatric and share common host plants. This suggests that transfer could have been effected by some external vector such as a eukaryotic or viral parasite, which might still host the transferred fragment. Reviewers This article was reviewed by Eric Bapteste, Adam Eyre-Walker and Alexey Kondrashov.

International audience

chap. 6

chap. 13

International audience

International audience

International audience; Current phylogenomic methods are still a long way from implementing a realistic genome evolution model. An ideal approach would require a general joint analysis of genomic sequences, while including coding sequence annotation, protein evolution or gene transfer, among other mechanisms , to infer the complete evolutionary history of the studied genomes. Such an approach is computationally intractable and currently approximated by phylogenomic pipelines that implement a series of independent steps ranging from gene annotation to species tree inference or positive selection detection. Here we review the virtues and limits of current phylogenomic methods compared to what could be expected from an ideal method. We present five case studies to illustrate various issues and limits in current phylogenomic practices, while assessing their relative importance. We argue that data error is pervasive in modern datasets and models are still too simplistic compared to the complexity of biological and evolutionary processes. Importantly , joint analyses should be a research focus as the many steps of phylogenomic pipelines are not mutually independent. It is essential to recognize the hidden assumptions of the many types of analysis available to our community so as to circumvent model misspecifications and critically evaluate the relevance of their results. In conclusion, the quality of datasets should be enhanced via numerous, rigorous checkpoints, while also boosting the capability of models to handle biological complexity by the development of better models, particularly through joint analyses.

International audience; Bayesian inference is now routinely used in phylogenomics and, more generally, in macro-evolutionary studies. Beyond the philosophical debates it has raised concerning the choice of the prior and the meaning of posterior probabilities, Bayesian inference, combined with generic Monte Carlo algorithms, offers a flexible framework for introducing subjective or context information through the prior, but also, for designing hierarchical models formalizing complex patterns of variation (across sites or branches) or the integration of multiple levels of evolutionary processes. In this chapter, the principles of Bayesian inference, such as applied to phylogenetic reconstruction , are first introduced, with an emphasis on the key features of the Bayesian paradigm that explain its flexibility in terms of model design and its robustness in inferring complex patterns and processes. A more specific focus is then put on the question of modeling pattern-heterogeneity across sites, using both parametric and non-parametric random-effect models. Finally, the current computational challenges are discussed.

International audience; In the last decade, we witnessed the ascent of reconciliations as an important tool to model and study the evolution of gene families. Reconciliations model discordance between gene trees and species trees caused by gene-level processes: duplications, losses and transfers of genes, Incomplete Lineage Sorting among others can be combined to generate a panoply of different models. In this review article, we give an overview of this vast topic by skimming over the different models and methods that have been proposed, and presenting some of their applications in phylogenomics. We also present the pros and cons of these methods and give some directions for future research that we are convinced will enhance their efficiency and use.

Gene tree discordance is now recognized as a major source of biological heterogeneity. How to deal with this heterogeneity is an unsolved problem, as the accurate inference of individual gene tree topologies is difficult. One solution has been to simply concatenate all of the data together, ignoring the underlying heterogeneity. Another approach infers gene tree topologies separately and combines the individual estimates in order to explicitly model this heterogeneity. Here we discuss the advantages and disadvantages of both approaches-using the gene trees singly or in concatenation-paying special attention to the sources of variance and implicit assumptions. We make it clear that all methods are likely to have their assumptions violated, though the consequence of these violations differs in different parts of parameter space. The main conclusion of our review is that different sources of error are more or less important in different settings, such that phylogenetics researchers should be using the methods most appropriate to their problems rather than stick to one dogmatically.

International audience; Species delimitation is the process of determining which groups of individual organisms constitute different populations of a single species and which constitute different species. The problem goes back to the earliest days of taxonomy and formalized processes for describing new species exist and are widely used, although the methods are time-intensive and problematic for some species. Genomic data carries extensive information about the degree of genetic isolation among species and about ancient and recent introgression. For this reason, genomic data can play an important role in species delimitation under many existing species concepts. Here we review the history of molecular species delimitation leading up to the current genomic era. We then describe the most widely used computational methods for species delimitation using single-and multi-locus genomic data. Relative strengths and weaknesses of the approaches are discussed and a new method for delimiting species based on empirical criteria is proposed.

International audience; The number of possible unrooted binary trees (phylogenies) increases super-exponentially with the number of taxa. To find the Maximum Likelihood (ML) tree one has to enumerate and evaluate all these trees. As we will see, this is computationally not feasible. Therefore, one predominantly deploys ad hoc tree search methods that strive to find a "good" ML tree in the hope that it will be close, either with respect to the likelihood score or the topological structure, to the globally optimal ML tree. In this chapter we provide an overview over the most popular and efficient ML tree search techniques. How to cite: Alexandros Stamatakis and Alexey M.

International audience; Modern methods to detecting adaptive evolution from interspecific protein-coding gene alignments rely on statistical models of sequence evolution formulated at the level of codons. By performing model comparisons, one measures the evidence for signals of adaptive substitution processes, relative to a null model that disallows any adaptive regime. In this chapter, we present the detailed form of these models of sequence evolution, and how they are applied to real data sets. The classical codon substitution models are based on evaluating the relative nonsynonymous to synonymous substitution rates, and the main focus has traditionally been placed on devising models allowing for increasingly more subtle manifestations of adaptive substitution processes. We also overview a contrasting modeling direction that has emerged in the last decade-although with roots two decades back-in which the emphasis is placed on devising a richer modeling of purifying selection. Using simulations, we expand the characterization of this latter approach, followed by a contrasting of its conclusions on real data with those of classical codon models. Finally , we discuss the numerous model violations that can lead to erroneous inferences on various tests, and potential future directions meriting attention.

RAxML-NG is a new phylogenetic inference tool that replaces the widely-used RAxML and ExaML tree inference codes. Compared to its predecessors, RAxML-NG offers improvements in accuracy, flexibility, speed, scalability, and user-friendliness. In this chapter, we provide practical recommendations for the most common use cases of RAxML-NG: tree inference, branch support estimation via non-parametric bootstrapping, and parameter optimization on a fixed tree topo-logy. We also describe best practices for achieving optimal performance with RAxML-NG, in particular, with respect to parallel tree inferences on computer clusters and supercomputers. As RAxML-NG is continuously updated, the most up-to-date version of the tutorial described in this chapter is available online at: https://cme.h-its.org/exelixis/raxml-ng/tutorial How to cite: Alexey M. Kozlov and Alexandros Stamatakis (2020).

Recombination breaks up the evolutionary history between genomic regions and, as a result, the evolutionary history of different genomic regions may differ. In fact, conflicting phylogenetic signal between genes is commonplace. The reasons for conflicting signal may be statistical or systematic in nature. In order to avoid strongly supported but incorrect inferences driven by systematic error, use of appropriate phylogenetic methods accounting for these processes is of fundamental importance. This chapter reviews possible causes of phylogenetic conflict between genes. Processes generating conflict, including gene duplication and loss, horizontal gene transfer, hybridization, and incomplete lineage sorting are presented using classic examples. In particular, we discuss compelling evidence for whole genome duplications in fish, as well as plants, and the role of horizontal transfer in the spread of antibiotic resistance. Finally, building on the material presented, we show how these processes lead to phylogenetic conflict, and how they can be described by phylo-genetic models.

International audience; Orthology is a central concept in evolutionary and comparative genomics, used to relate corresponding genes in different species. In particular, orthologs are needed to infer species trees. In this chapter, we introduce the fundamental concepts of orthology relationships and orthologous groups, including some non-trivial (and thus commonly misunderstood) implications. Next, we review some of the main methods and resources used to identify orthologs. The final part of the chapter discusses the impact of orthology methods on species phylogeny inference, drawing lessons from several recent comparative studies.

International audience

One of the basic questions of phylogenomics is how gene function evolves, whether among species or inside gene families. In this chapter, we provide a brief overview of the problems associated with defining gene function in a manner which allows comparisons which are both large scale and evolutionarily relevant. The main source of functional data, despite its limitations, is transcriptomics. Functional data provides information on evolutionary mechanisms primarily by showing which functional classes of genes evolve under stronger or weaker purifying or adaptive selection, and on which classes of mutations (e.g., substitutions or duplications). However, the example of the "ortholog conjecture" shows that we are still not at a point where we can confidently study phylogenomically the evolution of gene function at a precise scale., Comment: To be published in book "Phylogenomics" (Nicolas Galtier, C\'eline Scornavacca, Fr\'ed\'eric Delsuc, Eds.)

Multiple sequence alignment (MSA) is a prerequisite for most phylogenetic analyses. Aligning sequences to unravel residue homology is a challenging task that has been the focus of much attention in recent decades. Research in this field has been extremely active from both theoretical and practical standpoints. Numerous tools have been developed to align sequences and, more recently, to post-process those alignments and filter out their most dubious parts. Whether or not the inclusion of alignment filtering in a phylogenetic pipeline improves the quality of the inferred phylogenies is still debatable. The goal of this chapter is not to provide an exhaustive list of all tools available to produce or filter an MSA, but rather to cover the limitations of current alignment methods and their causes, to highlight key differences among MSA filtering methods and provide some practical MSA filtering guidelines. We consider that filtering methods can be subdivided into two main categories. The first one includes methods that filter MSA by entirely removing some sites or sequences from the MSA. The second category contains MSA filtering methods that mask residues and are able to extract some pieces of information from a site or sequence, while disregarding the remaining information-we called these picky-filtering methods. In our benchmark, the filtering methods that perform best are, as expected, in the picky category. When inferring phylogenies, MSA filtering impacts the inferred tree topology but it also seems to significantly improve branch length estimations, especially when a picky-filtering method is used. How to cite: Vincent Ranwez and Nathalie Chantret (2020).

The study of the tempo and mode of molecular evolution has played a key role in evolutionary biology, both as a stimulant for theoretical enrichment and as the foundation of useful analytical tools. When protein and DNA sequences were first produced, the surprising constancy of rates of change brought molecular evolution into conflict with mainstream evolutionary biology, but also stimulated the formation of new theoretical understanding of the processes of genetic change, including the recognition of the role of neutral mutations and genetic drift in genomic evolution. As more data were collected, it became clear that there were systematic differences in the substitution rate between species, which prompted further elaboration of ideas such as the generation time effect and the nearly neutral theory. Comparing substitution rates between species continues to provide a window on fundamental evolutionary processes. However, investigating patterns of substitution rates requires attention to potential complicating factors such as the phylogenetic non-independence of rates estimates and the time-dependence of measurement error. This chapter compares different analytical approaches to study the tempo and mode of molecular evolution, and considers the way a richer biological understanding of the causes of variation in substitution rate might inform our attempts to use molecular data to uncover evolutionary history.

A large body of research is dedicated to model sequence evolutionary dynamics. The evolutionary process may vary within groups of genes, among sites within a gene, between populations and among diverged species. Evolutionary models aiming to describe these dynamics must account for base pair substitutions as well as insertion and deletion (indel) events. Here, we explain the fundamental of continuous time Markov models used to describe sequence evolution. We begin by describing discrete Markov models, and slowly progress towards more realistic and more computationally complicated continuous time Markov models. Among other topics, we discuss nucleotide, amino acid, and codon models, among site rate variation, model reversibility, stationary distributions, rate matrix normalization, mixture models, indel models, and models of gene family evolution. Understanding the concepts presented here is vital for various phylogenom-ics analyses such as the inference of positive selection, alignment and phylogeny reconstruction, ancestral sequence reconstruction, and molecular dating.

Gene tree discordance is now recognized as a major source of biological heterogeneity. How to deal with this heterogeneity is an unsolved problem, as the accurate inference of individual gene tree topologies is difficult. One solution has been to simply concatenate all of the data together, ignoring the underlying heterogeneity. Another approach infers gene tree topologies separately and combines the individual estimates in order to explicitly model this heterogeneity. Here we discuss the advantages and disadvantages of both approaches-using the gene trees singly or in concatenation-paying special attention to the sources of variance and implicit assumptions. We make it clear that all methods are likely to have their assumptions violated, though the consequence of these violations differs in different parts of parameter space. The main conclusion of our review is that different sources of error are more or less important in different settings, such that phylogenetics researchers should be using the methods most appropriate to their problems rather than stick to one dogmatically.

International audience; The reconstruction of the chromosomal organization of ancient genomes has many applications in comparative and evolutionary genomics. Here we propose a novel, methodological, use for these predicted ancestral syntenies, directly focused on phylogenomics. It is a way to assess the accuracy of gene trees and species trees. We use a method that reconstructs, from gene trees and extant gene orders, ancestral adjacencies, i.e. the immediate neighborhood between pairs of genes, independently for each pair. This independence allows to split the computations into many independent problems that can each be solved exactly using efficient algorithms, but might result in sets of ancestral adjacencies that are incompatible with the expected linear or circular structure of chromosomes. We show here that this drawback can actually be turned into a useful 1 This work benefited from the support of ComputeCanada.

International audience; Twenty years of standardized DNA barcoding practice have resulted in millions of sequences being produced for a handful of molecular markers in a wide range of fungi, animal and plant species. Despite some basic quality controls, reference barcoding data sets deposited in the Bar-code of Life Datasystem (BOLD) database are not immune to sequencing errors and undetected pseudogenes. Such database inaccuracies can significantly bias subsequent species delimitation and biodiversity estimation based on DNA barcoding. These potential problems are amplified in metabarcoding studies containing thousands of sequences produced using high throughput se-quencing technologies. Here, we propose a pipeline based on MACSE v2, an extended version of our codon-aware multiple sequence alignment software accounting for frameshifts and stop codons. The MACSE_BARCODE pipeline allows the accurate alignment of hundreds of thousands of protein-coding barcode sequences. Re-analyses of published data sets confirm that MACSE v2 is able to automatically detect most sequencing errors previously identified manually. The proposed alignment strategy hence alleviates the risk of incorrect species delimitation due the incorporation of sequencing errors or undetected pseudogenes. By applying the MACSE_BARCODE pipeline to mammal, ant, and flowering plant barcode sequences available in BOLD, we highlight several cases of database errors and provide curated reference alignments for the main protein-coding barcode genes. We anticipate our approach to be particularly useful for metabarcoding studies in which thousands of new sequences need to be compared to a reference database for subsequent taxonomic assignment. This might prove particularly helpful for diet characterization studies and large-scale biodiversity assessments through environmental DNA metabarcoding. The new MACSE_BARCODE pipeline is distributed as Nextflow workflows that are available from the MACSE project webpage (https://bioweb.supagro.inra.fr/macse/).

Protein sequence convergence refers to substitutions leading to the same amino acid residue at the same position of a protein in multiple independent evolutionary lineages. Protein sequence convergence is often viewed as adaptive signal so is of great interest to evolutionary biologists. In this article, we review complications in identifying sequence convergences, statistical tests of the null hypothesis that the observed convergence events in a protein are attributable to chance alone, interpretations of genome-wide observations of sequence convergence, and a comparison in the susceptibility of molecular and morphological characters to convergence and its phylogenetic implications. We highlight the substantial progresses made in the last two decades and point out the main challenges at the present.

International audience; PhyloBayes is a software program for Bayesian phylogenetic reconstruction. Compared to other programs, its main distinguishing feature is the implementation of the CAT model, which accounts for fine-grained variation across sites in amino acid preferences using a Bayesian non-parametric approach. This chapter provides a detailed step-by-step practical introduction to phylogenetic analyses using PhyloBayes, using as an example a previously published dataset addressing the phylogenetic position of Microsporidia within eukaryotes. Through this historically emblematic case of a long-branch attraction artifact, a complete analysis under site-homogeneous and site-heterogeneous models is conducted and interpreted, thus providing an illustration of why modeling pattern variation is so fundamental for reconstructing deep phylogenies.