Impact of recombination and biased gene conversion on genome evolution: application to animal comparative genomics – CoGeBi

The identification of functional elements (genes, regulatory elements) in large eukaryotic genomes such as those of animals and plants is a major issue. Sequence annotation relies heavily on comparative genomics: by analyzing patterns of sequence variation between different species or within populations, it is possible to detect the hallmarks of natural selection and hence to identify functional elements within genomic sequences.
Up to now, the methods that have been developed to detect selection assume that sequence evolution depends only on 3 processes: mutation, genetic drift and selection. We have recently shown that in some taxa (notably in mammals and birds), sequence evolution is strongly affected by recombination, because of the process of biased gene conversion (BGC). This process mimics the action of selection and can therefore confound the classical selection tests. Moreover, BGC can even counteract the effect of selection and lead to the fixation of deleterious mutations.
We propose here to investigate thoroughly the impact of BGC on the evolution of functional regions and genomic landscapes (gene density, base composition, …) in animal taxa. First we will conduct theoretical population genetics developments to investigate how BGC may interfere with selection and how BGC could affect the evolution of recombination. We will then develop new bioinformatics methods to detect selection. These methods will account for BGC, as well as non-homogenous and non-stationary substitution processes.
We will use these new tools to quantify the impact of BGC on the evolution of protein-coding genes in animal taxa. We will also evaluate the interference between BGC and selection by analyzing synonymous codon usage biases.We have shown that BGC can counteract the action of selection and lead to the fixation of deleterious mutations. Thus BGC may be responsible for the maintenance of deleterious alleles in human populations. To determine whether this effect is quantitatively important, we will analyze the spectrum of mutations involved in human genetic disorders. Moreover we will analyze the taxonomic distribution of BGC in animal phyla to determine which taxa are affected by this process and to study which are the factors that contribute to the prevalence of BGC. We will notably investigate the impact of chromosomal rearrangements on changes in recombination rate and hence on BGC. Furthermore, we will study whether recombination affects the pattern of insertion and deletion (and hence gene density) in animal genomes.
Finally, we will re-evaluate the strength of selection in different genomic compartments, in taxa for which genome-wide multiple alignments and or polymorphism data are available. We will notably quantify positive selection in promoter regions (that are believed to be major contributors of adaptative evolution) for which we suspect an important impact of BGC. We will also search for positive selection in all possible genomic compartments for which there is some evidence that they are functional.
By characterizing the functional and evolutionary consequences of the newly-discovered BGC, our project is expected to impact animal genomic research both conceptually and practically. From a fundamental viewpoint, we will understand whether, and how much, the dynamics of functionally relevant mutations (advantageous or deleterious) is affected by BGC. Next we will assess the impact of BGC on genomic landscapes in many animal taxa. In addition, our project will yield a variety of new theoretical and bioinformatics tools that will be essential for the annotation of functional elements in genomes that are affected by BGC (not only mammals and birds, but also many other eukaryotes: plants, fungi, and ciliates).