Société Française de Bioinformatique

Mots-Clés

Bio-analysis Population genetics Genes under selection

Description

The greater yam (Dioscorea alata L.) is the most widespread edible yam species and is cultivated throughout sub-tropical and tropical areas. The species strongly contributes to food security in West Africa, where its production is expanding because of its ease of cultivation, in the Caribbean and the Pacific where it has considerable social and cultural importance, and also in parts of upland Asia. Despite of its importance and the release of different genome references, little is known about the evolutionary biology of Dioscorea alata. Up to now, the investigation of its evolutionary history was hampered by the lack of rational sampling and by the ploidy levels of the species.

We have been gathering accessions of Dioscorea alata from different geographical origins to build up a collection representative of the worldwide diversity. This panel represents approximately 199 accessions for which whole sequencing data are available.

Within this internship, the candidate will apply population genomic approaches to this worldwide sample in order to understand the demographic history of the species and to identify the genes under selection. Dioscorea alata is a polyploid species (2X, 3X and 4X). The accessions will be analyzed using with multivariate approaches (e.g. PCA, sNMF), the structure will be investigated using population genetics approaches to characterize the spatial distribution of genetic diversity and identify genetically different populations. Several genetic diversity statistics (e.g. π, Tajima’s D) will be calculated for each population, to quantify the levels of nucleotide diversity and identify signatures of effective population changes (expansions or bottlenecks). Finally, using the annotated genome and the D. alata transcriptome, analyses will be also be conducted to identify genes under selection between the different populations. When possible, 3X and 4X accessions will be analysed with the same methods applied to the 2X accessions (Hardigan et al. 2017).