Origins and impacts of chromosomal inversions in the evolution and adaptation of pearl millet

Abstract

Structural variation such as inversions can play important roles in evolutionary processes such as adaptation and speciation. The aim of the PhD was to study inversions in pearl millet (Cenchrus americanus (L.) Morrone syn. Pennisetum glaucum (L.) R. Br.), an African cereal that plays an important role in food security in several Sub-Saharan African countries. We first improved the assembly of the reference genome using an approach combining long-read sequencing data and optical maps. We then used this new genome and detected inversions using a population genomics approach in a population of cultivated plants in Senegal and on a broader sample of wild and cultivated pearl millet across Africa. Some of the inversions identified were confirmed using long-read assembly or comparisons of assembled genomes. Our study showed that large structural variants of several tens of megabases in size are present in Senegalese cultivated populations, notably within the centromeric regions. In-depth study of a large region of almost 90 megabases in size on chromosome 3 showed that multiple large divergent haplotypes, harboring or not inversions, are present in the population across this region and that a large variant is probably recessive lethal. A hypothesis of pseudo-overdominance was proposed to explain the maintenance of this variation, in particular through the sheltering of deleterious mutations that could accumulate within inversions and genomic regions with reduced recombination. We also also showed that these divergent haplotypes have been probably introgressed in cultivated populations following hybridization with wild relatives. Our study brings new and surprising results about the maintenance of large structural variants in plant genomes. One of the main hypothesis for the maintenance of this diversity is a major role played by selection associated with introgressions.