Résumé
Duplications are one on the main evolutionary forces in angiosperm, especially in Poaceae. A large number of genes involved in various metabolisms and pathways originate from such duplications (whole genome, segmental or single gene). However, to detect such duplication may be complicated, costly and generally requires heavy human and material investments. Here, we propose an alternative approach for detecting putative recent segmental duplications in haploid or diploid homozygous organisms based on NGS data. We rely on abusive mappings of paralogous sequences that increase apparent heterozygous points at a given locus to identify such duplicated genomic regions. We test our tool on simulated data, then on truerice genomic sequences and were able to identify about 200 candidate duplicated genes in African rice (Oryza glaberrima) lineage compared to Asian one (O. sativa).