Bioinformatic Approaches for Comprehensive Variants Discovery on the 3,000 Rice Genomes Project

The 3,000 rice genomes project (3KRGP) is arguably the largest current sequencing effort to elucidate genomic variability in a staple crop. Information obtained from bioinformatic analysis of whole genome resequencing (WGS) of the 3,000 varieties selected for this project will make a great resource not only for rice biology and population dynamics but also for applied rice breeding. To maximize the information obtained from WGS data we recently developed the software package NGSEP, which combines state-of-the-art algorithms to identify and genotype different kinds of genomic variation including SNPs, SSRs, indels and copy-number variants (CNVs). We demonstrated the use of NGSEP to reveal millions of SNPs and thousands of structural variation events on a panel of 94 O. sativa varieties an 10 wild relatives, which included 54 elite cultivars both from North America and Latin America. As part of the IRIC consortium, we now have used the algorithms implemented in NGSEP to predict structural variants in the complete WGS dataset of the 3KRGP. We identified more than one thousand CNVs and 10 thousand large deletions with alternative allele frequencies greater than 0.01 in regions with low repeat density and covering 4.5 and 1 Mbp of the genome respectively. About 50% of these events span, and sometimes even completely cover annotated genes, suggesting significant functional variation within rice due to structural variants. We expect that the variants obtained by this analysis complement the SNPs previously released to assemble a comprehensive information resource useful for several groups across the rice research community.