Major CNV Detection By Read-Depth Analysis Using Capture-Based GBS in Cotton

Copy number variation (CNV) reflects differential abundance of genomic sequences relative to each other and ploidy. CNV arises by multiple mechanisms, and affects genomic regions that range widely in size. Traits such as nematode resistance and aluminum tolerance can involve CNVs. In Upland cotton (Gossypium hirsutum L.), CNV detection has been hindered by the polyploid and paleopolyploid heritage, abundance of repetitive sequence elements, and availability of suitable genomics information and analytical methods. We assessed a strategy for cotton genomic CNV analysis involving quantitative analysis of targeted NGS data obtained after DNA “capture”, by analyzing dosage differences among and within G. hirsutum and G. hirsutum x G. barbadense euploid, hypoaneuploid or radiation hybrid genotypes. BAC-end SNP sequences were used as baits for capture by Capture-Seq and as markers for CNV detection through read-depth analysis. Applying this to 9.6 Mbp of dispersed SNP-laden target regions, we aligned reads to BAC-end Sanger sequences, and called SNPs from aligned sequences using a previously proven bioinformatic pipeline. Specific genomic regions within intraspecific G. hirsutum hypoaneuploids exhibited reduced read depth indicating success. Moreover, analogous interspecific hypoaneuploid F1 hybrids exhibited expected CNV-related changes in read-depth AND concordant hemizygosity. Extension to G. hirsutum and G. barbadense F1 radiation hybrid lines revealed a similar concordance. These three experimental scenarios provided unequivocal support for the detection of cotton CNVs through read-depth analysis of NGS, and the applicability of a capture-based approach. Research supported by NSF IOS1444552, NIFA-NNF 2014-38420-21835, and Cotton Inc 13-694.