P0758Three-Fourths of Maize Presence-Absence Variants (PAVs) are not in Linkage Disequilibrium (LD) with Nearby SNPs
Three-Fourths of Maize Presence-Absence Variants (PAVs) are not in Linkage Disequilibrium (LD) with Nearby SNPs
Genomic structural variation is increasingly being recognized as a major contributor to phenotypic variation across species. In plants structural variants contribute to traits including resistance to soybean cyst nematode, boron tolerance in barley, dwarfism in wheat, flowering time in both barley and wheat, and aluminum tolerance in maize. Presence-absence variants (PAV), a form of structural variation where a DNA sequence is present in one genome but entirely missing from another, has been found to contribute to a number of genetic diseases in humans. While the association between PAVs and phenotype is still being explored in plants with some examples, including flavor quality in strawberries and submergence tolerance in rice have been uncovered. We enabled global explorations of the association between PAVs and phenotypic variation by conducting RNA-sequencing on the 27 inbred founders of the maize Nested Association Mapping (NAM) population. This exploration identified thousands of putative expressed PAVs that are absent from B73, the maize reference. In collaboration with Roche-Nimblegen these putative PAVs were used to design a Zeanome sequence capture array. Sequence capture of the NAM founders using the Zeanome array identified ~12,000 genomic PAVs that are absent from B73. Characterization of these PAVs shows that many have similarities to characterized proteins. Using the Fluidigm BioMark Dynamic Array, we genotyped for a subset of the PAVs and genetically mapped them in the NAM RILs. Additionally, a subset of ~200 PAVs were genotyped and linkage disequilibrium (LD) mapped using 400k single nucleotide polymorphisms (SNPs) in the Goodman Association Panel. Fewer than a quarter of PAVs were found to be in LD with SNPs. The lack of LD suggests that the power of association studies that include only SNPs could be improved by the inclusion of PAV genotyping.
