Improving the Maize Genome Via GBS (Genotyping by Sequencing)

Genotyping by sequencing (GBS) provides a robust and cost-effective means to genotype large numbers of individuals at high density by targeting sequence adjacent to restriction enzyme cut sites. We genetically mapped (binomial p-value <0.001) 589,002 segregating 64-base ApeKI GBS tags as presence/absence markers in the maize IBM population relative to a new, high density map of 239 IBM lines based on the Illumina MaizeSNP50 Genotyping BeadChip. Using a high stringency (p<10^-7) subset of these genetically mapped tags, we found that, of the GBS tags that segregated with B73, only 0.4% genetically mapped to a chromosome different from the one to which they physically align on B73 RefGen_v2. In contrast, for tags that segregated with Mo17, the comparable proportion was 9.3%. This difference likely results from structural variation combined with the ancient polyploidization that occurred in maize. We have also used the GBS tag segregation data to genetically map (1) contigs from chromosome 0 of B73 RefGen_v2, (2) contigs from de novo 454 sequencing of B73, and (3) full length cDNAs. We were able to genetically map 8 of the 17 chromosome 0 contigs, 3408 novel 454 contigs and 407 FLcDNAs that are not represented in B73 RefGen_v2. Segments of the B73 reference genome to which elevated proportions of novel 454 contigs genetically map are prime targets for future improvement. To improve resolution, we are currently mapping de novo, misplaced and unplaced contigs using GBS data from the maize NAM population in which B73 is the common parent.