P0076
Extensive DNA Variant Detection in a Cattle Population by Low-Coverage Re-Sequencing
Extensive DNA Variant Detection in a Cattle Population by Low-Coverage Re-Sequencing
Date: Monday, January 14, 2013
Room: Grand Exhibit Hall
Re-sequencing a limited number of key ancestors of a population allows to capture a major proportion of the population's genomic variation. Thirty-nine key ancestors and four contemporary animals of the Fleckvieh (FV) cattle breed were selected for re-sequencing. These animals explain 68% of the gene pool of the current FV population. An average effective coverage of 7.4x per animal was obtained by next-generation sequencing. After the sequences had been aligned to the UMD3.1-assembly of the bovine genome, multi-sample SNP- and Indel-calling was performed. In total, 17.3 million SNPs and Indels could be detected. After an imputation step, 96.0% of genotypes showed phred-scaled quality scores >10. Comparing the sequence-derived SNP genotypes with high-density array genotypes, non-reference sensitivity amounted at least to 98% and the non-reference discrepancy rate ranged from 0.3% to 5.2%. 91,773 variants were annotated within coding regions of 18,139 genes. Among them, 46.3% led to non-synonymous amino acid exchanges and 575 to a premature stop codon. Structural variant (SV) calling with a read-pair method yielded 14,770 deletions, 5742 insertions, 1374 duplications and 687 inversions. The average length of a SV was 900 bp. 8011 SVs affect genes, with 13% of the variants being located in coding and 12% in regulatory regions. The variant catalog will be imputed in thousands of animals genotyped at high density and will be used for subsequent fine-mapping of QTL regions to identify causal variants.
