Genome-Wide Association Study of Canine Bloat in WGS and SNP Arrays
Gastric dilatation volvulus (GDV) is the second most life-threatening condition that occurs in pet dogs, with a higher risk in large or giant-size breeds. This condition is caused by a volvulus, or a twist, in the stomach that leads to an accumulation of pressure and reduced blood flow throughout the cardiovascular system resulting often times in sudden death. A diagnosis based on symptoms such as swollen stomach and excessive salivation is followed by an invasive surgery called gastropexy that can cost up to $5000. The monetary cost added to the emotional burden on pet owners caused by the sudden loss of a dear mascot could be addressed by early predicting GDV.
In this study, we aim at finding a method to predict GDV through the detection of genetic risk factors. It has been found that environmental factors do not seem to affect a dog’s risk for GDV, but that first-degree relatives (parents, siblings, and offspring) are more at risk if a related dog has GDV, suggesting a genetic component. However, little is known about genetic risk factors for GDV. We are utilizing whole genome sequencing (WGS) and SNP array analysis to conduct a genome-wide association study in affected and breed/age/sex-matched unaffected controls to investigate if there are any single nucleotide polymorphisms (SNPs), structural variants, or/and copy number variants associated with risk of developing GDV.
DNA was extracted from 369 dogs, cases and controls. Thirty-three were whole genome sequenced as a discovery cohort and an additional undergoing genotyping used as an expansion cohort. The discovery cohort included 3 breeds: Borzoi, Great Dane, and German Shepherd, with 5 cases and 6 controls (11 samples) per breed. WGS was performed on the Illumina NovaSeq and analyzed in an established informatics pipeline and with custom scripts. These scripts include haplotype calling with conversion of file format to gVCF, Joint Variant Calling, and allele filtering with VCF tools and Variant Quality Score Recalibration. One of the preliminary analyses completed was Principal Component Analysis. This was conducted to ensure breed composition, and we found that each breed clustered independently using chromosomes 1, 3, and 7. We were then able to select a control sample from each breed to run a Joint Variant Call against the cases. Once this step was performed, we filtered the files obtained from the Joint Variant Call in order to distinguish the different genetic variations in the cases as compared to the controls. Detailed analysis of WGS data is ongoing while SNP arrays are being completed. These data may help identify genetic GDV risk factors that can be utilized for screening and may inform broader understanding of GDV biologic underpinnings and clinical management.