Adsa Logo White Adsa Title White

Profiles of causative SNP in a genome-wide association study.

I. Misztal




Profiles of causative SNP in a genome-wide association study.
I. Misztal*1, I. Pocrnic1,2, M. Perez-Enciso3, D. A. L. Lourenco1. 1University of Georgia Athens, GA, 2The Roslin Institute Midlothian, United Kingdom, 3CRAG Barcelona, Spain.

The purpose of this study was to see the impact of causative SNP on GWAS with different populations with different effective population size. Three populations were simulated assuming 100 equidistant causative SNP with identical substitutions effects. Causative SNP were included in 50 k SNP genotypes. Ten generations were simulated, with the last 3 genotyped. Population NE60 was composed of 2000 animals per generation with effective population size 60. Population NE600 was composed of the same number of animals but with effective population size 600. NE60_3x was as NE60 but with 6000 animals per generation. Analyses were performed using single step GBLUP, with solutions converted to SNP values and subsequently to p-values for each SNP; in a GBLUP context, p-values are equivalent to those in standard GWAS methodology, where each SNP is treated as fixed effect, and a genomic relationship matrix accounts for the population structure. Manhattan plots for standardized SNP solutions showed large values for few of the 100 causative SNP and were very noisy. Manhattan plots for p-values were similar to those for SNP solutions. The number of SNP effects with p-values over the statistical threshold was smallest for NE60, larger for NE60_3X, and the largest for NE600. SNP profiles were created by averaging SNP solutions � 100 SNP around causative SNP. The profiles showed distinct peak for the causative SNP, with smaller signals for adjacent SNP. The peak was smallest for NE60 and largest for NE600. Each causative SNP influenced about 50 adjacent SNP for NE60 and NE60_3X, and about 10 SNP for NR600. The profiles help understand that the effective use of causative SNPs requires knowing their exact positions and either boosting their variance in analyses or elimination of SNPs adjacent to causative SNPs.

Keywords: genomic selection, causative SNP, sequence data.