Recovering the obscured genetic architectures through deconvolution of GWAS statistics

The aim for this study was to propose a novel deconvolution technique to improve the recovery of genetic architecture information for a trait that was obscured by strong marker linkage disequilibrium. With a simulated datasets with sample size of 3,500, this deconvolution technique has reduced the departure in distribution between true and estimated marker test statistics by up to 5.82-folds on a trait with heritability of 0.3 and 2,000 quantitative trait loci. Under the same scenario, deconvolution also improved the correlation between true and haplotype significance of association by 24.9%. Therefore, deconvolution could serve as a promising approach in recovery of genetic architecture information that could help in its elucidation in livestock traits.