The SRKN_ML_GWAS_public.R include a Machine learning based GWAS algorithm to identify key SNPs that contribute to SRKN gall rating scales Reference:
The input dataset has 717 samples each with 6K SNPs.
The output dataset has the gall rates for the 717 samples. 1 < 10 galls per plant, 2 = 11 to 20, 3 = 21 to 30, 4 = 31 to 40, and 5 >40 galls. For classification purposes, lines were considered tolerant when < 20 galls per plant, moderate > 20, <40, and susceptible >40 galls per plant.