Skip to main content
. 2015 Sep 25;32(2):203–210. doi: 10.1093/bioinformatics/btv504

Fig. 1.

Fig. 1.

Effects of hidden subphenotypes on GWAS. (A) Classical CC GWAS wherein heterogeneity in the case population is hidden or is not accounted for; can recover only shared causal markers (SNP3) (B) Subphenotyping approach to the same data recovers both shared (SNP3) and subphenotype causal-specific markers (SNP2, SNP4 and SNP5)