After phenotypic association analysis, the quality control measures used should be evaluated. One method is to compare the observed number of statistically significant results with a phenotype to expectation. Too many significant results may indicate insufficient QC. Also because no QC will catch all problematic SNPs, the intensity plots for statistically significant SNPs must be reviewed to make sure there are no obvious clustering problems. Replication of results using different genotyping technology (such as TaqMan) and/or in another sample may be needed as well.
