A manuscript on "Development and application of genomic control methods for genome-wide association studies using non-additive models" was recently published on-line (PLOS ONE). The work reported in the manuscript extends the genomic control method to non-additive models; all methods are implemented in GenABEL.
(To continue reading this story click on 'read more' below.)
With methods described, it is possible to correct the inflated statistic for non-additive models (such as recessive, dominant, over-dominant, and genotypic) in case a study population has genetic (sub)structure e.g. due to cryptic relatedness. Estimation of the correction coefficient is possible via the Variance Inflation Factor (VIF) and via polynomial of fixed degree (polynomial genomic control, PGC). The VIF estimation allows the corrections of 1df tests only, while PGC has more general application and also works for 2df genotypic test.
All methods reported in the manuscript are available in the GenABEL-package (part of the GenABEL suite) as functions PGC, VIFGC, VIFGC_ovedom.
We also take a chance to congratulate to Yakov Tsepilov - the first author of this study, and the GenABEL project contributor - on his first-as-first-author paper! Well done, and looking forward to see your next work!