I was been working on epistasis in a couple of GWAS loci. Basically the theory is that some GWAS loci might not work in an additive manner, but more in an epistasic way and calculated this using R :
Formula <- formula(paste("pheno ~ PC1 + PC2 + PC3 + PC4 + PC5 + SEX + SNP1 * SNP2")) model_test <- glm(Formula, data = data, family = "binomial") summary(model_test)
Now I was wondering are there already "proven" robust epistasic associations out there that in any disease/trait between GWAS loci? Have been googling a bit but haven't really found anything, the majority of the papers out there are methods and theory papers..