Leveraging putative enhancer-promoter interactions to investigate two-way epistasis in Type 2 Diabetes GWAS.

The FUSION p-value was higher in this case, but remained borderline significant (p=0.155562).

also in 11,409 other papers

Computational approaches include methods such as ReliefF and its derivatives, 8 – 10 MDR, 11 and “greedy” approaches which first identify SNPs with main effects (significant or marginally significant) in a GWAS and then use models involving only these SNPs. 12 – 13 The latter is certainly a reasonable approach; however it will miss potential interactions which involve SNPs with no main effects (i.e. it will miss what is referred to as ‘pure and strict epistasis’ 14 ).

also in 26,082 other papers

Here the pair did not quite reach significance, but its p-value remained relatively low.

also in 350 other papers