Guidance for Gene-level test Gene_P

[zyh4482]

The questions might be naive.

May I ask how should I interpret the p value and strongest mask of GENE_P results? Should this pVal be taken similarly like ADD for significance discovery(e.g. setting an empirical cutoff of 5*10^-8)? Should I interpret the Strongest mask (for example mask1 is stronger) as Mask1 set significantly correlated to trait comparing with mask2?

Thank you so much if you could provide any guidance on this.

[joellembatchou]

Hi,

So when you look at the strongest mask (ie the one with the smallest p-value) among the masks included in GENE_P, you need to account for multiple testing as GENE_P encompasses multiple masks & statistical tests.
GENE_P is used for an initial scan across sets (eg gene) of which ones provide evidence of association with a given trait. It accounts for multiple testing (across masks and statistical tests) using the Cauchy combination test ACAT so you only need to correct for the number of sets/genes tested when looking at GENE_P.

Cheers,
Joelle