Highly correlated, non-significant modules
0
0
Entering edit mode
@bramverstockt-11079
Last seen 7.3 years ago
KUL, Belgium

Dear all

While performing WGCNA analysis, has anyone found modules/MEs highly correlated to a trait of interest (bicor 0.45-0.7) without being significant (p-values 0.1 - 0.3)? I think this might be a real biological finding but due to sample size (n=28) underpowered to be statistically significant. What do you think? Still relevant to further study these modules? 

Thanks for your advice. 

 

wgcna • 893 views
ADD COMMENT
0
Entering edit mode

I think without more power you will have a hard time convincing people, maybe you can find some public data to increase your power?

ADD REPLY
0
Entering edit mode

Well, that depends on how one is thinking to use those correlation values if it is just for deciding which modules analyze it is not so important. If you want to claim that those genes are related to X disease/variable/drug, then it should be looked carefully. However remember that the p.value just indicates which is the probability of happening event randomly. It can be common to have such event and still be significant.

ADD REPLY

Login before adding your answer.

Traffic: 782 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6