Question: P-value correction on DEGS and Functional Annotation
gravatar for Rick
7 weeks ago by
Rick0 wrote:

Hi all,

I am using microarray analysis (e.g. Affymetrix) to identify differentially express genes (DEGs). I set two conditions to define the DEGs: (1) fold changes of >1.5 or <0.5; (2) q-value < 0.05 (I am using q-value to account for multiple comparisons).

Once I have the list of DEGs defined, and I am doing functional annotation and enrichment analysis (e.g. gene ontology and pathway enrichment analysis) with DAVID. My question is: to define what functional groups are significantly enriched, should I use p-values or q-values in the second round of analysis?

My intuition says that, since I already selected the DEGs by using q-value, the resulting list of genes has already been corrected for multiple testing. Thus, I should just focus on p-values for gene ontology and pathway enrichment analysis.

I have gone through the literature, and I have read papers using both strategies: DEGS with q-values and functional annotation with p-values, and others using q-values always.

Any suggestion?


gene • 168 views
ADD COMMENTlink modified 7 weeks ago by rpolicastro4.0k • written 7 weeks ago by Rick0
gravatar for rpolicastro
7 weeks ago by
Bloomington, IN
rpolicastro4.0k wrote:

You should still use the q-value. When doing an enrichment analysis, you are running a test on every term (such as the commonly used hypergeometric test for this application). Since there were multiple tests run for multiple terms, the multiple comparison problem is encountered, which makes p-value correction a must.

ADD COMMENTlink written 7 weeks ago by rpolicastro4.0k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 2102 users visited in the last hour