What's the difference between p.adjust and qvalue?
1
0
Entering edit mode
tjiagoM ▴ 10
@tjiagom-13123
Last seen 4.4 years ago

In the result of an enrichGO() we have two columns, p.adjust and qvalue. I guess p.adjust is the pvalue modified after considering FDR, but then what actually is the qvalue?

 

clusterprofiler • 3.2k views
ADD COMMENT
1
Entering edit mode
Guido Hooiveld ★ 3.0k
@guido-hooiveld-2020
Last seen 29 minutes ago
Wageningen University, Wageningen, the …

From the help page (?enrichGO):

pAdjustMethod: one of "holm", "hochberg", "hommel", "bonferroni", "BH", "BY", "fdr", "none".

Note: default = "BH".

 

qvalue: alternative approach/algorithm to calculate FDR/control for multiple testing. For more details see the package qvalue here.

Note: default value is 0.2. 

 

ADD COMMENT
0
Entering edit mode

Exactly, I saw that, but isn't that the same meaning as of an adjusted p-value? Because calculations of FDR are meant to change the pvalue, right?

ADD REPLY
1
Entering edit mode

Yep, the aim of all 'FDR algorithms' is the same, but how they exactly work differ between the algorithms. See e.g. this thread + links conceptual question about FDR, FDR adjusted p-value and q-value for an easy accessible overview, and confusion regarding nomenclature. The thread @ Biostars here also discusses this question (and links to the same BioC thread).

ADD REPLY
0
Entering edit mode

Ok, the package mentions that they are not creating adjusted pvalues. I will look with more attention, thanks.

ADD REPLY

Login before adding your answer.

Traffic: 487 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