Search
Question: What's the difference between p.adjust and qvalue?
0
16 months ago by
tjiagoM10
tjiagoM10 wrote:

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?

modified 16 months ago by Guido Hooiveld2.3k • written 16 months ago by tjiagoM10
1
16 months ago by
Guido Hooiveld2.3k
Wageningen University, Wageningen, the Netherlands
Guido Hooiveld2.3k wrote:

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.

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?

1

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).