Question: ANOVA-like test via treat() in limma
0
gravatar for k.vitting.seerup
5 months ago by
European Union
k.vitting.seerup30 wrote:

Is it possible to do an anova style test (testing multiple contrast jointly) when using treat() instead of eBays() ?

Would be very useful to test against the log2FC cutoff across all conditions.

anova limma • 164 views
ADD COMMENTlink modified 5 months ago by Gordon Smyth36k • written 5 months ago by k.vitting.seerup30
Answer: ANOVA-like test via treat() in limma
3
gravatar for Gordon Smyth
5 months ago by
Gordon Smyth36k
Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia
Gordon Smyth36k wrote:

No, you can't use treat() to do an F-test. I'm not even sure what a fold-change threshold would mean for an F-test. Fold changes are only defined for individual contrasts, not for an F-test, which in effect tests all possible contrasts.

Of course, if you just want a gene to satisfy a treat() test for several different contrasts, then you could just run treat() for each of those contrasts. The decideTests() function provides an easy way to select genes satisfying several contrasts at once.

ADD COMMENTlink modified 5 months ago • written 5 months ago by Gordon Smyth36k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 342 users visited in the last hour