Comparing groups with paired samples by using limma
1
0
Entering edit mode
annadv77 • 0
@annadv77-19831
Last seen 5.1 years ago

Dear All,

I have the following experiment: There are 3 donors (D1, D2 and D3), 2 treatment types (A and B) and 2 growth conditions (B and M). Each donor contributed samples for both treatments and both growth conditions (AB, AM, BB, BM).

I am hoping to examine the difference between growth conditions for each of the treatments separately (AB vs. AM and BB vs. BM) but we are also interested in the comparison between AB and BB. If all samples were obtained from different donors, I would have done the analysis in the following way:

design<-model.matrix(~0+factor(c(rep(1,3),rep(2,3),rep(3,3),rep(4,3)))); colnames(design)<-c("AB","AM","BB","BM");

contrast.matrix<-makeContrasts(AB-AM, BB-BM, AB-BB,levels=design) fit<-lmFit(dataset,design); fit2<-contrasts.fit(fit,contrast.matrix); fit2<-eBayes(fit2);

However, since there is also pairing of the samples based on the donors (from each donor we have AB, AM, BB, and BM samples), I understand that I need to add that information into the analysis, and I am not sure how to do that.

I will greatly appreciate any advice.

Thank you very much for your help!

Sincerly, Anna

limma paired analysis • 666 views
ADD COMMENT
0
Entering edit mode
@gordon-smyth
Last seen 3 hours ago
WEHI, Melbourne, Australia

Just add + Donor to the model formula.

ADD COMMENT
0
Entering edit mode

Is it also necessary to do duplicateCorrelation in addition to the blocking?

ADD REPLY

Login before adding your answer.

Traffic: 541 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