Question: Differential expression of one group agains all samples
0
gravatar for ofonov
2.7 years ago by
ofonov0
ofonov0 wrote:

Hi,

I would like to do differential expression analysis of one group of samples (miRNA) against all samples in the dataset (e.g samples A,B,C, against samples A...K ). At the moment I have tried to use DESeq2 for this purpure. I did not find description of such possibility in DEseq2 manual, however it did not give me any errors while I was doing that. My question is if it is OK to use DESeq2 in such way, or there are perhaps some better options?

 

ADD COMMENTlink modified 2.7 years ago by Gavin Kelly560 • written 2.7 years ago by ofonov0

Can you tell me more about these samples? Is there any structure?

ADD REPLYlink modified 2.7 years ago • written 2.7 years ago by Michael Love23k
Answer: Differential expression of one group agains all samples
0
gravatar for Gavin Kelly
2.7 years ago by
Gavin Kelly560
United Kingdom / London / Francis Crick Institute
Gavin Kelly560 wrote:

I think it would be better to do a contrast of A,B,C against samples D...K  If you include e.g. sample A in the model twice (as it appears you are doing), then the residuals aren't going to be independent, which is one of the key requirements of regression. If there's a valid reason for including the sample in both groups (and I can't think of one), then there's no simple alternative to DESeq that will deal with correlated residuals any better.  You'd need to get statistical advice on how best to deal with the correlation in your specific case, but the easiest route to removing the problem is to split the samples disjointly.

 

ADD COMMENTlink written 2.7 years ago by Gavin Kelly560
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: 115 users visited in the last hour