Need help with a contrast design for limma with a missing group?
1
0
Entering edit mode
Ahdee ▴ 50
@ahdee-8938
Last seen 6 days ago
United States

This is an RNAseq experiment where I'm trying figure out how to setup the contrast and I can't seem to figure it out. I have 3 groups.

  1. ( Normal FGFR1, Normal ACO2 ) - ( High FGFR1, Normal ACO2 ) ==> this comparison will give me all FGFR1 dependent genes.
  2. ( High FGFR1, Normal ACO2 ) - ( High FGFR1, low ACO2 ) ==> this comparison will give me all ACO2 dependent gene in the presence of high FGFR1

the issue is I also want is to find ACO2 dependent genes in the presence of normal FGFR1 as well. However, I do not have a ( Normal FGFR1, low ACO2) group.

Thus the question: is there way to set up a contrast between the 3 groups to answer this? ACO2 dependent genes in the presence of normal FGFR1.

For example, would this contrast address this?

( Normal FGFR1, Normal ACO2 + High FGFR1, Normal ACO2 ) /2 - ( High FGFR1, low ACO2 )

limma VOOM • 296 views
ADD COMMENT
0
Entering edit mode
@gordon-smyth
Last seen 2 hours ago
WEHI, Melbourne, Australia

Sorry, I can't give advice regarding scientific conclusions specific to a particular experiment. I can only give advice on more general statistical and computing issues relating to limma.

ADD COMMENT
0
Entering edit mode

Gordon Smyth ok thank you for taking the time to answer the rest of question on this forum though; much appreciated.

ADD REPLY
0
Entering edit mode

I can add one comment. The contrast you propose does just what is appears to do, and you can see that it will give you some FGFR1 effect as well as the ACO2 effect. Only you make the judgement as to whether that is useful for your purposes. Obviously a contrast can't make up entirely for a missing experimental group.

ADD REPLY

Login before adding your answer.

Traffic: 409 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