Search
Question: edgeR 2-way ANOVA testing
0
gravatar for martyna.broda
6 days ago by
martyna.broda0 wrote:

Hi all,

I've been struggling with my 2-way ANOVA analysis. I have 4 genotypes, lets name them A, B, C and D, and 3 time points: 0, 1 and 3. I need to test for time specific and genotype specific responses in my RNAseq data. I am quite new to bioinformatic field. How do I specify groups, do I use normalized counts or row counts? And how to actually do 2-way ANOVA analysis? All I found so far, was one-way...

Any help will be appreciated.

Martyna

ADD COMMENTlink modified 6 days ago by Gordon Smyth35k • written 6 days ago by martyna.broda0
0
gravatar for Gordon Smyth
6 days ago by
Gordon Smyth35k
Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia
Gordon Smyth35k wrote:

A 2-factor analysis is really just the same as a one-way analysis. The easiest way is to make a variable with 12 levels (A.t0, A.t1, A.t3, B.t0, B.t1 ...) and then make contrasts between the levels in the usual edgeR way. For example, to compare C to D at time 3, form the contrast D.t3 - C.73).

There is no need to change the matrix of counts in any way -- edgeR will do the computations for you.

The edgeR workflow provides an example of 2-factor analysis (cell type x pregnancy status):

https://www.bioconductor.org/packages/RnaSeqGeneEdgeRQL

(Click on the "html" link).

ADD COMMENTlink modified 6 days ago • written 6 days ago by Gordon Smyth35k
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 2.2.0
Traffic: 400 users visited in the last hour