0
$\begingroup$

I was wondering if anyone had any advice on comparing two groups where the variance in once group is high and the variance in the other group is low.

Specifically, I am attempting differential gene expression analysis using EdgeR. I have been using the glmLRT function however while playing with other functions I discovered the BCV was 0.79. My understanding is that using a generalised linear model is probably not a valid method to compare these two groups (each with 5 replicates). Can anyone offer advice?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

So you have two groups, small sample sizes and different variances? What is the ratio between sample variances? You can find some advice from this list.

I would guess the distributions are far from normal too? I would maybe try some permutation test, some advice in this list.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.