TPM or FPKM values are only good for comparison gene expression within samples, but further cross-sample normalization is still needed for differential expression analysis between conditions.
Agree with you. So for RNA-seq data, I used combat-seq in SVA package to remove the batch effect, which is working on raw COUNT rather than TPM or FPKM.
This is a fantastic lecture explaining batch effect. Thank you
Extremely helpful. Thank you!
Really nice video explaining batch effect
hi Prof. Liu, this is a very informative lecture. Thank you.
TPM or FPKM values are only good for comparison gene expression within samples, but further cross-sample normalization is still needed for differential expression analysis between conditions.
Agree with you. So for RNA-seq data, I used combat-seq in SVA package to remove the batch effect, which is working on raw COUNT rather than TPM or FPKM.
Thank you
This is a great lecture. Thank you
Good lecture. Thank you Dr Liu
Hi Dr Liu, great lecture series. Is there Lecture 7 in this playlist. Could not find it and so was wondering.
So what should i do.if i have 600 samples of condition 1 and 50.for condition 2?
I think it is usual for your condition. You can do as it had balanced sample size.