I am doing gene expression profiling of 3 structures(A,B,C). I compared (A vs B&C), (B vs A&C), (C vs A&B). I found 1, 2 and 20 DEG respectively between groups comparison. Are there any problems with these results, do I need to remove non-coding RNA before DEG analysis? I tries to find document and script to do so but I couldn't find one. Can you help me solving these problems?
1 answer
Is it necessary in terms "must I do it" => no
Can it be helpful, yes. If you have low power and/or much dispersion and/or low sample size (that all is basically interconnected) then removing genes to lower the multiple testing burden can help. You could run filterByExpr() first, then do the testing and then remove all but the non-coding genes before calling topTags(). Be sure to do a QC, e.g. by PCA or MDS (see edgeR manual) to see why you have few DEGs.
Log in to answer this question.
-
How to combine different "groups" in DESeq2 constrast
written by biotrekker •
-
EdgeR, a few DEGs found between group comparison, Can I go to further downstream analysis like heat…
written by TM •
-
EdgeR, How to remove non-coding RNA before DEG analysis?
written by TM •
-
edgeR, how to write the model matrix for comparison?
written by TM •
-
DESeq2 multiple group comparisons
written by Bhumi •
-
How to pull pairwise comparison results after running DESeq() with more than two condition
written by imrankhanbioinfo •
-
DESeq2 contrasts multiple treatments two genoptypes
written by wd •
-
RNA-seq analysis design
written by illinois.ks
-
Filtering samples or not before normalisation for samples from several groups
written by Zhilong Jia
-
[DESeq2 multiple treatments vs. 1-by-1] Inconsistant p-adj
written by madkitty
