I have gene expression data from RNAseq, specifically: log2(x+1) transformed RSEM normalized counts.
How can I convert these data from multiple samples to determine fold change in gene expression?
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
3
You should use a proper statistical framework for RNA-seq dfferential analysis (which includes FC calculation). Standard tools for this are (among others) edgeR or DESeq2. You could use tximport to import RSEM outputs into R and then use its output for e.g. DESeq2. The linked manual provides example code for this. Please read the manuals of edgeR or DESeq2 towards the respective downstream analysis. Both are extensive and compreshensive.
-
$\begingroup$ Thanks a lot for this information! So, just to confirm, that means that there is no 'simple formula' that can convert RSEM to fold change? Thanks again. $\endgroup$ Mar 7, 2020 at 23:28
-
2$\begingroup$ No. By the way, fold changes without statistics are basically meaningless. Just run it through DESeq2 and then take the results for your analysis. $\endgroup$– user3051Mar 8, 2020 at 10:44
-
$\begingroup$ Thank you very much. I will give this a try. $\endgroup$ Mar 8, 2020 at 19:54