Entering edit mode
8.2 years ago
moxu
▴
510
edgeR requires the input of raw counts of RNA-seq WITHOUT ANY kind of normalization. But, it seems reasonable to me for the raw counts to be normalized by ERCC spike in, right? BTW, how do you do ERCC spike in normalization? And what's the algorithm to do the ERCC normalization?
Thank you!
Do you mean edgeR can handle ERCC normalization internally? That sounds great.
Thank you!
BTW, I used RSEM to get the sequence count, and the output of RSEM is not integer but with decimal points such as 4.26, 1.91, etc. Will this affect the performance of edgeR?
Thanks a lot!