Dear BioCommunities,
Nowadays, I want to utilize the pan-cancer dataset "EBPlusPlusAdjustPANCAN_IlluminaHiSeq_RNASeqV2" for DEGs analysis. Because of the normalized and batch corrected transformation, I would perfer to use limma::voom. However, the dataset contains negative value, which isn't accepted by limma::voom. Any advices for dealing with those negative values before DEGs analysis? Thanks in advance!
The values in the file appear to be estimated read counts, after some modification. However the values include both negative values and NAs, neither of which are acceptable by voom, or by any RNA-seq software that I know of.
You could perhaps set the NAs and the negative values to zero, but I don't honestly know how reliable that would be.
It is really the responsibility of the PANCAN people to explain to you how to process the data rather than for us here on Bioconductor to do so. I have tried to read their documentation but they do not entirely explain how the data has been processed. They say they have used their own algorithm "EB++" but that algorithm is not published or explained.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
It isn't clear what data you are trying to analyse. Please give a link to the data that you downloaded and explain what sort of data it is (counts, rpkm, tpm etc).
Thanks for your reply sir! Here is the data (EBPlusPlusAdjustPANCAN_IlluminaHiSeq_RNASeqV2.geneExp.tsv). And there is the data procession of this data. Thanks a lot sir!