DEG analysis of RNAseq TCGA data AFTER batch effect removal !!!

Question

[mohammad]

Hi everyone

I performed batch effect removal using the swamp package (kill.pc command). After performing this method, my raw count data contained negative values. Now I want to perform DEG analysis using DEseq2, but I faced the following error:

Error in DESeqDataSet(se, design = design, ignoreRank) :    some
 values in assay are negative

How I should solve this problem? Do I need to another method after batch effect removal?

Answer 1

[ATpoint]

Please read the manual. DESeq2 expects raw counts, and batch removal is accomplished by adding the batch information into the design. Corrected counts as input are not supported.

[mohammad]

Thank you very much. Would you please suggest is the best alternative method for DEG analysis in this situation?