Batch correction for DE analysis

Question

[Yoonji]

Hi,

I have tried DE analysis using total and mRNA seq data (5 total RNA data and 5 mRNA data) and making the MDS plot in which 5 total RNA data are included in 2 groups (A, B). In the MDS plot, the samples were divided according to the sequencing method. So, I want to correct the sequencing method effect when I obtain the expression value and make MDS plot. Additionally, I have tried doing it with limma::removebatcheffect function, but there were no changes. How should I correct it?

[ATpoint]

Please add some code and a table that illustrates which samples come from which batch, and then MDS plot.

[rpolicastro]

If samples from both A and B were sequenced with both sequencing modalities, you can include the method as a covariate in the DESeq2 or edgeR formula (~ group + method). When doing contrasts between groups this will provide you the DEGs while controlling for differences in sequencing method.