DE analysis for single cell data with strong batch effect or different sequencing depth

Question

[Cheng Wei]

Hi everyone,

I am trying to do DE analysis between data downloaded from SRA, and they possess either very strong batch effect or very different sequencing depth. Wilcoxon rank sum test is probably the most popular test and one of the most effective. However, I am not sure whether it is effective for data with strong batch effects, since batch effects will likely cause the wilcoxon test ot be ineffective. I know of other packages such as MAST which can account for batch which may be effective , but can anyone share some light on whether my take on wilcoxon is correct, or share some papers that compares DE analysis methods for batch data?

---