Hi,

I have no code to provide . Instead I have a background-related question.

What is the best/most feasible way to integrate RNA-seq data of different sourceS(cell vs. tissue) to further conduct unsupervised clustering analysis? (especially from the mathematical perspective)

Is it feasible to combine raw counts, do vst normalization with batch effect removal(blind =FALSE) and then conduct unsupervised clustering?

Or should I normalize 2 data sets individually, combine , and remove batch effect with a corresponding package(harmony, combat, limma).

Or shoul I combine raw counts, do batch effect removal, normalize or zscore scale?

Best!

``` limma() Deseq2()

Your question is off-topic for this site. You might try over on biostars.org instead.