Hello, I would like to know if the DESeq2 normalization is suitable to compare bulk RNA-seq samples with pseudo bulk single cell RNA-seq samples. In particular, I would like to sum the counts of the different single cell clusters, put these pseudo bulk sample with the bulk samples and normalize them in order to compare the expression of the genes without the differential expression analysis since there are no replicates. Is the DESeq2 normalization a good choice in this case? Thank you.

It sounds reasonable to me. I think median ratio works fine when there are genes that are expressed across all the samples. If the pseudo bulk samples have such a set of genes that overlaps with the bulk, then I think it would work.