Hi all,
I sequenced a blood cell type from blood and sputum fluids of the same patients with single-cell transcriptomics and I want to identify genes that drive disease severity (have it as both continuous and categorical values if needed). I will convert to pseudobulk RNA-seq counts and I wanted to ask whether the DIABLO with the origin (sputum/blood) as blocks or separate sPLS analysis for each compartment (and analysis of the intersect of genes) would be the right way to go. My hesitation stems from the fact that it is the same modality but the cell type origin is different and wanted to account for that in the prioritization of genes. Any explanation would be appreciated.