Closed cantaloupe5 closed 3 years ago
Hi @cantaloupe5 If that would be the only type of data you have to address your biological question, I think you could apply NicheNet to generate some hypotheses, but I would indeed be cautious in interpreting the results.
Hello,
Thank you for creating such an amazing tool. CIBERSORTx can produce cell-type specific gene expression profiles from bulk RNA-Seq. The authors have demonstrated that their algorithm's results correlate well with FACS-sorted cell-type transcriptomes. As you have used NicheNet on FACS-sorted bulk-RNA Seq data in this paper, I was wondering what your thoughts are on using NicheNet on these "in silico" sorted cell transcriptomes. Of course, I imagine NicheNet's results will not be as reliable (as you previously said: garbage in, garbage out) but are there any reasons to avoid this technique altogether?
Thank you.