Closed ebecht closed 5 years ago
I wouldn't worry about it.
Also, I don't know where these files are. Perhaps @MikeDMorgan might still have them, but I gave up trying to maintain the code in this repository. It was easier to make a new repo, as mentioned in the README
. That said, the new repo doesn't bother to define cell types for the unannotated data sets; mostly because the clusters (before and after correction) were so clear, it didn't really matter which clustering method you used.
Ok, I'll go with what I have and double check everything then. Thanks!
Hello and thank you for having this public repo! I have run the various scripts in the
/Pancreas/
folder as I am interested in studying how batch effects impact DEG methods. However I would like to make sure that I am getting the exact cell type labels you generated (as maybe I have slightly different package versions...). Although most appear to concur I also observed some discrepancies in the heatmap of marker genes for the clusters I obtained.Therefore, would you mind sharing the
*_metadata.tsv
tables that are inMNN2017/Pancreas/Data
so that I can make sure I have the right cell type labels?Thank you for your help and efforts!