R scripts don't produce same RDS files about scrna.seq.datasets HOT 2 OPEN

Hi Pavlin, thanks for your question. All of the datasets were created from the scripts provided in the repository. In fact, this was done automatically, I wasn't involved. The only reason for this being different I see that it maybe a new version of R/Bioconductor/scater/limma package, in which calculateCPM function has been updated. calculateCPM I think is from the limma package. Last time the datasets on our website were generated on February 27, 2018, which was quite a long time ago, many things have been updated since then. So, my recommendation would be to use the scripts with the newest versions of all of the packages. You will have slightly different numbers, but I am sure the final results won't change dramatically. Hope this helps.

from scrna.seq.datasets.

I am experiencing the same issue. I don't think it's the calculateCPM function (which comes from scater). This is a pretty straightforward transformation (equivalent to t(t(counts)*1e6/colSums(counts))) and I can't find any record of it changing in the scater documentation.

All of the datasets were created from the scripts provided in the repository. In fact, this was done automatically, I wasn't involved.

I'm unable to run your scripts without the changes in ForrestCKoch@09074c1. Are these the same scripts you used?

from scrna.seq.datasets.