wukevin / babel

Deep learning model for single-cell inference of multi-omic profiles from a single input modality.
https://www.pnas.org/content/118/15/e2023070118
38 stars 16 forks source link

rna_genes less genes than the input data #7

Open EsperanzaDai opened 3 years ago

EsperanzaDai commented 3 years ago

Hi Kevin,

I have trained a model with ./bin/train_model.py codes and raised the flag --nofilter. However, in the rna_genes.txt, I have 944 genes lost. And I checked the source code in the train_model.py and could not find the exact place where genes are filtered. Could you kindly inform me how those genes are filtered?

Thanks a lot! Yam

wukevin commented 2 years ago

Hi Yam - this shouldn't be happening. Can you pleas share the exact command you used (no need to share the files if you aren't comfortable doing that) and the log messages? This should help us narrow down what is happening.

Also, as a sanity check, are these 944 genes all cases where there is 0 expression across all cells in your data?