search

digitalcytometry / ecotyper

EcoTyper is a machine learning framework for large-scale identification of cell states and cellular ecosystems from gene expression data.
Other
184 stars 42 forks source link

Issue with cell type specific genes when performing scRNAseq discovery #63

Closed dkwon6 closed 1 year ago

dkwon6 commented 1 year ago

I have been trying to run EcoTyper scRNAseq discovery on a scRNAseq dataset. When I set the config Filter genes to "cell type specific", I get the following error:

Step 2 (cell state discovery on correrlation matrices): Calculating correlation matrices... Warning message: There are more than 2500 single cells available for cell type 'Epithelial.cells'. Subsampling to 2500 cells.

Filtering 'CD4.Naive' profiles for cell type specific genes... ---> Checking zero-variance data... ---> Total number of variables: 2500 ---> WARNING: 2500 variables found with zero variance ---> Maximum number of splits: floor(n/2) = 0 ---> WARNING: number of splits nSplit > 0 ---> WARNING: using maximum number of splits: nSplit = 0 Error in fastCor(scaled_data) : invalid nSplit: 0 Execution halted Error in RunJobQueue() : EcoTyper failed. Please check the error message above! Execution halted

However, when I manually set the Filter genes to a number (e.g. I've done 1000 and 2000 in my analyses), it runs without any issues.

I'd be happy to email an anonymized subsetted scRNAseq data if needed.

Thanks

BALuca commented 1 year ago

Hello,

Thanks for your interest in EcoTyper. Would it be possible to send us a reproducible example at ecotyper.stanford@gmail.com?

Best, The EcoTyper team