Formula to exclude certain clusters?

[yabeykoon]

Hi, Thank you for SCEPTRE. It has helped me immensely to analyze my high MOI perturb-seq data! From my single cell run, I have some cells that seemed to have bulked during the capture, that form a few low quality clusters in my UMAP when I analyze my cellranger output using Scanpy. Is there a way to eliminate certain low quality clusters by adding them as covariates in the SCEPTRE pipeline? I would appreciate an example code if so.

[ekatsevi]

Thank you for SCEPTRE. It has helped me immensely to analyze my high MOI perturb-seq data!

That's great to hear!

Is there a way to eliminate certain low quality clusters by adding them as covariates in the SCEPTRE pipeline?

The best way to do this is to extract a (one-based) integer vector of indices of cells in low-quality clusters from Scanpy, and then to pass this to the additional_cells_to_remove argument of the run_qc() function.

[yabeykoon]

Thank you! Is it possible to add this step to the Nextflow trans analysis? I don't see where I could add this argument in the Nextflow pipeline.