search

heiniglab / scPower

Experimental design framework for scRNAseq population studies (eQTL and DE)
45 stars 5 forks source link

Is it possible to add other cell types? #23

Closed pcm32 closed 1 year ago

pcm32 commented 1 year ago

How would one go about doing that? Thanks.

KatharinaSchmid commented 1 year ago

Hi,

Thanks for your interest in our tool. Yes, it is possible to add other cell types, our package allows fitting and applying a custom-made set of cell type priors. In the vignette introduction-scPower Part 2 is a step by step explanation how to fit the priors. For this, it is necessary to have a count matrix containing multiple cells of the cell type of interest measured with the same technology. In case you want to model also the dependency on the read depth, it is additionally necessary to include not only the original count matrix but also a few subsampled count matrices (in the example I generated 3 subsampled matrices). In case you want to stay with the same read depth as in your example count matrix, a simplified version of the workflow is sufficient (subsection „Simplified gamma model without optimizing the read depth“).

Please let me know in case you have any questions or issues with the vignette. I am also always happy for suggestions how to improve the package.

In general, I would even recommend to get as closely matched priors as possible in order to get the best possible estimates. So fitting custom-made priors is a good idea to improve the experimental design.

Best regards, Katharina