Sig. cell type-disease associations and Sig. individual cells

[hopehan12]

Hi Martin,

Thanks for the great work! I finished the analysis and I am kinda stuck with interpreting the results. I have 0 significant individual cells (0 out of 252452 cells) but significant cell type-disease associations in some cell types (2 out of 12 cell types). I am wondering if this is because of generally higher disease scores across cells within that cell type?

Please just let me know, many thanks!

[martinjzhang]

You are correct. Cell type-level analysis can be more powerful if cells within a cell type have consistently higher disease scores (which may not be significant for individual disease cells).

[martinjzhang]

We plan to release a significantly more powerful version in a few months. Stay tuned.

[hopehan12]

Hi Martin, Thank you! Super helpful. I will keep an eye out for it!