search

martinjzhang / scDRS

Single-cell disease relevance score (scDRS)
https://martinjzhang.github.io/scDRS/
MIT License
98 stars 11 forks source link

use scdrs in a tissue seemlingly unrelated with the disease #92

Closed Davidseye closed 1 month ago

Davidseye commented 1 month ago

when i use scdrs in a tissue seemlingly unrelated with the disease, such as , if i use scdrs to lung/kidney with the IBD statistics, and some cells are enriched and the p-value is <0.05, does that make sense in a biological way?

martinjzhang commented 1 month ago

Hi, did you apply multiple testing correction? https://en.wikipedia.org/wiki/False_discovery_rate#History

Davidseye commented 1 month ago

FDR < 0.05? yeah, it passed

martinjzhang commented 1 month ago

there are immune cells in lung/kidney. if those are the associated cells, it is biologically plausible. If the cells are not immune cells, the discoveries may be spurious.

Davidseye commented 1 month ago

thanks, i got it. Appreciating for your prompt reply,thanks again!