search

GfellerLab / EPIC

Repository for the R package EPIC, to Estimate the Proportion of Immune and Cancer cells from bulk gene expression data.
https://gfellerlab.shinyapps.io/EPIC_1-1/
Other
74 stars 21 forks source link

other cells #3

Closed marceaf23 closed 5 years ago

marceaf23 commented 6 years ago

Hi,

The software is great but I am still confused on what cells are in other cells? can normal cells also be mixed in there with cancer cells or is it assumed that the "other cells" portion is all cancer cells?

jracle85 commented 6 years ago

Hello,

Thanks for your interest in EPIC. So the other cells correspond to all the cell types that are not part of the reference samples. If you use our defined reference profiles, we defined most of the cell types that should be found in a tumor (except the cancer cells and the normal cells from which the cancer developed as these other normal cells aren't much different from the cancer cells). Therefore it can be assumed that the "other cells" are a good proxy for the proportion of the true cancer cells.

If you use your own reference profiles or use EPIC on non-tumor bulk samples, then the "other cells" will reflect the various cell types that are present in this bulk but not defined by the reference profiles.

I hope this helps!

Cheers,

Julien