gaobio
commented
2 years ago i, and thanks for this tool!
In this case, copykat failed to tell them apart. It happens when the aneuploidy in tumor cells is very weak comparing to random noise. I added some confidence cutoffs, but I clearly haven't found a better method to always classify this situation.
Hi, and thanks for this tool! I'm running copykat on a dataset that contain tumor and normal epithelial cells, together with other normal cell types. I have an annotation of the cell types and I have ran the copykat analysis giving the known normal cells as normal reference in the fuction as a vector of the cells names. Now ,the copykat analysis returned to me an annotation in which the cells i given as normal are annotated as aneuploid. It is possible this kind of output happens? Maybe is due to sample characteristics? i double checked the colnames of the count matrix and the vector i give as normal, and everything looks fine, moreover the same analysis on Dataset of the same kind give to me more "reasonable" output.