search

caravagnalab / revolver

REVOLVER - Repeated Evolution in Cancer
https://caravagnalab.github.io/revolver/
64 stars 25 forks source link

How to assign cluster id to samples if using binary data? #45

Closed sunandinisharma closed 2 years ago

sunandinisharma commented 2 years ago

I am using the binary input data for REVOLVER. Not sure how to assign the cluster id to patients? In my case, I am not using PyClone as I have Copy number information available and want to keep everything in binary format. Can I get some explanation about how to assign patients into clusters while preparing the input? Is it necessary to use the sub-clonal deconvolution using Pyclone or other tolls?

caravagn commented 2 years ago

Hi @sunandinisharma,

Say you have samples A, B and C in a patient, with 3 CNAs (X, Y and Z) observed

   A | B | C
X: 1 | 1 | 1
Y: 1 | 1 | 1
Z: 0 | 1 | 0

you clusters should be 2, one for X and Y, and one only for Z. This is because the rows are identical among and X and Y, and are different from the ones in Z.

The cluster label obviously does not matter.