How to obtain cancer cell fraction of each mutation in subclones of a single sample

Question

[fantastycrane]

Hi All,

I am working on a project focusing on phylogeny of a specific type of tumor using single sample from multiple patients. It seems REVOLVER (manuscript link) could serve this purpose. However, its input requires cancer cell fraction (CCF) of point mutations in subclones of each sample. I know that expands the R package characterizes subpopulations in a single tumor sample. Still, further analysis can not be performed due to lack of CCF information. Is there any way to get this kind of data for REVOLVER?

Answer 1

[gcaravagn]

Hi fantastycrane, the GitHub space for REVOLVER contains a wiki with some FAQ, and vignettes that you might want to look at if you decide to use the tool. Concerning your CCF-related question, to explain it shortly CCF estimates - and their clusters - are computed doing what we call "subclonal deconvolution", which requires 1) SNV 2) copy number calls, and 3) tumour purity estimation (contamination of non-tumour cells in the bulk sample). With that data there are several tools that you can use to compute the CCF values, and their clusters. Two tools that I like are pyClone and sciClone etc, for which you find support either here, or on the tool webpage.

So that bit of the analysis does not happen within REVOLVER because there are dedicated tools for it. Feel free to drop a line if you try that analysis and something does not work out for you. For questions regarding REVOLVER etc, you can also use the GitHub issue page.

REVOLVER Wiki: https://github.com/caravagn/revolver/wiki Issues: https://github.com/caravagn/revolver/issues sciClone: https://github.com/genome/sciclone pyClone: https://github.com/aroth85/pyclone