Suggestion for improvement: VAF clustering

[caravagn]

Hi, nice tool.

This is just a comment and suggestion for improvement, since you reference ref.14 and neutrality etc.

Why do you perform VAF clustering (mutCluster) using a Gaussian mixture, instead of using a more appropriate model like MOBSTER? Can't you interface your tool to other tools etc and obtain the same results (i.e., mutation clustering)?

ref: Note that, “low VAF/CCF” populations might be a mixture of subclones mutations represent admixed polyphyletic lineages 14.