calling the CNV in tumor-only samples (WGS and WES)

Question

[Bogdan]

Dear all,

please, which algorithm(s) would you recommend for calling the CNV in tumor-only samples (obtained by either WES or WGS technology ?

I would gladly use GATK4, although hmmm ... we do not have so many samples in order to generate a PoN (a panel of normals).

many thanks,

bogdan

Answer 1

[Eric T.]

CNVkit can do CNV calling from tumor-only samples, either in isolation or using a small pool of unrelated normal samples that were sequenced with the same protocol.

[Bogdan]

Dear Eric, great to hear from you ;) if it is OK by you, will get back to you by email, as we have a few questions about CNVkit ;).

[Eric T.]

Sure, happy to help. Also try CNVkit's issue tracker if you think there's a bug other people might be experiencing.

Answer 2

[Bogdan]

Thanks, Eric. will write to you as soon as i have a free minute; essentially CNVkit did run on our samples, although i would have to talk with you about the output. Thanks again !