Alternatives to CNVnator?

Question

[QVINTVS_FABIVS_MAXIMVS]

I think I have put over 36 man hours in trying to get CNVnator to install. I have tried nearly every avenue, conda, environments, etc... and still cannot get it to install.

I am at my wits end, what are comparable CNV calling algorithms for whole genomes that use read depth. Basically I'm looking for a CNVnator clone that can install easily.

I am using a server that does not give me root privileges, if I had this I wouldn't have this issue.

Thanks

[German.M.Demidov]

ClinCNV: https://github.com/imgag/ClinCNV/blob/master/doc/germline_CNV_analysis.md

[astulaaa]

Any suggestions specifically for long read (nanopore) non-cancer CNV calling? the only option I came across was nanogladiator

[ATpoint]

Please do not use the answer field for questions.

Note: Answers should ONLY be used to respond to the original question at the top of this page!
Use the ADD COMMENT or ADD REPLY buttons above to respond to someone, ask for clarifications, request more details etc,

[German.M.Demidov]

but since you've already asked - there are many well-established tools, I personally like sniffles

Answer 1

[jared.andrews07]

I have been pleasantly surprised by CNVkit. Easy install via pip, simple to run, good plotting features, can handle arrays or wgs. I've been using low coverage wgs, and it works quite well.

[meilan.he]

How low is your coverage, I have 3x to 10x WGS data generated from tumor FFPE tissues without normal controls, and exploring DNAnexus platform to do CNV analysis, they have both CNVkit and CNVnator, CNVkit recommend to use at least 10x coverage while CNVnator is sensitive for low coverage WGS data, do you have suggestion for my case especially regarding if I need or where to find a normal controls, thanks

[jared.andrews07]

I had about 10x coverage and used controls from the 1000 genome project. I just grabbed ~20 samples and combined them into a normal reference in CNVkit. Not sure how CNVnator works on that front, but it seemed to work well enough for CNVkit for samples where I didn't have a matched normal control.

[German.M.Demidov]

I would say for lower coverage data you have to bin your genome into larger regions, then CNVkit can work just as good as for higher coverage (but with lower resolution). As far as I know CNVnator is designed for Germline samples - I'd discourage usage of it in favor of CNVKit.

[jared.andrews07]

Oh yes, this is a great point. I found 100kb bins worked fairly well for 10x coverage, but you might have to go to 500kb or 1 Mb bins for your lower coverage samples.