Copy Number Variation Tools

Question

[nayshool]

Hi,

We are searching for the best tool for detection of Copy Number Variants and/or Structural Variants from exome seq. We would like to incorporate that tool into out current pipeline. Which are the best tools for exome seq available as for today?

thank you

Omri

[GenoTechUser]

Hi Guys,

I'm new here

Can someone please provide an update to the tools or even publications (if any) to be used to detect CNVs from WES data?

Thank You

[German.M.Demidov]

hi, there are publications linked in this topic, did you read them? an update to which of them you need?

[geocarvalho]

Check it out

https://github.com/geocarvalho/sv-cnv-studies

[GenoTechUser]

Thank you for the lead. Will read on the available literature.

Answer 1

[Eric T.]

The bcbio-nextgen documentation describes a pipeline for structural variant calling that is consistent with best practices. Here are the callers that are suitable for exomes:

• Lumpy, Manta, and DELLY for paired end and split read calling
• CNVkit for read-depth based CNV calling
• WHAM for association testing

Answer 2

[geocarvalho]

I'm trying to study Structural Variants in Genome first, but I already noted some softwares for Exome: cnvScan, CoNIFER, exomeCopy, CONTRA, EXCAVATOR2, CODEX, CLAMMS.

Answer 3

[Charles Yin]

I have developed an algorithm to compute the copy number of DNA repeats, but not sure if the algorithm can fit your need. You can check the paper and its source code. [Yin, C. (2017). Identification of repeats in DNA sequences using nucleotide distribution uniformity. Journal of Theoretical Biology, 412, 138-145.], Python source code

Answer 4

[markus.riester]

There are a couple of good reviews with links to all important tools, for example a fairly recent one is https://doi.org/10.1093/bib/bbv055.

There is no gold standard and different tools were optimized for very different scenarios. Some tools require matched normals. Some use pool of normals for better coverage normalization. Some support off-targets reads for assays without tiling probes. Some are tumor purity and ploidy aware and provide integer copy numbers. Some tools work better in high-throuput settings than others. Some work better in clinical samples (low purity FFPE samples for example) than others.

A few recent tools not mentioned in the review or in the previous posts: CopywriteR, EXCAVATOR2, GATK4 beta, Sequenza, Facets, PureCN.

Answer 5

[Titus]

I used XHMM which pretty easy to use, looks working well but you need at least around 50 samples to have strong statistics and reduces false positive. As you can see there is a lot of software and no good demonstration (to my knowledge) of the best one. Even this publication https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3484655/ is a bit old i found it a good abstract of the diversity/algorithm.