CNV caller software

Question

[Bogdan]

Dear all,

any suggestions regarding a good algorithm to use to call CNV (copy number variants) in the cancer genomes ? many thanks !

-- bogdan

[Jared Evans]

Do you need a CNV caller for whole genome sequencing data? Or exome sequencing data?

Do you have a paired normal sample to compare with? Or just a tumor sample?

[Bogdan]

Thank you Jared. We would need a CNV caller on WGS data, we do have paired normal-tumor samples, and any suggestions would be welcome. Many thanks !

Answer 1

[cbst]

ASCAT, FACETS, Theta, ABSOLUTE, ADTex, CopyCat, copynumber, depending on the type of data you have.

I personally use ASCAT for array data, and it has been used in several articles as the gold standard to compare tools, and FACETS for exome sequencing, it was published this summer, and my first impression is quite good.

Answer 2

[Eric T.]

At UCSF's cancer genome sequencing service we've been using CNVkit with good results.

Answer 3

[EagleEye]

For array,

http://penncnv.openbioinformatics.org/en/latest/

Answer 4

[Jeffin Rockey]

Please see cn.MOPS also.

Jf

Answer 5

[ssv.bio]

GISTIC (http://portals.broadinstitute.org/cgi-bin/cancer/publications/pub_paper.cgi?mode=view&paper_id=216&p=t), DNAcopy (https://bioconductor.org/packages/release/bioc/html/DNAcopy.html)

Answer 6

[Bogdan]

Thank you all for suggestions ! Has anyone experience with HMMcopy and TITAN ? how well do these tools do ?