Canvas for non-tumorous data?

[P4sm]

Hi

I was just wondering, can you use Canvas to detect copy number variants from any WGS data? The manual and publication talks almost explicitly about tumor and normal samples. Shouldn't e.g. Somatic-WGS (CNV calling of a somatic sample from whole genome sequencing data) mode be apt for any human somatic-WGS data?

[StephenTanner]

Yes, Canvas can call germline copy number variants from normal (non-tumor) WGS data.

The manual and the article focus a lot of attention on the somatic case in part because it is more complicated (the signal is noisier, tumor genomes can be more extensively rearranged, tumor + paired normal sequencing is needed,).

[P4sm]

Thank you for your quick reply. So Canvas supports only germline copy number variant detection for non-tumor WGS data, but not somatic? So let's say I have just some non tumor somatic WGS sample (test) and then some normal (control) sample. Can you recommend some other tool for that purpose?

[sivakhno]

Names of Canvas workflows represent most-popular use cases: germline for single-sample and tumour-normal for matched samples. You can use the latter for any paired samples including non-tumour, though TN workflow was optimized for detection of large CNV and extensive rearrangements observed in cancer.

What Canvas still doesn’t do is a join detection of germline and somatic variants, i.e. the variants in the control sample are not detected prior to subtraction, though you can run the germline workflow first to detect them.