consensus SV calls from multiple SVcaller?

[bioteksampath]

Hi,
Thanks for the combiSV SV tool, it seems intersting and helpful.
I have vcfs called using cuteSV, Sniffle and NanoSV using ONT reads and I like to get consensus SV (Same SV called by two are more SV caller). I understand that your script comined SV from multiple caller based on the depth.
But, I was wondering is there possible to get consensus calls for a line using multiple callers.

Thanks in advance
Sam

[ndierckx]

Hi,

Depth is just a parameter you can change but it is not the main factor to call a SV
It will look at consensus calls supported by multiple callers, but will also include some calls only supported by one caller
These calls depend on the type of SV, the caller, the size, genotype,... and was determined statistically from extensive benchmark results.

If you already have the vcf files, you can try combisv easily, just takes a few seconds to run

[bioteksampath]

Thanks, so, is that possible to get only consensus calls supported by minimum two callers?
I also have another question, is there is an option in combiSV to merge VCF calls from multiple lines to make a meta SV dataset?

[ndierckx]

"Thanks, so, is that possible to get only consensus calls supported by minimum two callers?"

For the moment not, but I can add an option so that you can select that, will do it with the next update
It would reduce the recall though...

"is there is an option in combiSV to merge VCF calls from multiple lines to make a meta SV dataset?"

Not at the moment, but you would be interested in merging similar SVs as one, but also leave the SVs unique to one line in the meta SV dataset? Could add that option too, would be easy to include