Variants inside and outside of target region - what's the difference?

[NikTuzov]

Hello:

I am trying to understand whether there is some drastic difference between on-target and off-target reads. I assume that even if there is a true variant outside of target region, it is not of interest anyway. Apart from that, is there any difference between the two sets of reads and the variants detected by such reads?

In terms of (quantitative) metrics that can be computed for VCF files, is there any difference between the variants inside and outside of target region? For instance, if I were to compute an average depth across variants inside the target region, will it be higher than the depth outside the target region?

Thanks in advance,
Nik

[Brian Bushnell]

I assume you are talking about exome-capture?

There is no important difference between on- and off-target reads. Variants outside target regions may be of interest, though it's less likely; and of course 99.99% of variants inside the target are of no interest, but presumably they will tend to be more interesting than the ones outside the targets. Note that the gene annotations are continuously evolving as new genes and isoforms are discovered, while baits for capture kits may be based on annotation from many years earlier.

If the capture was successful, then yes, the coverage will be much higher within the targets than outside.