Filter by ONT indels quality #100
Comments
|
Hi @ilivyatan We have been improving the SNP-indel co-phasing internally for quite some sometime. The version in the develop branch has improved the SNP-indel co-phasing accuracy by ignoring/unphasing some unreliable indels internally in the phasing algorithm. Just you have to checked out that branch and compile by yourself at this moment. Having said that, we still can't guarantee all the false indels can be rule out. Your suggestion will be considered in our to-do list. Thanks. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi,
I appreciate the --indels option when using longphase on ONT data, since the indel calling of this LRS platform has a very high FP rate, so it's sometimes a good idea to disclude it from phasing. Nevertheless, in a clinical setting, sometimes the variant of interest is an indel and having it co-phased automatically with surrounding variants is useful.
Can I make a suggestion to add a flag for indel quality filtering, such that indels could be considered during phasing, but only if they pass minimum quality threshold?
I can also achieve this by pre-filtering the VCF before running phasing, just wondering if any similar request have come up from the field, and if there are other attempts to fine-tune the approach to indels during phasing?
The text was updated successfully, but these errors were encountered: