Closed gilhornung closed 6 years ago
Hi Gil,
yes, that is the correct behavior, for now. Variants (even if closely located) are not phased, but reported as separate entries in the output VCF. We plan to add support for phasing in the future, but there is no timeline for when this will be implemented.
The latest release now reports multiple consecutive variants as single events (via VCF replacement operations). Cheers!
Thanks!
On Fri, 20 Jul 2018, 7:28 AM Giuseppe Narzisi, notifications@github.com wrote:
The latest release now reports multiple consecutive variants as single events (via VCF replacement operations). Cheers!
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/nygenome/lancet/issues/20#issuecomment-406485772, or mute the thread https://github.com/notifications/unsubscribe-auth/AYRgKdvK1UQWnl4Ga1-WjhZk4bu38XZ-ks5uIVxVgaJpZM4Tjtrv .
Hi Giuseppe,
I noticed a couple of cases where there is a di-nucleotide variant which is reported as two single-base variants. This is common in many variant callers, but I did not expect it to happen after the local assembly.
Below are the relevant lines from a VCF. bam files to reproduce it are
normal_CADPS2.bam
andtumor_CADPS2.bam
from this link