search

cortes-ciriano-lab / savana

Somatic structural variant caller for long-read data
Apache License 2.0
43 stars 2 forks source link

Variant with POS=0 in savana vcf output #33

Closed waltergallegog closed 4 weeks ago

waltergallegog commented 11 months ago

Hi, during some tests with SAVANA I got the following variant:

GL000226.1  0   ID_147659_1 N   ]hs37d5:30138859]N  .   PASS    SVTYPE=BND;MATEID=ID_147659_2;NORMAL_SUPPORT=0;TUMOUR_SUPPORT=3;SVLEN=0;BP_NOTATION=-+;ORIGINATING_CLUSTER=359770c35c0b46b3b2b0c5342a7a9e3b;END_CLUSTER=4befddd8f90e4bad8644ae3df9686a0f;ORIGIN_STARTS_STD_DEV=0.47;ORIGIN_MAPQ_MEAN=31.33;ORIGIN_EVENT_SIZE_STD_DEV=0;ORIGIN_EVENT_SIZE_MEDIAN=1;ORIGIN_EVENT_SIZE_MEAN=1;END_STARTS_STD_DEV=0.47;END_MAPQ_MEAN=31.33;END_EVENT_SIZE_STD_DEV=0;END_EVENT_SIZE_MEDIAN=1;END_EVENT_SIZE_MEAN=1;TUMOUR_DP=12510,459;NORMAL_DP=7724,285    GT  0/1

Which has position 0. AFAIK, a value of zero for the POS field in VCF should not be possible, as VCF is 1-based. Perhaps SAVANA is calculating the position zero based (for some variant types) ?

BR Walter

helrick commented 4 weeks ago

Hi there,

I suspect this has something to do with the fact that this variant falls on a short alternate contig. I wasn't able to replicate this issue with the latest release of SAVANA, so am closing for now, but feel free to re-open the issue if you still encounter it in the latest version of SAVANA.

All the best, Hillary