search

arq5x / lumpy-sv

lumpy: a general probabilistic framework for structural variant discovery
MIT License
309 stars 118 forks source link

All INV have "++" as the strands in the output of vcfToBedpe #264

Open weishwu opened 6 years ago

weishwu commented 6 years ago

I converted LUMPY output VCF to BEDPE using vcfToBedpe in the lumpy package. All variants of type "INV" have strands "+ +". Is this a bug?

For example, should this one be "- -"?

1       45398796        45398805        1       45399208        45399228        386     .       +       +       INV
     .       SVTYPE=INV;SVLEN=414;END=45399219;STRANDS=++:6,--:19;IMPRECISE;CIPOS=-8,0;CIEND=-10,9;CIPOS95=-8,0;CIEND95=-1,0;SU=25;PE=22;SR=3        GT:SU:PE:SR     ./.:25:22:3     ./.:0:0:0
ryanlayer commented 6 years ago

It's not really a bug, more of a limitation of BEDPE. As expected, Your variant had evidence with both ++ and -- strand configurations. In BEDPE we can only pick one, so we went with ++.

On Tue, Aug 21, 2018 at 3:10 PM weishwu notifications@github.com wrote:

I converted LUMPY output VCF to BEDPE using vcfToBedpe in the lumpy package. All variants of type "INV" have strands "+ +". Is this a bug?

For example, should this one be "- -"?

1 45398796 45398805 1 45399208 45399228 386 . + + INV . SVTYPE=INV;SVLEN=414;END=45399219;STRANDS=++:6,--:19;IMPRECISE;CIPOS=-8,0;CIEND=-10,9;CIPOS95=-8,0;CIEND95=-1,0;SU=25;PE=22;SR=3 GT:SU:PE:SR ./.:25:22:3 ./.:0:0:0

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/arq5x/lumpy-sv/issues/264, or mute the thread https://github.com/notifications/unsubscribe-auth/AAlDUdzYwhZ92fbtLRiOBpdAEHp9FyFqks5uTHc6gaJpZM4WGmVt .

weishwu commented 6 years ago

Thanks. I thought LUMPY would pick one that had more evidence.