search

arq5x / lumpy-sv

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

pe+sr question #263

Open LEE1214 opened 6 years ago

LEE1214 commented 6 years ago

Hi, In the lumpy paper, the performance measurements for LUMPY were based on pe and sr alignments. So in a vcf file, how do we know that this result is supported by pe and sr. If the info contains the keyword ‘imprecise' in the vcf file, does this result get no pe+sr support? Thanks in advance! image

ryanlayer commented 6 years ago

Then next three fields in the info give this information. pe is given by PE, sr is given by sr, and SU gives the sum of the PE and SR.

On Aug 10, 2018, at 8:28 AM, LEE1214 notifications@github.com wrote:

Hi, In the lumpy paper, the performance measurements for LUMPY were based on pe and sr alignments. So in a vcf file, how do we know that this result is supported by pe and sr. If the info contains the keyword ‘imprecise' in the vcf file, does this result get no pe+sr support? Thanks in advance!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

LEE1214 commented 6 years ago

Then why PE>0, SR>0, there is still the 'imprecise' keyword in info. image

ryanlayer commented 6 years ago

IMPRECISE tags SVs with breakpoints that are not resolved to a single base pair. Check out the lumpy paper for a description of how these breakpoints are defined.

On Aug 11, 2018, at 2:56 AM, LEE1214 notifications@github.com wrote:

Then why PE>0, SR>0, there is still the 'imprecise' keyword in info.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.