You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
fgbio and fgpyo provide implementations for a Template struct, collecting all alignment records associated with the same queryname and template, and a TemplateIterator, which can iterate over a queryname-grouped BAM and yield Template instances. It would be useful to carry these objects and patterns over to Rust.
To the extent possible and sensible, the API should be consistent with those implemented by fgbio and fgpyo.
See also this recent PR - records marked as both secondary and supplementary should be placed in the list of supplementary alignments
I think that on reflection @nh13 and I have a difference of opinion on that PR and feel that secondary-supplementary records should end up in the list of secondary alignments. This is how it's done in fgbio, and having thought more on it, I prefer that.
Summary
fgbio
andfgpyo
provide implementations for aTemplate
struct, collecting all alignment records associated with the same queryname and template, and aTemplateIterator
, which can iterate over a queryname-grouped BAM and yieldTemplate
instances. It would be useful to carry these objects and patterns over to Rust.To the extent possible and sensible, the API should be consistent with those implemented by
fgbio
andfgpyo
.Prior art
Template
templateIterator
Template
TemplateIterator
Suggested solution
An initial MVP to address this issue would implement:
Template
struct to contain R1 and R2 primary alignments, and lists of R1 and R2 supplemental and secondary alignments.TemplateIterator
that consumes a queryname-grouped BAM and yields an iteration ofTemplate
structsAdditional logic associated with
Template
, e.g. fixing mate info, may be addressed in later PRsThe text was updated successfully, but these errors were encountered: