search

ga4gh / vrs

Extensible specification for representing and uniquely identifying biological sequence variation
https://vrs.ga4gh.org
Apache License 2.0
80 stars 34 forks source link

`sequence` in Sequence Location and LSE/RLE is confusing #440

Closed korikuzma closed 1 year ago

korikuzma commented 1 year ago

It's kind of confusing that sequence is referring to both the accession in a SequenceLocation and the biological sequence in LiteralSequenceExpression and ReferenceLengthExpression https://github.com/ga4gh/vrs/blob/a445074b1e8928fab846f32083f29ce58868abc1/schema/vrs-source.yaml#L405-L410 https://github.com/ga4gh/vrs/blob/a445074b1e8928fab846f32083f29ce58868abc1/schema/vrs-source.yaml#L504-L506 https://github.com/ga4gh/vrs/blob/a445074b1e8928fab846f32083f29ce58868abc1/schema/vrs-source.yaml#L527-L529

ahwagner commented 1 year ago

Agreed, will address this.

korikuzma commented 1 year ago

Hey @ahwagner , any updates on this? In a standup meeting yesterday, @larrybabb showed us this where SequenceLocation.sequence was replaced with SequenceLocation.sequenceReference. I wasn't sure if this was the official decision. I'm making changes in Variation Normalizer 2.0-alpha, but the digests will change again once the field name changes.

ahwagner commented 1 year ago

We are working to make VRS flexible for use in many different schemas. In @larrybabb's example, they have a project-specific schema that leverages VRS IRIs to point to other objects within their project-specific JSON document. Due to the flexibility of IRIs, resolution of such references should be handled separately. I don't think VRS-Python should worry about IRI resolution right now.

ahwagner commented 1 year ago

as for the field name, I think we should change this to sequenceReference.

korikuzma commented 1 year ago

as for the field name, I think we should change this to sequenceReference.

Okay, I will make this change in vrs-python

korikuzma commented 1 year ago

This has been resolved