SEP 010 -- simplify description of sequence features and sub-parts

[graik]

This proposal aims to clearly separate and simplify the description of part - subpart relations on the one hand, and the annotation of (genbank-style) sequence features on the other hand.

Proposal document: https://github.com/SynBioDex/SEPs/blob/master/sep_010.md

[graik]

Updated after very productive discussion at COMBINE breakout session. The current formulation of the SEP prescribes changes for SBOL 2.2 (adding and deprecating fields) and for SBOL 3.0 (removing fields and renaming SequenceAnnotation to SequenceFeature.

[jakebeal]

On discussion, the editors have determined that this is an SBOL 3.0 issue, because as soon as it is possible to write a file that does not have a SequenceAnnotation --> Component link, then it will not be possible for SBOL 2.0 or SBOL 2.1 tools to correctly interpret the location. This missing information means it is not a backward compatible change.

[NeilWipat]

Update from COMBINE 2018

Needs more discussion

[cjmyers]

This was already approved, but the remove of Component link from SA has not been executed yet or the adding or Location to Component. These were deferred to SBOL 3.

[bbartley]

Once this SEP is implemented, can we relax the requirement that the definition property is required? I would like that property to be optional. For example, I would like to be able to represent sub-structure using only Components and their respective child Locations.

[cjmyers]

That would definitely need to be a different SEP. Not sure I'm excited about this idea. I guess it could allow you to avoid creating dummy CDs when constructing a design, which could be good. Not sure what the difference would be though in simply using SequenceAnnotations with Roles instead. I think that you end up in the same place.