Genbank file outputs linear seqs instead of plasmids

[nickdelkis]

Distribution.gb sequences are linear instead of circular.

Exporting circular sequences would make it less confusing for teams (although they can be used for cloning purposes as linear seqs as well)

[jakebeal]

Is this an actual issue, or a test issue?

[nickdelkis]

It is a true issue, wanted to add context after the webinar and forgot. Adding now.

[nickdelkis]

Actually, I think this shouldn't be too hard to tackle. I will look into it Jake

[jakebeal]

I suspect that the problem will turn out to be that the SBOL3 to SBOL2 converter isn't handling the SequenceOntology "circular" topology type correctly.

[nickdelkis]

Are you referring to this function?

https://github.com/SynBioDex/SBOL-utilities/blob/cd8f27f96de27aae044296204c67917f5672aa72/sbol_utilities/conversion.py#L169-L201

[jakebeal]

That's right.

SBOL2 and SBOL3 both use SequenceOntology term SO:0000988 (Circular), but recommended canonical form has been simplified by identifiers.org from http://identifiers.org/so/SO:0000988 to http://identifiers.org/SO:0000988. With pySBOL3, using tyto to compare terms will find that these are equivalent.

With the old SBOL2 to GenBank converter, however, it's a string comparison, which won't recognize http://identifiers.org/SO:0000988 as the circular type: https://github.com/SynBioDex/libSBOLj/blob/d99912e57e2d734d86576f6cdb6b24ce88cd6583/core2/src/main/java/org/sbolstandard/core2/GenBank.java#L449

I thus believe this issue should be able to be patched by adding entries for the topology types (circular, linear, single-stranded, and double-stranded) to the type-remapping table in the conversion function you point to.

The longer-term fix will be the direct SBOL3 to GenBank converter that @mohitdmak is working on in Google Summer of Code right now, but a) that's not yet ready, and b) this patch will make sure these test cases are there for him to match.