test instantiating biosamples from NCBI

[turbomam]

• illustrate how NCBI Biosamples could be instantiated from either efetch or LBL's Biosample SQL database #147

Instantiating against submission-schema instead of nmdc-schema because submission-schema is already flat, like NCBI Biosample attributes.

• There are some required slots in SoilInterface (for example) that are unlikely to be asserted for NCBI Biosamples
  • so I'm overlaying NCBI Biosample attributes over a known good data file
• There are some NCBI Biosample attributes don't would require minor tweaking, like a lat_lon of '71.323 N 156.6114 W'
• other NCBI Biosample attributes may be much more further away from our specification

[turbomam]

@pkalita-lbl have you worked on anything like this yet?

I'm getting the Attributes for one NCBI Biosample here from the efetch API. We also have a Postgres version of the NCBI Biosample database on SPIN.

[turbomam]

There's some code in https://github.com/INCATools/biosample-analysis that tries to clean up some NCBI Biosample attributes. I don't think it uses nmdc-chema or submission-schema classes as its target.

It was some of the earliest and worst code I wrote for BBOP.

Maybe in https://github.com/INCATools/biosample-analysis/blob/mam-envo-mapping/src/mixs-envo-mapping/biosample-triad-mapping.py?

It might not be much use doing any more research in there.

[turbomam]

I don't know why the actions are saying

Because nmdc-submission-schema depends on xmltodict (^0.13.0) which doesn't match any versions, version solving failed.
Error: Process completed with exit code 1.

https://pypi.org/project/xmltodict/0.13.0/

[turbomam]

I have the sense that team members like @aclum and @sujaypatil96 are developing code for this in nmdc-runtime or other repos and I propose closing this PR

[aclum]

Efforts of the current squad have been focused on nmdc biosamples -> ncbi samples + SRA records, not the other way around.