Make a list of Mongo queries people find difficult to write

[turbomam]

This ticket is deprecated. Please refer to this Github issue and this Google Sheet

There are some NMDC information retrieval tasks that do not seem intuitive.

Let's collect some use cases that call out

• what the task is
  • list the input knowledge and the output knowledge, like getting the identifiers for all Biosamples that are part_of a given Study
• who the intended audience is
  • NMDC team members
  • workflows
  • external users

Possible solutions:

• fancy pure-MongoDB queries over the existing collections
• ad-hoc Python scripts that integrate results form multiple queries
• MongoDB views
• SPARQL
• existing API endpoints
• new API endpoints

[turbomam]

I added some people who might have some thoughts on this to the issue assignees, but please add or remove others if necessary.

[brynnz22]

@turbomam here is a mongo query I wrote for one of the notebooks to go from biosample to a data object: https://github.com/microbiomedata/notebook_hackathons/blob/soil-contig-tax/taxonomic_dist_by_soil_layer/python/mongodb_query.txt.js

[aclum]

@turbomam see microbiomedata/nmdc-runtime#355
My preference would be consolidate tickets, no preference which repo we work out of.

[PeopleMakeCulture]

@dwinston and I have been collecting examples of difficult queries in microbiomedata/nmdc-runtime#355

[mslarae13]

I'm going to close this issue. @brynnz22 @turbomam please copy your comments to the existing issue in runtime.