Associating SRA accession numbers with trait scoring within a given trial

[Nuanda]

Sent by @teatree1212:

The user wants to select all measurements (trait_scores.score_values) from a given plant_trial.plant_trial_name.

The user wants to select all SRA-accession numbers ( plant_lines.sequence_identifier) from the same plant_trial.

In order for the measurements to make sense, the score_values need to be available in context with their trait_descriptors.desriptor_name.

Another important metadata-field would be the plant_accession.plant_accession. Since plant_parts is a separate table, this should be added to the trait_descriptors information. No more information of that table would be necessary.

The API needs to access 7 tables for this- or rather, there needs to be joins across these 6 tables, of which the API picks up the relevant data ( and why not more metadata, too).

So the user-GET request would be something like: given this plant_trial_name, under the condition that it is processed data,

give me all

1. {trait_scores with associated trait descriptors, including the plant_part}

2. information on the plant_accession

3. information on the plant_lines, probably from an SRA-identifier’s point of view ( if that makes sense?)

4. information on plant_trial itself.

5. Minimal info on the plant_scoring_units, maybe just plant_scoring_units.date_planted.

I can see two other ways to get this data as it stands now, access the data from the database using joins ( which would be the first step for ruby, too) , or using a script to merge all json outputs.

I think though that this extension of the API may be quite a necessary step, in order to provide analytical tools and workflows with this information.

[Nuanda]

@teatree1212 Annemarie, please tell me if the new zip export of PlantTrials is enough to cover this use case.