Add data integrity tests for MIRIAM URIs/URNs

[cthoyt]

I found that there were several URNs pointing to prefixes in MIRIAM that do not exist. This pull request adds unit tests that check that all entries in the uri column starting with urn:miriam: are indeed valid MIRIAM prefixes.

Entries for the following were updated to use the correct MIRIAM prefixes:

• [x] FlyBase
• [x] MGI
• [x] OMIM
• [x] Pathway Ontology
• [x] PATO
• [x] Systems Biology Ontology
• [x] WormBase

Entries for the following are not available in MIRIAM and were removed:

• [x] BIND (this resource doesn't exist anymore)
• [x] IPI (this resource doesn't exist anymore)
• [x] SPRINT
• [x] Unipathway (MIRIAM has unipathway.reaction and unipathway.compound, neither seem appropriate)

For SPRINT and Unipathway, there are alternative prefixes from the Bioregistry (https://bioregistry.io/sprint and https://bioregistry.io/upa). For BIND and IPI, we can add prefixes to the Bioregistry and mark them as "deprecated", if that helps.

[egonw]

ping @Finterly, you last year some issues with upper/lower case things in the datasources.csv, so please take notice of the changes in this patch. no action is needed otherwise whatsoever.

[cthoyt]

I'm not quite sure I understood your feedback @egonw, are there any changes needed for this PR? Or do you have to first consider the downstream impact of standardizing terminology here?

[DeniseSl22]

@egonw : any time to check this PR from @cthoyt ?

[egonw]

cc @Finterly

[egonw]

@cthoyt:

E   ModuleNotFoundError: No module named 'bs4'

[cthoyt]

getting there!

[egonw]

ping @cthoyt, there was overlap with the other PR. I pulled in one patch from this PR.

Can you have a look? And salvage other bits? For the changes in the datasources.tsv, please put this in an isolated PR because I want @Finterly to have a look at that, because of PathVisio depending on that info.

[cthoyt]

@egonw not sure if makes sense to put the PR for tests that are going to fail - the idea would be that once you implement some tests, you update the data to follow that standard and never let the tests fail again. But if that's what you want, I could separate these.

in the mean time I resolved the diff

[egonw]

@Finterly, @mkutmon, please take note of the changes in upper/lower casing in the datasources.tsv. Fin, if I remember correctly, there was something about upper/lower case last year, but I cannot quite remember what it was. Are these changes going to cause any downstream issues?

[cthoyt]

@Finterly @mkutmon feedback would be appreciated

[egonw]

Okay, there are merge conflicts, so will have to do this one manually.

[cthoyt]

@egonw not sure what the issue is, in the most recent commit on this branch 289e54b I addressed all conflicts with the main branch

[egonw]

This branch cannot be rebased due to conflicts

It says "This branch cannot be rebased due to conflicts" ... i will explore, but after the meeting I am in.

[cthoyt]

Are you looking at the github interface for merging or trying to do this locally?

[egonw]

github interface

[cthoyt]

then I'd suggest doing "Squash and merge" which would turn all commits into this branch into a single one as it's merged. There isn't really any valuable information in the small commits made during this PR