Source for Edge with Primary KS not Displayed

[sstemann]

I think this issue is because of a mismatch between the primary KS value tag "hetio" and the infores catalog, which is automat-hetio or hetionet

Steps:

1. Use the Test UI > What gene is upregulated by "lisdexamfetamine"

2. View the first result slc6a3

3. on the first/shortest result, click the edge "Affected by +5"

4. note that the "showing evidence for" lists Slc6a3 Directly Physically Interacts With Lisdexamfetamine
   

5. Note: the publications tab does not have relationship Directly Physically Interacts With, neither does the Sources tab
   

6. In ARAX UI, it looks like this edge is coming from BTE> primary KS > Hetio: https://arax.ncats.io/?r=761ba21f-3935-479a-85c4-c8384264adc6
   

7. It's also coming back from Robokop:

[sierra-moxon]

from TAQA:
mismatch between infores:hetio and infores:automat-hetio.
hetio is a source should be the primary, automat should be the aggregator
should be fixed in test; but we will check.
infores:hetio-net and infores:automat-hetio-net
Sierra: infores catalog is broken AGAIN.

[sierra-moxon]

update: infores catalog has been migrated to YAML (old TSV is still in the repository to help folks transition without breaking BMT installations). biolink/biolink-model#1318 addresses the hetio-net consistency issue brought up last time.

[sierra-moxon]

retested - 06/22/23 on ci.
https://ui.ci.transltr.io/results?l=Lisdexamfetamine&t=3&q=37271f9d-7cb9-48b9-aed3-d12f2915197b
https://arax.ncats.io/?r=37271f9d-7cb9-48b9-aed3-d12f2915197b

No longer seeing "affects 5+" in the slc6a3 result.

I see the "directly physically interacts with" predicate in the evidence display for the first, direct edge.

I think this is working now :) closing as complete.