search

NCATSTranslator / Feedback

A repo for tracking gaps in Translator data and finding ways to fill them.
7 stars 0 forks source link

ATP is top result for Adams-Oliver syndrome, based on weak SemMedDB evidence #924

Open amykglen opened 2 months ago

amykglen commented 2 months ago

on ui.test.transltr.io, I ran the query "What drugs may treat conditions related to Adams-oliver Syndrome?"

which produced these results: https://ui.test.transltr.io/results?l=Adams-oliver%20Syndrome&i=MONDO:0007034&t=0&r=0&q=cc818ef5-ff34-45ad-bee6-0079c6f2df52

ATP is the first result, and the three edges that link ATP to an Adams-Oliver-associated gene are all from SemMedDB.

due to #923, I can't view which publications these SemMedDB assertions were extracted from, however, I can see that all three edges are based on 2 or fewer publications.

and oddly, two of the SemMedDB edges seem to contradict each other: one says that ATP causes increased activity of DLL1, and another says that ATP causes decreased activity of DLL1:

Screenshot 2024-08-26 at 6 56 08 PM

is that correct that ATP both increases and decreases activity of DLL1?

and should ATP really be the top result here?

sierra-moxon commented 2 months ago

@kaiwenho - I saw unsecret in the inferred edge provenance - would you be able to look into this a bit further and reassign if necessary? :).