Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[bioontology-support] Questions regarding API search

Michael Dorf mdorf at stanford.edu
Thu Jun 25 13:17:34 PDT 2020


Hi John,

Glad to hear from you again! I can confirm that this behavior is by design. Our default search does not perform “starts with” partial matches, but only the whole word matches. It does match on single words in a multi-word expression, but those results are usually ranked lower than the full expression matches. If you want to do a “starts with” partial match, you can use the asterisk in the query, as you did below, оr pass a parameter “suggest=true”:

http://data.bioontology.org/search?q=ubi&ontologies=PSIMOD&page=1&pagesize=40&display_context=false&suggest=true

I think the confusion comes from this statement: prefLabel (match on partial pref label). What I meant here is if one searches for “abnormal cell”, for example:

http://data.bioontology.org/search?q=abnormal%20cell&ontologies=NCIT

You can see that the exact match “Abnormal Cell” is at the top; the matches that contain both words “abnormal” and “cell” are a bit lower; and the partial matches - those that contain either word - are there but ranked even lower. That’s what I meant by the “partial pref label”, NOT partial word matches. Sorry for the confusion.

Hope this helps!

Michael


On Jun 25, 2020, at 12:33 PM, John Zobolas <john.zobolas at ntnu.no<mailto:john.zobolas at ntnu.no>> wrote:

Hi,

This is a continuation of some other Q&A I had raised some time ago.
My question back then was about the fields that a query such as `search?q=something` looks for to make a match. Michael Dorf had told me that:

"These are the fields being searched in their order of rank priority:

id
prefLabelExact (match on the full pref label)
prefLabel (match on partial pref label)
synonymExact (match on the full synonym(s))
synonym (match on the partial synonym(s))
notation (last fragment of id)
cui (for UMLS ontologies)
semantic_types"

We have been beta-testing this curation tool<https://mi2cast.github.io/causalBuilder/>, in which (long-story-short) some boxes use the BioPortal API (via a dictionary<https://github.com/UniBioDicts/vsm-dictionary-bioportal> interface I've built for it), and some curators expected something to show in one of these boxes when they were writing 'ubi' (for ubiquitination) which pretty much translated to the query:

http://data.bioontology.org/search?q=ubi&ontologies=PSIMOD&page=1&pagesize=40&display_context=false

, which returns nothing. But if you add the asterisk (dirty little trick 😎):

http://data.bioontology.org/search?q=ubi*&ontologies=PSIMOD&page=1&pagesize=40&display_context=false<http://data.bioontology.org/search?q=ubi*&ontologies=PSIMOD&page=1&pagesize=40&display_context=false&apikey=%3CTestAPIkeyToFillin%3E>

you can see that there is both prefLabel and synonym partial matches, so these items/classes should have been returned in the first query!

So my question is if this a bug (the partial match does not work properly) or something has changed in the above fields (and I should probably implement the asterisk hack cause the curators are somewhat dissatisfied with writing the whole word 😊)?

==================
Another example: Curator wants to find 'physical association', nothing is returned while they type:
http://data.bioontology.org/search?q=physi&ontologies=MI&page=1&pagesize=40&display_context=false

There is though a partial prefLabel match:
http://data.bioontology.org/search?q=physi*&ontologies=MI&page=1&pagesize=40&display_context=false
===================

BR, John.
-----------------
John Zobolas
PhD Student
Department of Biology, Faculty of Natural Sciences, NTNU
Department of Clinical and Molecular Medicine, Faculty of Medicine and Health Sciences, NTNU


_______________________________________________
bioontology-support mailing list
bioontology-support at lists.stanford.edu<mailto:bioontology-support at lists.stanford.edu>
https://mailman.stanford.edu/mailman/listinfo/bioontology-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/bioontology-support/attachments/20200625/cc871da0/attachment-0001.html>


More information about the bioontology-support mailing list