Closed erikyao closed 6 years ago
Hi, that’s because the standard way representing a disease ontology is DOID:1498, rather than the number itself. You might try http://biothings.io/explorer/api/v2/directinput2output?input_prefix=do&input_value=DOID:1498&output_prefix=hp to see if it works.
Btw, are you Yao Yao from Steve’s group? Looking forward to seeing u in the hackathon.
Best, Kevin
Oh, I got it. Thank you very much!
Yes, I work for Steve. See you later!
Hi Kevin,
In terms of OMIM, I have to use input_prefix=omim.disease
and input_value=605543
, not input_value=OMIM:605543
. Could you please share me the standard you have? Thanks!
I see we need to follow the patterns on Identifiers.org. Thanks!
You might check out our ID_Mapping file located at: https://github.com/NCATS-Tangerine/translator-api-registry/blob/openapi_2.0/ID_MAPPING_NEW.csv. When determining what is the standard syntax representing an ID, we first refer to 'http://identifiers.org'. For biological ID, it has a field called 'Identifier pattern' to specify the standard syntax. You might check http://identifiers.org/doid/. It specifies the standard syntax should be ^DOID\:\d+$. Also, you might check http://identifiers.org/omim/. It specifies the standard syntax should be ^[*#+%^]?\d{6}$. But I do understand that it might cause some confusion for new users. So I will also add support for both the CURIE form (e.g. OMIM:605543) as well as the short form (e.g. 605543) in the BioThings Explorer, which will make life easier for the users.
Thanks!
Kevin
The problem is that we use CURIE forms all over our application. I'll adapt our IDs when using BioThings Explorer. Thank you very much!
BTW, do you happen to know if there is any python library that can convert CURIE form to short form? Thank!
E.g. for request url:
Response is:
However, for underlying request endpoint using "ONID" instead of "do":
The response is not empty.
Thank you!