Skip to content

What I want to know is…

What are knowledge graphs, why do we use them and how do we build them?

Check out our introduction to knowledge graphs?

How can I start playing with Translator?

Translator is accessible to translational science end users via the Translator user interface. Python-savvy data scientists can try out the “Hello Translator” Jupyter Notebook.

How can I develop software to use and/or contribute to Translator?

We composed a set of tutorials for you.

How do I find an identifier for a biomedical concept/entity?

CURIE’s (Compact Uniform Resource Identifiers) are used in TRAPI messages to identify particular biomedical objects and concepts. Here are two ways to convert from natural language to a CURIE:

  1. The SRI Name Resolver, specifically the /lookup endpoint, can take natural language and find a CURIE associated with it.
  2. The ARAX UI has a Synonyms feature on the left bar under the heading Tools. In it, you can type a term (with autocomplete) and find equivalent CURIES. This approach uses both the SRI Name Resolver and additional information derived from RTX-KG2.

How do I compute the Normalized Google Distance (NGD) between two concepts

The endpoint https://arax.transltr.io/api/arax/v1.4/ui/#/PubmedMeshNgd/pubmed_mesh_ngd takes two natural language inputs (terms occurring in UMLS) and returns the NGD between them.

How can I quickly view a TRAPI response?

The ARAX UI https://arax.transltr.io/ can take a TRAPI response and render the results. This is mainly helpful for debugging and development use. You can also enter a PK in the ID field .