MEDIE & Enju API

 MEDIE is an intelligent search engine to retrieve biomedical correlations from MEDLINE, based on indexing by Natural Language Processing and Text Mining techniques.  Enju is a syntactic parser for English. MEDIE's excellent search function can be programmatically accessed using its API. In addition, parsing results of Enju on the all MEDLINE records can be accessed programmatically.

Attention: Please cite  MEDIE when using the API.

•  Enju API Document
• Enju API endpoint:  http://www-tsujii.is.s.u-tokyo.ac.jp/enju-medline/enju-medline.cgi
• MEDIE API endpoint:  http://www-tsujii.is.s.u-tokyo.ac.jp/medie/dbcls.cgi
  • Arguments
    • subject: alpha-numerics
    • verb: ditto
    • object: ditto
    • count: 1 to 100 (the number of the results, default is 3)
    • sm: 0 or 1 (sm means "semantics")
    • position: value of the <next> tag
    • pmid: PubMed ID
  • Sample URLs
    •  http://www-tsujii.is.s.u-tokyo.ac.jp/medie/dbcls.cgi?subject=p53&verb=activate&sm=1
    •  http://www-tsujii.is.s.u-tokyo.ac.jp/medie/dbcls.cgi?subject=p53&verb=activate&sm=1&count=10
    •  http://www-tsujii.is.s.u-tokyo.ac.jp/medie/dbcls.cgi?subject=p53&verb=activate&sm=1&position=medline09n0893-7388931
    •  http://www-tsujii.is.s.u-tokyo.ac.jp/medie/dbcls.cgi?pmid=19116711&count=1

The above services are provided by  Tsujii Laboratory.

Obtaining Subject, Verb, and Object of each sentence of MEDLINE titles/abstracts

• Endpoint:  http://docman.dbcls.jp/medieconv
  • argument: pmid (PubMed ID, multiple IDs can be accepted using bar "|" as a delimiter)
  • Example:  http://docman.dbcls.jp/medieconv?pmid=17551671|18424370

Back to TextMining Wiki