SemRep is a UMLS-based program that extracts three-part propositions, called semantic predications, from sentences in biomedical text. Predications consist of a subject argument, an object argument, and the relation that binds them. For example, from the sentence in (1), SemRep extracts the predications in (2).

  1. We used hemofiltration to treat a patient with digoxin overdose that was complicated by refractory hyperkalemia.
  2. Hemofiltration-TREATS-Patients
    Digoxin overdose-PROCESS_OF-Patients
    hyperkalemia-COMPLICATES-Digoxin overdose
    Hemofiltration-TREATS(INFER)-Digoxin overdose
The subject and object arguments of each predication are concepts from the UMLS Metathesaurus and their binding relationship (in uppercase) is a relation from the UMLS Semantic Network. For a detailed description of SemRep, see [1].

Holders of a UMLS license can run SemRep interactively or in batch mode using the SKR Scheduler. SemRep is also available as a stand-alone program on the Linux platform.


  1. Rindflesch, T.C. and Fiszman, M. (2003). The interaction of domain knowledge and linguistic structure in natural language processing: interpreting hypernymic propositions in biomedical text. Journal of Biomedical Informatics, 36(6):462-477.