This view aims to "triangulate" evidence between MR estimates (from MR-EvE) and published scientific papers (from MEDLINE using SemMedDB).
For a given trait the query will map this to UMLS concepts, then identify literature-derived triples that link this to any other trait for which we have pairwise MR evidence with the original trait.
The MR relationship will link a pair of traits directly based on MR evidence.
The literature triples will link a pair of traits via mapped UMLS concepts, with a predicate (eg AFFECTS or ASSOCIATED_WITH) that has been derived from text-mining of the literature.
The papers underpinning a particular literature triple are provided with their PubMed ID. We strongly recommend reviewing these papers to understand the context of the relationship!
NOTE: this view is still in alpha stage.
Causal estimates are from MR-EvE (Hemani et al, Automating Mendelian randomization through machine learning to construct a putative causal map of the human phenome, bioRxiv 10.1101/173682)
Literature triples (Subject-Predicate-Object) are from SemMedDB (1-4)
How to use
Enter your trait of interest outcome in the Trait text box.
You can choose a prediacte (eg AFFECTS or STIMULATES) in the Semmed predicate text box.
Top results are presented in the Network plot tab, with full results searchable in the Table tab. The Query tab provides information to enable you to download full results using the API.
Kilicoglu, H, Shin D, Fiszman M, Rosemblat G, Rindflesch TC. (2012) SemMedDB: A PubMed-scale repository of biomedical semantic predications. Bioinformatics, 28(23), 3158-60.
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.
Kilicoglu, H. et al. (2008). Semantic MEDLINE: A Web Application to Manage the Results of PubMed Searches. In Proceedings of the Third International Symposium on Semantic Mining in Biomedicine (SMBM 2008), 69-76.
Rindflesch, T.C. et al. (2011) Semantic MEDLINE: An advanced information management application for biomedicine. Information Services & Use, 31, 15-21.