Extraction of Conditional Probabilities of the Relationships Between Drugs, Diseases, and Genes from PubMed Guided by Relationships in PharmGKB

Guided by curated associations between genes, treatments (i.e., drugs), and diseases in pharmGKB, we constructed n-way Bayesian networks based on conditional probability tables (cpt’s) extracted from co-occurrence statistics over the entire Pubmed corpus, producing a broad-coverage analysis of the r...

Descripción completa

Detalles Bibliográficos
Autores principales: Theobald, Martin, Shah, Nigam, Shrager, Jeff
Formato: Texto
Lenguaje:English
Publicado: American Medical Informatics Association 2009
Materias:
Articles
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3041559/
https://www.ncbi.nlm.nih.gov/pubmed/21347183
Descripción
Sumario:Guided by curated associations between genes, treatments (i.e., drugs), and diseases in pharmGKB, we constructed n-way Bayesian networks based on conditional probability tables (cpt’s) extracted from co-occurrence statistics over the entire Pubmed corpus, producing a broad-coverage analysis of the relationships between these biological entities. The networks suggest hypotheses regarding drug mechanisms, treatment biomarkers, and/or potential markers of genetic disease. The cpt’s enable Trio, an inferential database, to query indirect (inferred) relationships via an SQL-like query language.

Ejemplares similares