Mining Genotype-Phenotype Associations from Public Knowledge Sources via Semantic Web Querying

• Published in: AMIA Summits on Translational Science proceedings Vol. 2013; pp. 118 - 122
• Main Authors: Kiefer, Richard C, Freimuth, Robert R, Chute, Christopher G, Pathak, Jyotishman
• Format: Journal Article
• Language: English
• Published: United States American Medical Informatics Association 2013
• ISSN: 2153-4063; 2153-4063

Gene Wiki Plus (GeneWiki+) and the Online Mendelian Inheritance in Man (OMIM) are publicly available resources for sharing information about disease-gene and gene-SNP associations in humans. While immensely useful to the scientific community, both resources are manually curated, thereby making the data entry and publication process time-consuming, and to some degree, error-prone. To this end, this study investigates Semantic Web technologies to validate existing and potentially discover new genotype-phenotype associations in GWP and OMIM. In particular, we demonstrate the applicability of SPARQL queries for identifying associations not explicitly stated for commonly occurring chronic diseases in GWP and OMIM, and report our preliminary findings for coverage, completeness, and validity of the associations. Our results highlight the benefits of Semantic Web querying technology to validate existing disease-gene associations as well as identify novel associations although further evaluation and analysis is required before such information can be applied and used effectively.