PosMed: Ranking genes and bioresources based on Semantic Web Association Study

Area of Science:

• Biomedical Informatics
• Genomics
• Bioinformatics

Background:

• Positional MEDLINE (PosMed) is a Semantic Web Association Study engine used since 2005 for in silico positional cloning.
• It ranks biomedical resources like genes, metabolites, and diseases based on associations with phenotypic keywords.
• These associations are derived from a Semantic Web of biological databases.

Purpose of the Study:

• Redesign PosMed as a workbench to discover functional interpretations for genetic variants from exome sequencing.
• Enhance the value of mouse bioresources by inferentially associating them with phenotypes.
• Improve the discovery of candidate disease genes and functional interpretations of genetic variants.

Main Methods:

• Established text-mining rules for biomedical documents through human curation.
• Created extensive, accurate links between genes and documents.
• Integrated 20 public databases and four original datasets to associate phenotypic keywords with mouse resources.

Main Results:

• PosMed is redesigned as a workbench for functional interpretation of genetic variants.
• Demonstrated enhanced utility of mouse bioresources through inferential phenotype association.
• Successfully linked phenotypic keywords to mouse resources using a comprehensive data integration.

Conclusions:

• The redesigned PosMed workbench facilitates the interpretation of genetic variants in human disease.
• PosMed enhances the utility of mouse models in biomedical research by enabling phenotype-based associations.
• The Semantic Web approach provides a powerful framework for integrating diverse biological data for discovery.