An evidence ontology for use in pathway/genome databases

Area of Science:

• Bioinformatics
• Database Development
• Ontology Engineering

Background:

• Model Organism Databases (MODs) and other bioinformatics databases increasingly rely on computationally predicted information.
• There is a growing need to capture the scientific evidence supporting database assertions, especially for predicted data like gene functions and metabolic pathways.

Purpose of the Study:

• To present an ontology for encoding the type and degree of scientific support for database assertions.
• To enable the encoding of literature sources for reported evidence.
• To address the need for robust evidence tracking in bioinformatics databases.

Main Methods:

• Developed an ontology with a hierarchy of 35 evidence codes.
• The codes model various types of wet-lab and computational evidence.
• Implemented the ontology within the Pathway Tools software environment.

Main Results:

• The ontology effectively models different types of evidence for gene functions, operons, and metabolic pathways.
• The implementation allows for querying and updating evidence within Pathway/Genome Databases (DBs).
• Successfully integrated into software used for major databases like EcoCyc, MetaCyc, and HumanCyc.

Conclusions:

• The proposed ontology provides a standardized method for representing scientific evidence in bioinformatics databases.
• This approach enhances data reliability and transparency, particularly for computationally derived information.
• The implementation facilitates better data management and querying in critical biological databases.