GOLink: Finding Cooccurring Terms across Gene Ontology Namespaces.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• The Gene Ontology (GO) is a crucial resource for annotating genes and gene products, utilized by major public repositories.
• GO comprises three subontologies: Cellular Component, Molecular Function, and Biological Process.
• Current querying limitations hinder integrated analysis across GO subontologies.

Purpose of the Study:

• To develop a novel method for querying and integrating information across GO subontologies.
• To address the challenge of finding co-occurring terms within the GO database.
• To facilitate prediction of gene functions and processes.

Main Methods:

• GOLink identifies co-occurring terms from all GO subontologies within the GO database.
• Genes annotated with a specific term are exported and their annotations are categorized.
• The software utilizes Molecular Function terms to predict Biological Process for gene sets.

Main Results:

• GOLink successfully links terms across GO subontologies, enabling cross-domain queries.
• The software achieved 80-90% accuracy, sensitivity, and specificity in predicting Biological Processes from Molecular Functions.
• This demonstrates a robust method for inferring gene functions.

Conclusions:

• GOLink provides a standardized approach to querying the Gene Ontology database.
• The tool enhances gene annotation and facilitates biological discovery by integrating diverse GO terms.
• GOLink is freely available for non-commercial use, promoting wider research applications.