Ontological Discovery Environment: a system for integrating gene-phenotype associations

Area of Science:

• Genomics
• Bioinformatics
• Systems Biology

Background:

• Genomic technologies allow rapid characterization of biological traits.
• Effective biological investigation requires clear operational definitions of processes.
• Current biological characterization often relies on external constructs and semantics.

Purpose of the Study:

• To propose a new framework for categorizing biological characters based on endogenous processes and biological networks.
• To introduce the Ontological Discovery Environment (ODE) for phenotype-centered genomic data analysis.
• To demonstrate ODE's utility in integrating diverse genomic datasets for hypothesis discovery.

Main Methods:

• Utilizing the Ontological Discovery Environment (ODE) for data storage, sharing, retrieval, and analysis.
• Inputting various gene-phenotype relationship datasets, including QTL, literature, microarray, and ontological data.
• Employing a bipartite network of gene-phenotype relations for gene set similarity, distance, and hierarchical analysis.
• Leveraging cross-species homology and metadata for data synthesis and interpretation.

Main Results:

• ODE enables the analysis of phenotype-centered genomic data across species and model systems.
• The bipartite network approach allows for set-set matching of non-referential data.
• Demonstrated a use case analyzing alcoholism genomic studies, showcasing ODE's homology and data synthesis capabilities.
• Integrated computationally derived gene sets into hierarchical trees based on phenotype interdependencies.

Conclusions:

• ODE facilitates data integration and hypothesis discovery across diverse experimental contexts.
• The proposed framework moves beyond external semantics to define biological characters based on natural processes.
• ODE provides a robust platform for comparative genomics and understanding gene-phenotype relationships.