GENIES: gene network inference engine based on supervised analysis.

Area of Science:

• Bioinformatics
• Systems Biology
• Computational Biology

Background:

• Understanding gene regulatory networks is crucial for deciphering biological processes.
• Existing methods often struggle to integrate diverse genomic data effectively.
• Predicting unknown gene interactions remains a significant challenge in systems biology.

Purpose of the Study:

• To develop a web server, GENIES, for predicting gene networks using supervised learning.
• To integrate heterogeneous genomic data through kernel methods for enhanced network inference.
• To identify candidate genes for missing enzymes in metabolic pathways.

Main Methods:

• Utilizes supervised network inference framework.
• Employs kernel methods for integrating heterogeneous data like gene expression and phylogenetic profiles.
• Trains predictive models using partially known network information (e.g., KEGG PATHWAY database).

Main Results:

• GENIES predicts novel gene pairs and maps them onto known pathway diagrams.
• The server identifies candidate genes potentially acting as missing enzymes.
• Provides a user-friendly interface for uploading gene profiles or similarity matrices.

Conclusions:

• GENIES offers a powerful approach for gene network inference by integrating diverse data.
• The tool facilitates the discovery of novel gene interactions and functional insights.
• GENIES is a valuable resource for researchers in genomics and systems biology.