Building Molecular Interaction Networks from Microarray Data for Drug Target Screening

Area of Science:

• Bioinformatics
• Systems Biology
• Genomics

Background:

• Microarray studies reveal differential gene expression between patients and healthy individuals.
• Identifying potential drug targets is crucial for effective disease treatment.
• Existing methods require robust computational approaches for network analysis.

Purpose of the Study:

• To develop and validate a computational method for identifying potential drug targets using differentially expressed genes.
• To construct molecular interaction networks for pinpointing pivotal molecules.
• To demonstrate the method's application using a type 2 diabetes mellitus microarray dataset.

Main Methods:

• Differential gene expression analysis using R packages (simpleaffy, affy, limma).
• Construction of molecular interaction networks with InnateDB.
• Identification of significant gene modules using Cytoscape (jActiveModules).
• Association testing of gene interactions using the psych package.
• Selection of drug targets based on significant associations and known protein structures (Protein Data Bank).

Main Results:

• A systematic pipeline was established for drug target identification from microarray data.
• Key genes and molecular interactions were identified within disease-specific networks.
• The method successfully identified potential drug targets in a type 2 diabetes mellitus case study.
• Significant gene associations and protein structure availability were used for target prioritization.

Conclusions:

• The described method provides a robust framework for discovering novel drug targets from gene expression data.
• Network-based analysis of differentially expressed genes enhances the identification of clinically relevant molecules.
• This approach holds promise for accelerating drug discovery in various diseases, including type 2 diabetes mellitus.