A Web-based and Grid-enabled dChip version for the analysis of large sets of gene expression data

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Microarray techniques are crucial for studying gene expression in complex diseases.
• Existing analysis tools like Bioconductor and dChip have limitations for large datasets.
• The dChip software's gene expression analysis component was adapted for distributed computing.

Purpose of the Study:

• To develop a Grid-enabled application for analyzing large-scale microarray data.
• To enable researchers to analyze extensive datasets without hardware or software constraints.
• To modify and parallelize dChip for execution on cluster and Grid infrastructures.

Main Methods:

• Algorithms from dChip were used to create a new application.
• The dChip software was ported to Linux and parallelized.
• The application was tested on cluster and Grid environments using diverse datasets.

Main Results:

• A Grid-enabled software application for large microarray dataset analysis was successfully developed.
• Validation tests were performed using Human Umbilical Vein Endothelial Cells and Fibroblast data.
• Performance tests on cluster and Grid environments showed good scalability with a 1000-sample breast cancer dataset.

Conclusions:

• A Grid-enabled software application for analyzing large microarray datasets has been proposed.
• The modified dChip software effectively utilizes Grid technologies for distributed data analysis.
• The application demonstrates good scalability and performance on both cluster and Grid infrastructures.