BADGE

BADGE applies Bayesian statistical analysis to identify and quantify differential gene expression from microarray experiments for investigation of biological responses and disease-associated expression changes.

Key Features:

• Bayesian Statistical Framework: Implements Bayesian methods to provide probabilistic interpretation and quantify uncertainty in differential gene expression estimates.
• Differential Expression Criteria: Calls differential expression using a probability >0.999 for greater than 1.5-fold change in signal intensity with a predicted score between 70 and 100.
• Cross-Validation: Validates differential expression predictions using cross-validation techniques.
• Whole-Genome Microarray Compatibility: Supports whole-genome microarrays, explicitly including the Affymetrix U133A-B GeneChip.

Scientific Applications:

• Sickle Cell Disease (SCD) research: Applied to study gene expression changes associated with SCD and to investigate endothelial cell (EC) dysfunction as a potential modifier of clinical variability.
• Endothelial cell response analysis: Used to analyze human pulmonary artery endothelial cells (HPAEC) exposed to plasma from acute chest syndrome (ACS) patients, steady-state SCD patients, normal volunteers, and serum-free media to identify exposure-specific expression changes.
• Pathogenesis and phenotype characterization: Identified 50 genes (steady-state SCD versus normal) and an additional 58 genes (ACS exposure) implicating cholesterol biosynthesis, lipid transport, cellular stress response, extracellular matrix proteins, and a shift toward anti-apoptotic and enhanced cholesterol biosynthesis phenotypes potentially relevant to sickle vasoocclusion.

Methodology:

Performs Bayesian analysis of microarray signal intensities, applies a probability >0.999 and >1.5-fold change threshold with predicted scores 70–100 to define differential expression, and validates results via cross-validation.