Fold change rank ordering statistics: a new method for detecting differentially expressed genes.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Analysis of differentially expressed (DE) genes in microarray data is crucial for biological research.
• Existing methods often rely on statistical tests incorporating expression variability or fold change (FC).
• FC-based results are noted for higher reproducibility and biological relevance compared to variability-based methods.

Purpose of the Study:

• To propose a novel method for identifying DE genes in microarray data.
• To address limitations of existing methods, particularly in handling noisy datasets and multiple testing issues.

Main Methods:

• Developed Fold Change Rank Ordering Statistics (FCROS) method.
• FCROS utilizes combinatorial pairs of biological conditions to exploit variations in calculated FC levels.
• A statistic is associated with the ranks of FC values for each gene to identify DE genes.

Main Results:

• The FCROS method is deterministic and computationally efficient.
• FCROS effectively addresses the multiple testing problem common in microarray analysis.
• Performance comparison using synthetic and real microarray datasets demonstrated FCROS's suitability.

Conclusions:

• FCROS is well-suited for identifying DE genes, particularly in noisy microarray datasets.
• The method provides a robust alternative to existing FC-based approaches for DE gene analysis.