Analysis of high density expression microarrays with signed-rank call algorithms.

Area of Science:

• Bioinformatics
• Genomics
• Molecular Biology

Background:

• High-density oligonucleotide microarrays are crucial for gene expression analysis.
• Accurate detection and comparison of gene expression are essential across experiments.
• Robustness against outliers and user-adjustable parameters are needed for microarray data.

Purpose of the Study:

• To develop rank-based algorithms for gene expression detection and comparison calls.
• To ensure algorithms are robust against data outliers and offer adjustable specificity/sensitivity.
• To provide reliable gene expression analysis for oligonucleotide microarrays.

Main Methods:

• Utilized rank-based algorithms employing discrimination scores for detection calls.
• Employed intensity differences and Wilcoxon's signed-rank test for comparison calls.
• Incorporated adjustable parameters, p-value calculations, and novel normalization factors for robust analysis.

Main Results:

• Developed rank-based algorithms for robust gene expression detection and comparison.
• Algorithms allow user adjustment of specificity and sensitivity levels.
• Successfully handled scanner saturation and provided confidence levels via p-values.

Conclusions:

• The presented rank-based algorithms offer a robust method for gene expression analysis using microarrays.
• User-adjustable parameters enhance the flexibility of gene expression comparison.
• These algorithms improve the reliability of detecting and comparing expressed genes across diverse experimental conditions.