LyM: a tool to reach the best factor in gene expression comparison

Area of Science:

• Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• Analyzing gene expression changes across libraries is crucial for understanding biological processes.
• Existing tools like DGED require manual interaction and may miss subtle expression differences.

Purpose of the Study:

• To develop and validate LyM, a Perl-based tool for determining gene expression fold changes.
• To compare LyM's performance against the gold standard DGED using SAGE libraries.

Main Methods:

• LyM utilizes a Bayesian framework to analyze prior and posterior probability density functions.
• A binary search algorithm is implemented in LyM to efficiently find the optimal fold change factor.
• Performance was validated using SAGE libraries from human breast, colon, lung, and stomach tissues (normal vs. tumor).

Main Results:

• LyM quantitatively matched results generated by DGED.
• LyM accurately retrieved the best fold change factor (F) for gene expression differences.
• LyM demonstrated significant time economy due to its binary search algorithm.
• LyM revealed differential expression levels exceeding 100-fold, which were previously hidden in DGED.

Conclusions:

• LyM provides an accurate, fast, and convenient method for gene expression analysis in molecular biology.
• The tool requires minimal user interaction, making it accessible for standard laboratory workstations.
• LyM enhances the discovery of significant differential gene expression.