Electrophoretic signal comparison applied to mRNA differential display analysis

Area of Science:

• Molecular Biology
• Bioinformatics
• Genomics

Background:

• Gene expression analysis using electrophoretic methods is often hindered by time-consuming manual interpretation of signal profiles.
• Objective evaluation of quantitative electrophoretic patterns is crucial for accurate gene expression profiling.

Purpose of the Study:

• To develop a flexible, computer-assisted approach for comparing quantitative electrophoretic patterns across multiple gene expression signals.
• To automate the identification and quantification of significant gene expression patterns.

Main Methods:

• Fitting Gaussian curves to complex electrophoretic peak mixtures.
• Aligning and comparing the fitted signals on a peak-by-peak basis according to user-defined patterns.
• Assigning a score to each pattern to prioritize potential findings for visual analysis.

Main Results:

• The method successfully compresses complex electrophoretic data into a list of exceptional expression patterns with associated numeric features.
• Automated identification of gene expression patterns showed strong agreement with manual visual evaluation in human colonic carcinoma mRNA differential display experiments.
• The developed scoring system effectively highlights the most significant findings for researchers.

Conclusions:

• The computer-assisted method offers an efficient and accurate alternative to manual evaluation of electrophoretic gene expression data.
• This approach streamlines the analysis of gene expression profiling, particularly for large datasets.
• The general comparison framework has potential applications in various gene expression profiling instruments beyond mRNA differential display.