Comparison and consolidation of microarray data sets of human tissue expression

Area of Science:

• Genomics and Bioinformatics
• Molecular Biology
• Human Tissue Expression Analysis

Background:

• Human tissues exhibit diverse structures and functions, necessitating gene expression studies to understand their development from common DNA.
• Microarray technologies have generated large-scale human tissue expression datasets, crucial for biomedical research.
• Microarray data are prone to noise and experimental artifacts, requiring critical comparison and validation.

Purpose of the Study:

• To compare and integrate four publicly available human tissue expression datasets generated on three different microarray platforms.
• To assess the reliability and consistency of gene expression data across platforms.
• To construct consolidated, platform-independent lists of tissue-specific genes.

Main Methods:

• Comparative analysis of four distinct human tissue expression datasets comprising 377 microarray hybridizations.
• Statistical assessment of gene expression profile similarity across different microarray platforms.
• Development of complementary measures to generate consolidated, platform-independent gene lists.

Main Results:

• Gene expression analysis outcomes are significantly influenced by the chosen dataset.
• Statistically significant similarities in gene expression profiles were observed across different microarray platforms.
• Consolidated gene lists derived from integrated data demonstrated increased reliability in follow-up analyses.

Conclusions:

• Consolidating multiple tissue expression datasets enhances data quality and biological interpretability.
• The developed compendium of platform-independent gene lists facilitates the identification of novel tissue-specific marker genes.
• This integrated approach provides a more robust foundation for human gene expression research.