Data mining for biomarker development: a review of tissue specificity analysis

Area of Science:

• Biomarker Discovery
• Genomics
• Translational Medicine

Background:

• Biomarker development is resource-intensive, necessitating early selection of high-quality candidates.
• High-throughput gene expression data are commonly used to find differentially expressed transcripts in disease versus normal samples.

Purpose of the Study:

• To outline a strategy for selecting high-quality biomarker candidates early in the development pipeline.
• To leverage existing gene expression databases for enhanced biomarker candidate evaluation.

Main Methods:

• Data-mining various gene expression databases including Expressed Sequence Tag (EST), Serial Analysis of Gene Expression (SAGE), Massively Parallel Signature Sequencing (MPSS), and microarrays.
• Computing quantitative measures of tissue-specific gene expression to assess candidate biomarker specificity.
• Utilizing these quantitative measures to guide the selection of promising biomarker candidates.

Main Results:

• Gene expression databases offer valuable cross-tissue and cross-disease expression information for candidate biomarkers.
• Quantitative assessment of tissue-specific gene expression can effectively differentiate potential biomarker candidates.
• This approach aids in prioritizing candidates with desirable expression profiles for further development.

Conclusions:

• Integrating data-mining of gene expression databases into early biomarker development improves candidate selection efficiency.
• Quantifying tissue specificity is a key metric for identifying robust biomarkers with clinical potential.
• This strategy can reduce resource commitment by focusing on high-quality biomarker candidates from the outset.