Informatic selection of a neural crest-melanocyte cDNA set for microarray analysis.

Area of Science:

• Genomics and Bioinformatics
• Molecular Biology
• Developmental Biology

Background:

• Simultaneous gene expression comparison is possible using cDNA microarrays.
• Selecting tissue-appropriate cDNA sets enhances the discovery of altered gene expression.
• The expressed sequence tag (EST) database offers extensive sequence information.

Purpose of the Study:

• To leverage the dbEST database for identifying a neural crest-derived melanocyte cDNA set for microarray analysis.
• To develop and validate an approach for selecting tissue-appropriate cDNAs for gene expression studies.

Main Methods:

• Utilized dbEST to identify a cDNA library (library 198) enriched for neural crest-expressed genes.
• Selected 852 clustered ESTs from library 198 based on their tissue distribution profile.
• Performed microarray analysis comparing the Mel1 array (neural crest-selected) against a non-biased array.

Main Results:

• Library 198 showed a significant bias towards neural crest-expressed genes compared to ubiquitously expressed control genes.
• The Mel1 array (852 ESTs) revealed significant differential gene expression in melanoma cell lines versus kidney epithelial cells (P < 1 x 10(-8)).
• A control array of 1,238 ESTs selected without library bias did not show significant differential expression (P = 0.204).

Conclusions:

• The study presents a validated approach for selecting tissue-specific cDNAs using EST database information.
• This method enables the examination of gene expression profiles relevant to developmental processes and diseases, such as melanoma.
• Tissue-appropriate cDNA selection significantly improves the ability to detect biologically relevant gene expression changes.