Exprtarget: an integrative approach to predicting human microRNA targets

Area of Science:

• Genetics
• Molecular Biology
• Bioinformatics

Background:

• Gene expression variation influences complex traits and disease susceptibility.
• MicroRNAs (miRNAs) regulate gene expression by binding to mRNA targets.
• Accurate identification of miRNA targets is crucial for understanding gene regulatory networks.

Purpose of the Study:

• To develop a novel computational approach, ExprTarget, for improved miRNA target prediction.
• To integrate existing miRNA target prediction algorithms with genome-wide miRNA and mRNA expression data.
• To create a comprehensive database (ExprTargetDB) of predicted human miRNA targets.

Main Methods:

• Integration of miRanda, PicTar, and TargetScan algorithms.
• Utilized HapMap lymphoblastoid cell line miRNA and mRNA expression datasets.
• Evaluated computational tools using experimentally validated targets from TarBase.
• Analyzed ExprTarget performance using ROC curve analysis and cross-validation.

Main Results:

• ExprTarget significantly enhances miRNA target prediction accuracy (sensitivity and specificity) compared to individual algorithms.
• The study presents the first miRNA expression profiling in HapMap lymphoblastoid cell lines.
• An online database, ExprTargetDB, was developed for accessing predicted human miRNA targets.

Conclusions:

• ExprTarget offers a more sensitive and specific method for miRNA target identification.
• Integrating gene expression data improves the reliability of miRNA target predictions.
• ExprTargetDB provides a valuable resource for researchers studying miRNA-mediated gene regulation.