Accurate microRNA target prediction correlates with protein repression levels

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• MicroRNAs (miRNAs) are non-coding RNA molecules regulating gene expression.
• miRNA-target interactions are crucial for biological processes and disease.
• Experimental identification of miRNA targets is challenging, necessitating computational prediction.

Purpose of the Study:

• To introduce DIANA-microT 3.0, an advanced algorithm for microRNA target prediction.
• To provide a reliable computational tool for deciphering miRNA functions in development and disease.

Main Methods:

• DIANA-microT 3.0 algorithm integrates conserved and non-conserved miRNA recognition elements.
• Prediction accuracy is assessed using signal-to-noise ratio and precision scores.
• The algorithm was benchmarked against experimentally validated miRNA targets.

Main Results:

• DIANA-microT 3.0 calculates prediction scores correlating with protein production changes.
• The algorithm provides signal-to-noise and precision scores to estimate false positive rates.
• DIANA-microT 3.0 demonstrated high precision in predicting miRNA targets.

Conclusions:

• DIANA-microT 3.0 achieved the highest precision (approx. 66%) among tested miRNA target prediction programs.
• The tool offers a user-friendly web server for accessing prediction results.
• DIANA-microT 3.0 is a valuable resource for miRNA research.