MiRPlex

MiRPlex predicts microRNAs from small RNA (sRNA) high-throughput sequencing datasets, enabling genome-independent miRNA discovery in organisms without sequenced genomes.

Key Features:

• Genome-Independent Analysis: Operates without a sequenced genome by predicting miRNA candidates directly from sRNA sequencing data and without requiring identification of flanking hairpin structures from genomic alignments.
• High-Throughput Sequencing Input: Accepts small RNA (sRNA) datasets produced by high-throughput sequencing technologies as primary input.
• Multi-Stage Prediction Process: Implements a pipeline of filtering, miRNA:miRNA* duplex generation, and duplex classification to produce candidate miRNAs.
• Duplex Classification (SVM): Uses a support vector machine to classify predicted miRNA:miRNA* duplexes and distinguish genuine miRNA candidates from other sRNAs.
• Validation on Model Animals: Has been tested on sRNA datasets from model animals to evaluate its classification performance.

Scientific Applications:

• Gene Regulation Studies: Enables identification of novel miRNAs involved in gene regulation via mRNA decay and translational repression, including roles during embryonic development.
• Disease Research: Supports discovery of miRNAs implicated in cancer, cardiovascular disorders, and neurodegenerative diseases to inform studies of disease mechanisms and potential therapeutic targets.

Methodology:

Performs a multi-stage computational workflow comprising filtering of sRNA data, generation of candidate miRNA:miRNA* duplexes, and support vector machine–based duplex classification; the approach has been evaluated on sRNA datasets from model animals.