BosFinder: a novel pre-microRNA gene prediction algorithm in Bos taurus.

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• MicroRNAs (miRNAs) are small non-coding RNAs regulating gene expression with vital roles in plant and animal development.
• Bos taurus (cattle) is economically significant livestock, but its miRNA repertoire is underexplored due to recent genome assembly.
• Computational methods, particularly machine learning, are crucial for identifying novel miRNA genes from genomic data.

Purpose of the Study:

• To develop and apply a machine-learning-based algorithm for identifying and annotating pre-microRNA (miRNA) genes in the Bos taurus genome.
• To conduct the first comprehensive genome-wide survey of pre-miRNAs in cattle using computational approaches.
• To make the developed miRNA identification tool publicly available for further research.

Main Methods:

• Utilized public bovine expressed sequence tag (EST) data from various tissues and developmental stages.
• Developed a novel algorithm, BosFinder, employing a support vector machine (SVM) model.
• The SVM model was trained to classify pre-miRNA genes based on 33 structural and thermodynamic features.

Main Results:

• Identified a substantial number of 18,776 highly potential pre-miRNA sequences within the Bos taurus genome.
• BosFinder successfully annotated pre-miRNA genes across the bovine genome.
• This study represents the first genome-wide survey of pre-miRNAs in Bos taurus using a machine-learning approach.

Conclusions:

• The BosFinder algorithm provides an effective computational tool for discovering pre-miRNA genes in the Bos taurus genome.
• This work significantly expands the known repertoire of bovine miRNAs, facilitating future research into their functions.
• The findings contribute to a better understanding of cattle biology, with implications for livestock improvement and industry.