miRNAfe: A comprehensive tool for feature extraction in microRNA prediction

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• MicroRNA (miRNA) identification and analysis are crucial for understanding gene regulation.
• Existing tools for miRNA feature extraction are often fragmented or lack comprehensive functionality.
• A unified, accessible platform for extracting diverse miRNA-related features is needed.

Purpose of the Study:

• To develop and present miRNAfe, a web-based tool for comprehensive feature extraction from RNA sequences.
• To provide a user-friendly interface for accessing state-of-the-art feature extraction methods.
• To facilitate downstream analysis and prediction tasks in miRNA research.

Main Methods:

• Development of a web service integrating multiple feature extraction algorithms.
• Implementation of a simple user interface for sequence file upload and feature selection.
• Categorization of up to 80 features into six distinct groups: primary sequence, secondary structure, thermodynamic stability, statistical stability, interspecies conservation, and substring analysis.
• Validation of extraction algorithms against original author software.

Main Results:

• miRNAfe successfully extracts a wide array of features (up to 80) from RNA sequences.
• Features are categorized for simplified analysis, covering sequence, structure, stability, conservation, and substrings.
• Pre-trained species-specific classifiers are available for miRNA prediction.
• Validated algorithms ensure accurate and reliable feature extraction.

Conclusions:

• miRNAfe offers a valuable, freely accessible resource for the bioinformatics community.
• The tool simplifies and unifies the process of miRNA feature extraction.
• It serves as a crucial input for various miRNA analysis and prediction pipelines.