Semi-supervised miRNA

The software tool 'Semi-supervised miRNA' addresses the resource-intensive process of microRNA (miRNA) identification by combining two semi-supervised machine learning approaches: active learning and multi-view co-training. Unlike traditional methods relying on large labeled training datasets, this tool maximizes the use of both labeled and abundant unlabeled RNA sequence data. The multi-stage semi-supervised approach demonstrates improved miRNA classification performance across six diverse species, effectively leveraging minimal labeled instances.

Topic

Functional, regulatory and non-coding RNA;Machine learning;Sequencing;Microarray experiment;Gene transcripts

Detail

  • Operation: Splitting;miRNA target prediction;miRNA expression analysis

  • Software interface: Command-line user interface

  • Language: Python

  • License: -

  • Cost: Free

  • Version name: -

  • Credit: The Natural Sciences and Engineering Research Council, Canada.

  • Input: -

  • Output: -

  • Contact: James R. Green jrgreen@sce.carleton.ca

  • Collection: -

  • Maturity: -

Publications

  • A semi-supervised machine learning framework for microRNA classification.
  • Sheikh Hassani M and Green JR. A semi-supervised machine learning framework for microRNA classification. A semi-supervised machine learning framework for microRNA classification. 2019; 13:43. doi: 10.1186/s40246-019-0221-7
  • https://doi.org/10.1186/S40246-019-0221-7
  • PMID: 31639051
  • PMC: PMC6805288

Download and documentation

< Back to DB search