Mammalian MicroRNA Prediction through a Support Vector Machine Model of Sequence and Structure
2007

Predicting Mammalian MicroRNAs Using a Support Vector Machine Model

Sample size: 976746 publication Evidence: moderate

Author Information

Author(s): Sheng Ying, Pär G. Engström, Boris Lenhard

Primary Institution: University of Bergen, Bergen, Norway

Hypothesis

Can a computational method effectively predict new mammalian microRNA candidates based on genomic sequence and structure?

Conclusion

The mirCoS method successfully predicts a significant number of new candidate miRNAs for experimental verification.

Supporting Evidence

  • 3476 mouse candidates and 3441 human candidates were found.
  • 68% of known conserved miRNAs were retained in the predictions.
  • Predictions showed a high level of conservation across vertebrates.

Takeaway

Scientists created a computer program to find tiny RNA molecules in animals that help control genes, and it found many new ones that need to be tested.

Methodology

The study used a support vector machine model to analyze genomic sequences and predict microRNA candidates based on sequence, secondary structure, and conservation.

Potential Biases

The method may miss lowly expressed or tissue-specific miRNAs due to its reliance on conservation.

Limitations

The predictions require experimental validation to confirm the presence of the predicted miRNAs.

Statistical Information

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1371/journal.pone.0000946

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication