Dinucleotide controlled null models for comparative RNA gene prediction
2008

New Algorithm for RNA Gene Prediction

Sample size: 1000 publication 10 minutes Evidence: moderate

Author Information

Author(s): Tanja Gesell, Stefan Washietl

Primary Institution: Center for Integrative Bioinformatics Vienna

Hypothesis

Can we create a dinucleotide-controlled model for RNA gene prediction that reduces false positives?

Conclusion

The SISSIz algorithm provides a more accurate method for RNA gene prediction by preserving dinucleotide content.

Supporting Evidence

  • The new algorithm was tested on vertebrate genomic alignments.
  • SISSIz can produce negative controls for machine learning-based programs.
  • Using SISSIz, the false positive rate in RNA predictions was found to be three times higher than with mononucleotide controls.

Takeaway

Scientists created a new computer program that helps find RNA genes more accurately by keeping track of certain DNA patterns.

Methodology

The study developed a program called SISSIz that simulates multiple alignments while preserving dinucleotide content.

Potential Biases

The algorithm may still underestimate false positive rates in RNA predictions.

Limitations

The model may lose some signal in true structured RNAs due to its conservative nature.

Statistical Information

P-Value

0.01

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2105-9-248

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication