RNAstrand: Predicting Reading Direction of Structured RNAs
Author Information
Author(s): Kristin Reiche, Peter F. Stadler
Primary Institution: University of Leipzig
Hypothesis
Can a support vector machine accurately classify the reading direction of structured RNAs from multiple sequence alignments?
Conclusion
The RNAstrand tool significantly improves the classification accuracy of structured RNAs' reading directions compared to previous methods.
Supporting Evidence
- RNAstrand achieves classification accuracies of 90% and above for most ncRNA families.
- The tool is integrated into the RNAz package, enhancing its usability.
- RNAstrand provides a two-fold reduction in misclassifications compared to naive approaches.
Takeaway
RNAstrand is a computer program that helps scientists figure out which way a piece of RNA is read, which is important for understanding how it works.
Methodology
The study used a support vector machine to classify RNA reading directions based on folding energies and structural conservation.
Potential Biases
The SVM model may be biased due to the training data being independent in different classes.
Limitations
RNAstrand may not perform well on alignments not recognized as structured RNA by RNAz.
Statistical Information
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website