RNA gene discovery through the classification of RNA secondary structure elements

A multi-layered exploratory look at the use of secondary structure elements in ribonucleic acid (RNA) gene finding is performed. Individual structural element metrics are analyzed for their ability to act as structural RNA gene signals. Additionally, each structural element is analyzed for its ability to detect structural RNA gene sequences by training and testing classifiers which utilize the structural element's metrics to classify candidate RNA sequences. Finally, groups of structural elements are examined, by voting the prediction results of the individual structural element classifiers together to determine if a candidate sequence is a structural RNA gene. The tests reveal that the external loop, structure, stemloop, hairpin loop, and tail structural elements produce significant signals for structural RNA genes. Many groups of structural elements were found to have potential but particularly the stemloop and hairpin loop structural element combination stood out for its practicality and strong results.