TITLE:
Prediction of human microRNA hairpins using only positive sample learning
AUTHORS:
Dang Hung Tran, Tho Hoan Pham, Kenji Satou, Tu Bao Ho
KEYWORDS:
MicroRNA, Hairpin, One-class SVM
JOURNAL NAME:
Journal of Biomedical Science and Engineering,
Vol.1 No.2,
August
25,
2008
ABSTRACT: MicroRNAs(miRNA) are small molecular non-coding RNAs that have important roles in the post-transcriptional mechanism of animal and plant. They are commonly 21-25 nucleotides (nt) long and derived from 60-90 nt RNA hairpin structures, called miRNA hairpins. A larger num-ber of sequence segments in the human genome have been computationally identified with such 60-90 nt hairpins, however a majority of them are not miRNA hairpins. Most computational meth-ods so far for predicting miRNA hairpins were based on a two-class classifier to distinguish between miRNA hairpins and other sequence segments with hairpin structures. The difficulty of these methods is how to select hairpins as negative examples of miRNA hairpins in the classifier-training datasets, since only a few miRNA hairpins are available. Therefore, their classifier may be mis-trained due to some false negative examples of the training dataset. In this paper, we introduce a one-class support vector machine (SVM) method to predict miRNA hair-pins among the hairpin structures. Different from existing methods for predicting miRNA hairpins, the one-class SVM model is trained only on the information of the miRNA class. We also illus-trate some examples of predicting miRNA hair-pins in human chromosomes 10, 15, and 21, where our method overcomes the above disad-vantages of existing two-class methods.