HuntMi: an efficient and taxon-specific approach in pre-miRNA identification

Abstract

BackgroundMachine learning techniques are known to be a powerful way of distinguishing microRNA hairpins from pseudo hairpins and have been applied in a number of recognised miRNA search tools. However, many current methods based on machine learning suffer from some drawbacks, including not addressing the class imbalance problem properly. It may lead to overlearning the majority class and/or incorrect assessment of classification performance. Moreover, those tools are effective for a narrow range of species, usually the model ones. This study aims at improving performance of miRNA classification procedure, extending its usability and reducing computational time.ResultsWe present HuntMi, a stand-alone machine learning miRNA classification tool. We developed a novel method of dealing with the class imbalance problem called ROC-select, which is based on thresholding score function produced by traditional classifiers. We also introduced new features to the data representation. Several classification algorithms in combination with ROC-select were tested and random forest was selected for the best balance between sensitivity and specificity. Reliable assessment of classification performance is guaranteed by using large, strongly imbalanced, and taxon-specific datasets in 10-fold cross-validation procedure. As a result, HuntMi achieves a considerably better performance than any other miRNA classification tool and can be applied in miRNA search experiments in a wide range of species.ConclusionsOur results indicate that HuntMi represents an effective and flexible tool for identification of new microRNAs in animals, plants and viruses. ROC-select strategy proves to be superior to other methods of dealing with class imbalance problem and can possibly be used in other machine learning classification tasks. The HuntMi software as well as datasets used in the research are freely available at http://lemur.amu.edu.pl/share/HuntMi/.

Genome-wide discovery of miRNAs using ensembles of machine learning algorithms and logistic regression

“…Of the three tree-based methods included, MiPred (Jiang et al, 2007) and HuntMi (Gudyś et al, 2013) are based on random forests and CID-miRNA (Tyagi et al, 2008) on decision trees. The hairpin classifier used by CID-miRNA is using a stochastic contextfree grammar and is implemented with J48 (Quinlan, 1993).…”

“…Examples of early machine learning classifiers for miRNA discovery are Triplet-SVM (Xue et al, 2005) and mir-abela (Sewer et al, 2005), which are based on Support Vector Machines (SVMs). Subsequently, miRNA prediction algorithms relying on many different classifiers have been published, including Hidden Markov Models (HMMs) (Terai et al, 2007;Agarwal et al, 2010), random forests (Jiang et al, 2007;Gudyś et al, 2013), artificial neural networks (Rahman et al, 2012) and decision trees (Tyagi et al, 2008). The emergence of next-generation sequencing (NGS) technologies led to development of prediction methods using read mapping in genomes.…”

Data mining for microrna gene prediction: On the impact of class imbalance and feature number for microrna gene prediction

“…To be able to examine class imbalance effect, the largest available negative dataset for human was used (http://adaa.polsl.pl/agudys/huntmi/huntmi.htm) [15]. Of this dataset about 50000 sequences were used for testing.…”

miRBoost: boosting support vector machines for microRNA precursor classification

“…Such methods include microPred (Batuwita and Palade 2009), MiRenSVM (Ding et al 2010), mirExplorer (Guan et al 2011), HeteroMirPred (Lertampaiporn et al 2013), and HuntMi (Gudys et al 2013).…”

“…Especially, we failed to validate HuntMi (Gudys et al 2013), which has recently appeared as an efficient tool, implementing ROC-select and applied random forest to get the best balance between sensitivity and specificity. Twenty gigabytes of memory on our machine was not enough to build its models on our data.…”