The prediction of novel pre-microRNA (miRNA) from genomic sequence has received considerable attention recently. However, the majority of studies have focused on the human genome. Previous studies have demonstrated that sensitivity (correctly detecting true miRNA) is sustained when human-trained methods are applied to other species, however they have failed to report the dramatic drop in specificity (the ability to correctly reject non-miRNA sequences) in non-human genomes. Considering the ratio of true miRNA sequences to pseudo-miRNA sequences is on the order of 1:1000, such low specificity prevents the application of most existing tools to non-human genomes, as the number of false positives overwhelms the true predictions. We here introduce a framework (SMIRP) for creating species-specific miRNA prediction systems, leveraging sequence conservation and phylogenetic distance information. Substantial improvements in specificity and precision are obtained for four non-human test species when our framework is applied to three different prediction systems representing two types of classifiers (support vector machine and Random Forest), based on three different feature sets, with both human-specific and taxon-wide training data. The SMIRP framework is potentially applicable to all miRNA prediction systems and we expect substantial improvement in precision and specificity, while sustaining sensitivity, independent of the machine learning technique chosen.

Methods for the de novo identification of microRNA (miRNA) have been developed using a range of sequence-based features. With the increasing availability of next generation sequencing (NGS) transcriptome data, there is a need for miRNA identification that integrates both NGS transcript expression-based patterns as well as advanced genomic sequence-based methods. While miRDeep2 does examine the predicted secondary structure of putative miRNA sequences, it does not leverage many of the sequence-based features used in state-of-the-art de novo methods. Meanwhile, other NGS-based methods, such as miRanalyzer, place an emphasis on sequence-based features without leveraging advanced expression-based features reflecting miRNA biosynthesis. This represents an opportunity to combine the strengths of NGS-based analysis with recent advances in de novo sequence-based miRNA prediction. We here develop a method, microRNA Prediction using Integrated Evidence (miPIE), which integrates both expression-based and sequence-based features to achieve significantly improved miRNA prediction performance. Feature selection identifies the 20 most discriminative features, 3 of which reflect strictly expression-based information. Evaluation using precision-recall curves, for six NGS data sets representing six diverse species, demonstrates substantial improvements in prediction performance compared to three methods: miRDeep2, miRanalyzer, and mirnovo. The individual contributions of expression-based and sequence-based features are also examined and we demonstrate that their combination is more effective than either alone.

No abstract

Methods for the de novo identification of microRNA (miRNA) have been developed using a range of sequence-based features. With the increasing availability of next generation sequencing (NGS) transcriptome data, there is a need for miRNA identification that integrates both NGS transcript expression-based patterns as well as advanced genomic sequence-based methods. While miRDeep2 does examine the predicted secondary structure of putative miRNA sequences, it does not leverage many of the sequence-based features used in state-of-the-art de novo methods.Meanwhile, other NGS-based methods, such as miRanalyzer, place an emphasis on sequence-based features without leveraging advanced expression-based features reflecting miRNA biosynthesis. This represents an opportunity to combine the strengths of NGS-based analysis with recent advances in de novo sequence-based miRNA prediction. We here develop a method, microRNA Prediction using Integrated Evidence (miPIE), which integrates both expression-based and sequence-based features to achieve significantly improved miRNA prediction performance. Feature selection identifies the 20 most discriminative features, 3 of which reflect strictly expression-based information. Evaluation using precision-recall curves, for six NGS data sets representing six diverse species, demonstrates substantial improvements in prediction performance compared to miRDeep2 and miRanalyzer. The individual contributions of expression-based and sequence-based features are also examined and we demonstrate that their combination is more effective than either alone.

No abstract