Coding Potential Calculator
Home Prediction Training Documents Contact




Introduction

miRNAs are ~21-nucleotide noncoding RNAs, some of which are known to sit at the heart of regulating gene expression in plant growth, development and response to environmental biotic and abiotic stresses. The most commonly used method for discovering miRNAs is based on nest-generation sequencing (NGS) technologies. However, this type of experimental method requires the identification of expressed miRNAs and has a limited ability to detect miRNAs that exhibit low, linkage, stress, developmental and/or cell-specific expression. Therefore, computational tools are urgently required to locate the precise mature miRNAs from pre-miRNA sequences. Here, we present an ML-based system with random forest algorithm named miRLocator for the computational prediction of mature miRNAs within plant pre-miRNAs. miRLocator was constructed based on 440 sequence and structural features extracted from miRNA duplexes. A ten-fold cross-validation evaluation using 5854 experimentally validated miRNAs from 19 plant species demonstrated that the prediction performance of miRLocator was comparable to or better than that achieved with the state-of-art miRNA predictor miRdup.



CITE

Cui, H., Zhai, J., & Ma, C. (2015). miRLocator: machine learning-based prediction of mature microRNAs within plant pre-miRNA sequences. PloS one, 10(11), e0142753.

Zhang, T.#, Ju, L.#, Song, Y., Zhai, J., Song, J., & Ma, C. (2018). miRLocator: A python package and web server for predicting miRNAs from pre-miRNA sequences. (Submitted to Methods in Molecular Biology, # Co-first author)