MicroRNA categorization using sequence motifs and k-mers

Abstract

BackgroundPost-transcriptional gene dysregulation can be a hallmark of diseases like cancer and microRNAs (miRNAs) play a key role in the modulation of translation efficiency. Known pre-miRNAs are listed in miRBase, and they have been discovered in a variety of organisms ranging from viruses and microbes to eukaryotic organisms. The computational detection of pre-miRNAs is of great interest, and such approaches usually employ machine learning to discriminate between miRNAs and other sequences. Many features have been proposed describing pre-miRNAs, and we have previously introduced the use of sequence motifs and k-mers as useful ones. There have been reports of xeno-miRNAs detected via next generation sequencing. However, they may be contaminations and to aid that important decision-making process, we aimed to establish a means to differentiate pre-miRNAs from different species.ResultsTo achieve distinction into species, we used one species’ pre-miRNAs as the positive and another species’ pre-miRNAs as the negative training and test data for the establishment of machine learned models based on sequence motifs and k-mers as features. This approach resulted in higher accuracy values between distantly related species while species with closer relation produced lower accuracy values.ConclusionsWe were able to differentiate among species with increasing success when the evolutionary distance increases. This conclusion is supported by previous reports of fast evolutionary changes in miRNAs since even in relatively closely related species a fairly good discrimination was possible.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-017-1584-1) contains supplementary material, which is available to authorized users.

Categorization of species based on their microRNAs employing sequence motifs, information-theoretic sequence feature extraction, and k-mers

“…Therefore, only sequence-based features were used for parameterization in this study. Sequence motifs (200) as in [31] were used as well as 84 k-mers and their information-theoretic transformations (91). In the following, the parameters used in this study are detailed.…”

“…All hairpins were filtered for sequence similarity as in Yousef et al [31] before training machine learning models using the Usearch tool [47].…”

“…Therefore, both positive and negative classes for training and testing were derived from known pre-miRNAs, effectively removing the need for pseudo negative data. We have previously proposed the same strategy [31] using sequence motifs and k-mers. In this study, we further introduced information-theoretic approaches and important additional analyses.…”

“…However, we realized that using positive examples to represent the negative class from different species holds a number of promises [31]. One of the promises is that it enables the categorization of pre-miRNAs into species.…”

“…Due to the large impact of k-mers on the categorization in our previous work, in an attempt to add more discriminating power, here, we added information theory (IT)-based features. Apart from our previous study [31], only Lopes and colleagues attempted to use pre-miRNAs to discriminate between species [33]. However, they resorted to establishing ab initio pre-miRNA detection models with the same bias on negative data as existing pre-miRNA detection methods [26,[39][40][41][42]; using the same training and testing strategies [32,[42][43][44].…”