miRBind: A Deep Learning Method for miRNA Binding Classification

MiRNAs as potential therapeutic targets and biomarkers for non-traumatic intracerebral hemorrhage

Non-traumatic intracerebral hemorrhage (ICH) is the most common type of hemorrhagic stroke, most often occurring between the ages of 45 and 60. Hypertension is most often the cause of ICH. Less often, atherosclerosis, blood diseases, inflammatory changes in cerebral vessels, intoxication, vitamin deficiencies, and other reasons cause hemorrhages. Cerebral hemorrhage can occur by diapedesis or as a result of a ruptured vessel. This very dangerous disease is difficult to treat, requires surgery and can lead to disability or death. MicroRNAs (miRNAs) are a class of non-coding RNAs (about 18-22 nucleotides) that are involved in a variety of biological processes including cell differentiation, proliferation, apoptosis, etc., through gene repression. A growing number of studies have demonstrated miRNAs deregulation in various cardiovascular diseases, including ICH. In addition, given that computed tomography (CT) and/or magnetic resonance imaging (MRI) are either not available or do not show clear signs of possible vessel rupture, accurate and reliable analysis of circulating miRNAs in biological fluids can help in early diagnosis for prevention of ICH and prognosis patient outcome after hemorrhage. In this review, we highlight the up-to-date findings on the deregulated miRNAs in ICH, and the potential use of miRNAs in clinical settings, such as therapeutic targets and non-invasive diagnostic/prognostic biomarker tools.

Exploring miRNA-target gene pair detection in disease with coRmiT

A wide range of approaches can be used to detect micro RNA (miRNA)-target gene pairs (mTPs) from expression data, differing in the ways the gene and miRNA expression profiles are calculated, combined and correlated. However, there is no clear consensus on which is the best approach across all datasets. Here, we have implemented multiple strategies and applied them to three distinct rare disease datasets that comprise smallRNA-Seq and RNA-Seq data obtained from the same samples, obtaining mTPs related to the disease pathology. All datasets were preprocessed using a standardized, freely available computational workflow, DEG_workflow. This workflow includes coRmiT, a method to compare multiple strategies for mTP detection. We used it to investigate the overlap of the detected mTPs with predicted and validated mTPs from 11 different databases. Results show that there is no clear best strategy for mTP detection applicable to all situations. We therefore propose the integration of the results of the different strategies by selecting the one with the highest odds ratio for each miRNA, as the optimal way to integrate the results. We applied this selection-integration method to the datasets and showed it to be robust to changes in the predicted and validated mTP databases. Our findings have important implications for miRNA analysis. coRmiT is implemented as part of the ExpHunterSuite Bioconductor package available from https://bioconductor.org/packages/ExpHunterSuite.

Analysis of chimeric reads characterises the diverse targetome of AGO2-mediated regulation

Argonaute proteins are instrumental in regulating RNA stability and translation. AGO2, the major mammalian Argonaute protein, is known to primarily associate with microRNAs, a family of small RNA 'guide' sequences, and identifies its targets primarily via a 'seed' mediated partial complementarity process. Despite numerous studies, a definitive experimental dataset of AGO2 'guide'-'target' interactions remains elusive. Our study employs two experimental methods-AGO2 CLASH and AGO2 eCLIP, to generate thousands of AGO2 target sites verified by chimeric reads. These chimeric reads contain both the AGO2 loaded small RNA 'guide' and the target sequence, providing a robust resource for modeling AGO2 binding preferences. Our novel analysis pipeline reveals thousands of AGO2 target sites driven by microRNAs and a significant number of AGO2 'guides' derived from fragments of other small RNAs such as tRNAs, YRNAs, snoRNAs, rRNAs, and more. We utilize convolutional neural networks to train machine learning models that accurately predict the binding potential for each 'guide' class and experimentally validate several interactions. In conclusion, our comprehensive analysis of the AGO2 targetome broadens our understanding of its 'guide' repertoire and potential function in development and disease. Moreover, we offer practical bioinformatic tools for future experiments and the prediction of AGO2 targets. All data and code from this study are freely available at https://github.com/ML-Bioinfo-CEITEC/HybriDetector/ .

Using Attribution Sequence Alignment to Interpret Deep Learning Models for miRNA Binding Site Prediction

MicroRNAs (miRNAs) are small non-coding RNAs that play a central role in the post-transcriptional regulation of biological processes. miRNAs regulate transcripts through direct binding involving the Argonaute protein family. The exact rules of binding are not known, and several in silico miRNA target prediction methods have been developed to date. Deep learning has recently revolutionized miRNA target prediction. However, the higher predictive power comes with a decreased ability to interpret increasingly complex models. Here, we present a novel interpretation technique, called attribution sequence alignment, for miRNA target site prediction models that can interpret such deep learning models on a two-dimensional representation of miRNA and putative target sequence. Our method produces a human readable visual representation of miRNA:target interactions and can be used as a proxy for the further interpretation of biological concepts learned by the neural network. We demonstrate applications of this method in the clustering of experimental data into binding classes, as well as using the method to narrow down predicted miRNA binding sites on long transcript sequences. Importantly, the presented method works with any neural network model trained on a two-dimensional representation of interactions and can be easily extended to further domains such as protein–protein interactions.