MirLocPredictor: A ConvNet-Based Multi-Label MicroRNA Subcellular Localization Predictor by Incorporating k-Mer Positional Information

Abstract

MicroRNAs (miRNA) are small noncoding RNA sequences consisting of about 22 nucleotides that are involved in the regulation of almost 60% of mammalian genes. Presently, there are very limited approaches for the visualization of miRNA locations present inside cells to support the elucidation of pathways and mechanisms behind miRNA function, transport, and biogenesis. MIRLocator, a state-of-the-art tool for the prediction of subcellular localization of miRNAs makes use of a sequence-to-sequence model along with pretrained k-mer embeddings. Existing pretrained k-mer embedding generation methodologies focus on the extraction of semantics of k-mers. However, in RNA sequences, positional information of nucleotides is more important because distinct positions of the four nucleotides define the function of an RNA molecule. Considering the importance of the nucleotide position, we propose a novel approach (kmerPR2vec) which is a fusion of positional information of k-mers with randomly initialized neural k-mer embeddings. In contrast to existing k-mer-based representation, the proposed kmerPR2vec representation is much more rich in terms of semantic information and has more discriminative power. Using novel kmerPR2vec representation, we further present an end-to-end system (MirLocPredictor) which couples the discriminative power of kmerPR2vec with Convolutional Neural Networks (CNNs) for miRNA subcellular location prediction. The effectiveness of the proposed kmerPR2vec approach is evaluated with deep learning-based topologies (i.e., Convolutional Neural Networks (CNN) and Recurrent Neural Network (RNN)) and by using 9 different evaluation measures. Analysis of the results reveals that MirLocPredictor outperform state-of-the-art methods with a significant margin of 18% and 19% in terms of precision and recall.

Exosome-Mediated miR-21 Was Involved in the Promotion of Structural and Functional Recovery Effect Produced by Electroacupuncture in Sciatic Nerve Injury

“…EA is not only effective in treating PNI [ 53 ] but can also be used for the treatment of neuropathic pain [ 54 ] and central nervous disease [ 55 ]. Since miRNAs are closely associated with the development of PNI [ 15 , 32 , 33 , 41 , 56 – 59 ] and nerve repair, while exosomes are valuable carriers of miRNAs [ 60 ], our study focused on exosomal delivery of miR-21 to explore the effects of EA. Exosomes can minimize safety concerns associated with the administration of live cells as cell-free therapy and has a great therapeutic potential for the treatment of brain, heart, liver, lung, skin, and bone diseases [ 61 ].…”

Histone-Net: a multi-paradigm computational framework for histone occupancy and modification prediction

“…Non-overlapping higher order residues loose the deep correlation information of basic residues, whereas overlapping residues retain deep correla- tion information of residues. Following existing sequence analysis work [41,[50][51][52][53], we generate overlapping higher order residues with 10 different degrees ranging from 2to-12 to comprehensively assess the impact of higher order residues in disentangling rich contextual information for 3 distinct histone sequence analysis tasks.…”

Advances in Computational Methodologies for Classification and Sub-Cellular Locality Prediction of Non-Coding RNAs

“…Up until now, only four methodologies, namely miRGOFS [185], MIRLocator [176], miRnALoc [186], and MirLocPredictor [187], have been proposed for the prediction of miRNA sub-cellular localization. MIRLocator utilized an attention based sequence to sequence a neural network with pre-defined information of label order to discover subcellular localities of human miRNA.…”

“…Apart from semantics, the position of k-mers is another key component that defines the function of RNA molecules. Recently, Asim et al [187] developed a novel sequence descriptor kmerPR2Vec that fused positional information of higher order residues with randomly initialized higher order embedding. Using kmerPR2Vec statistical representation, they developed a deep learning based end-to-end predictive methodology "MirLocPredictor" which utilized CNN for accurate determination of miRNA sub-cellular localization.…”

“…Performance of LncLocator [206] is assessed on a seven sub-cellular locations based dataset, the distribution of which is shown in the top rightmost pie chart (Figure 4). Both MirLocPredictor [187] and miRLocator [176] are evaluated on the same dataset comprised of six sub-cellular locations including Cytoplasm, Nucleus, Circulating, Microvesicle, Exosome, and Mitochondrion. DeepLncRNA [204] is evaluated on a small dataset annotated against only two sub-cellular locations, namely Cytosol and Nucleus.…”