Fast and accurate clustering of noncoding RNAs using ensembles of sequence alignments and secondary structures

Recent trends in RNA informatics: a review of machine learning and deep learning for RNA secondary structure prediction and RNA drug discovery

Computational analysis of RNA sequences constitutes a crucial step in the field of RNA biology. As in other domains of the life sciences, the incorporation of artificial intelligence and machine learning techniques into RNA sequence analysis has gained significant traction in recent years. Historically, thermodynamics-based methods were widely employed for the prediction of RNA secondary structures; however, machine learning-based approaches have demonstrated remarkable advancements in recent years, enabling more accurate predictions. Consequently, the precision of sequence analysis pertaining to RNA secondary structures, such as RNA-protein interactions, has also been enhanced, making a substantial contribution to the field of RNA biology. Additionally, artificial intelligence and machine learning are also introducing technical innovations in the analysis of RNA-small molecule interactions for RNA-targeted drug discovery and in the design of RNA aptamers, where RNA serves as its own ligand. This review will highlight recent trends in the prediction of RNA secondary structure, RNA aptamers and RNA drug discovery using machine learning, deep learning and related technologies, and will also discuss potential future avenues in the field of RNA informatics.

Reference-based read clustering improves the de novo genome assembly of microbial strains

Constructing accurate microbial genome assemblies is necessary to understand genetic diversity in microbial genomes and its functional consequences. However, it still remains as a challenging task especially when only short-read sequencing technologies are used. Here, we present a new read-clustering algorithm, called RBRC, for improving de novo microbial genome assembly, by accurately estimating read proximity using multiple reference genomes. The performance of RBRC was confirmed by simulation-based evaluation in terms of assembly contiguity and the number of misassemblies, and was successfully applied to existing fungal and bacterial genomes by improving the quality of the assemblies without using additional sequencing data. RBRC is a very useful read-clustering algorithm that can be used (i) for generating high-quality genome assemblies of microbial strains when genome assemblies of related strains are available, and (ii) for upgrading existing microbial genome assemblies when the generation of additional sequencing data, such as long reads, is difficult.

Informative RNA base embedding for RNA structural alignment and clustering by deep representation learning

Effective embedding is actively conducted by applying deep learning to biomolecular information. Obtaining better embeddings enhances the quality of downstream analyses, such as DNA sequence motif detection and protein function prediction. In this study, we adopt a pre-training algorithm for the effective embedding of RNA bases to acquire semantically rich representations and apply this algorithm to two fundamental RNA sequence problems: structural alignment and clustering. By using the pre-training algorithm to embed the four bases of RNA in a position-dependent manner using a large number of RNA sequences from various RNA families, a context-sensitive embedding representation is obtained. As a result, not only base information but also secondary structure and context information of RNA sequences are embedded for each base. We call this ‘informative base embedding’ and use it to achieve accuracies superior to those of existing state-of-the-art methods on RNA structural alignment and RNA family clustering tasks. Furthermore, upon performing RNA sequence alignment by combining this informative base embedding with a simple Needleman–Wunsch alignment algorithm, we succeed in calculating structural alignments with a time complexity of O(n²) instead of the O(n⁶) time complexity of the naive implementation of Sankoff-style algorithm for input RNA sequence of length n.

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

Apart from protein-coding Ribonucleic acids (RNAs), there exists a variety of non-coding RNAs (ncRNAs) which regulate complex cellular and molecular processes. High-throughput sequencing technologies and bioinformatics approaches have largely promoted the exploration of ncRNAs which revealed their crucial roles in gene regulation, miRNA binding, protein interactions, and splicing. Furthermore, ncRNAs are involved in the development of complicated diseases like cancer. Categorization of ncRNAs is essential to understand the mechanisms of diseases and to develop effective treatments. Sub-cellular localization information of ncRNAs demystifies diverse functionalities of ncRNAs. To date, several computational methodologies have been proposed to precisely identify the class as well as sub-cellular localization patterns of RNAs). This paper discusses different types of ncRNAs, reviews computational approaches proposed in the last 10 years to distinguish coding-RNA from ncRNA, to identify sub-types of ncRNAs such as piwi-associated RNA, micro RNA, long ncRNA, and circular RNA, and to determine sub-cellular localization of distinct ncRNAs and RNAs. Furthermore, it summarizes diverse ncRNA classification and sub-cellular localization determination datasets along with benchmark performance to aid the development and evaluation of novel computational methodologies. It identifies research gaps, heterogeneity, and challenges in the development of computational approaches for RNA sequence analysis. We consider that our expert analysis will assist Artificial Intelligence researchers with knowing state-of-the-art performance, model selection for various tasks on one platform, dominantly used sequence descriptors, neural architectures, and interpreting inter-species and intra-species performance deviation.

Deep forest ensemble learning for classification of alignments of non-coding RNA sequences based on multi-view structure representations

Non-coding RNAs (ncRNAs) play crucial roles in multiple biological processes. However, only a few ncRNAs’ functions have been well studied. Given the significance of ncRNAs classification for understanding ncRNAs’ functions, more and more computational methods have been introduced to improve the classification automatically and accurately. In this paper, based on a convolutional neural network and a deep forest algorithm, multi-grained cascade forest (GcForest), we propose a novel deep fusion learning framework, GcForest fusion method (GCFM), to classify alignments of ncRNA sequences for accurate clustering of ncRNAs. GCFM integrates a multi-view structure feature representation including sequence-structure alignment encoding, structure image representation and shape alignment encoding of structural subunits, enabling us to capture the potential specificity between ncRNAs. For the classification of pairwise alignment of two ncRNA sequences, the F-value of GCFM improves 6% than an existing alignment-based method. Furthermore, the clustering of ncRNA families is carried out based on the classification matrix generated from GCFM. Results suggest better performance (with 20% accuracy improved) than existing ncRNA clustering methods (RNAclust, Ensembleclust and CNNclust). Additionally, we apply GCFM to construct a phylogenetic tree of ncRNA and predict the probability of interactions between RNAs. Most ncRNAs are located correctly in the phylogenetic tree, and the prediction accuracy of RNA interaction is 90.63%. A web server (http://bmbl.sdstate.edu/gcfm/) is developed to maximize its availability, and the source code and related data are available at the same URL.

The rare lncRNA GOLLD is widespread and structurally conserved among Mycobacterium tRNA arrays

Noncoding RNA (ncRNA) genes produce transcripts involved in a wide range of functions, including catalytic and regulatory functions. Besides, some transcripts have highly complex structures that may impact their activities. Among the largest bacterial ncRNAs, there is the rare GOLLD RNA, which is associated with tRNA genes and supposed to be chromosome- and phage-encoded in specialized groups of bacteria, including those from Lactobacillales and Actinomycetales orders. The only GOLLD structure was inferred from a variety of sequences, including many marine metagenomes. To explore GOLLD RNA in bacterial genomes, we mined the GOLLD gene in thousands of Mycobacterium and virus genomes using Infernal software. We identified this gene in 350 mycobacteria, including megaplasmids, and 39 bacteriophages, mainly in the genomic context of tRNA arrays. Mycobacterium GOLLD genes presented a high diversity and were distributed in three phylogenetic groups: (i) Mycobacterium exclusive; (ii) Mycobacterium and mycobacteriophages; and (iii) mycobacteriophage exclusive. We also determined the GOLLD secondary structure of each group using R2 R software based on GOLLD alignments generated by Infernal software. All GOLLD groups displayed a 3' half conserved structure, including utter E-loops pseudoknots substructures, also shared by non-Mycobacterium GOLLD while the 5' half motif was different among the groups. Here, we showed that the lncRNA GOLLD is widespread in Mycobacterium within tRNA arrays and corroborated the previously predicted GOLLD secondary structure.

Convolutional neural networks for classification of alignments of non-coding RNA sequences

Motivation

The convolutional neural network (CNN) has been applied to the classification problem of DNA sequences, with the additional purpose of motif discovery. The training of CNNs with distributed representations of four nucleotides has successfully derived position weight matrices on the learned kernels that corresponded to sequence motifs such as protein-binding sites.

Results

We propose a novel application of CNNs to classification of pairwise alignments of sequences for accurate clustering of sequences and show the benefits of the CNN method of inputting pairwise alignments for clustering of non-coding RNA (ncRNA) sequences and for motif discovery. Classification of a pairwise alignment of two sequences into positive and negative classes corresponds to the clustering of the input sequences. After we combined the distributed representation of RNA nucleotides with the secondary-structure information specific to ncRNAs and furthermore with mapping profiles of next-generation sequence reads, the training of CNNs for classification of alignments of RNA sequences yielded accurate clustering in terms of ncRNA families and outperformed the existing clustering methods for ncRNA sequences. Several interesting sequence motifs and secondary-structure motifs known for the snoRNA family and specific to microRNA and tRNA families were identified.

Availability and implementation

The source code of our CNN software in the deep-learning framework Chainer is available at http://www.dna.bio.keio.ac.jp/cnn/, and the dataset used for performance evaluation in this work is available at the same URL.

Accurate Classification of RNA Structures Using Topological Fingerprints

While RNAs are well known to possess complex structures, functionally similar RNAs often have little sequence similarity. While the exact size and spacing of base-paired regions vary, functionally similar RNAs have pronounced similarity in the arrangement, or topology, of base-paired stems. Furthermore, predicted RNA structures often lack pseudoknots (a crucial aspect of biological activity), and are only partially correct, or incomplete. A topological approach addresses all of these difficulties. In this work we describe each RNA structure as a graph that can be converted to a topological spectrum (RNA fingerprint). The set of subgraphs in an RNA structure, its RNA fingerprint, can be compared with the fingerprints of other RNA structures to identify and correctly classify functionally related RNAs. Topologically similar RNAs can be identified even when a large fraction, up to 30%, of the stems are omitted, indicating that highly accurate structures are not necessary. We investigate the performance of the RNA fingerprint approach on a set of eight highly curated RNA families, with diverse sizes and functions, containing pseudoknots, and with little sequence similarity-an especially difficult test set. In spite of the difficult test set, the RNA fingerprint approach is very successful (ROC AUC > 0.95). Due to the inclusion of pseudoknots, the RNA fingerprint approach both covers a wider range of possible structures than methods based only on secondary structure, and its tolerance for incomplete structures suggests that it can be applied even to predicted structures. Source code is freely available at https://github.rcac.purdue.edu/mgribsko/XIOS_RNA_fingerprint.

SPARSE: quadratic time simultaneous alignment and folding of RNAs without sequence-based heuristics

Motivation: RNA-Seq experiments have revealed a multitude of novel ncRNAs. The gold standard for their analysis based on simultaneous alignment and folding suffers from extreme time complexity of O(n6). Subsequently, numerous faster ‘Sankoff-style’ approaches have been suggested. Commonly, the performance of such methods relies on sequence-based heuristics that restrict the search space to optimal or near-optimal sequence alignments; however, the accuracy of sequence-based methods breaks down for RNAs with sequence identities below 60%. Alignment approaches like LocARNA that do not require sequence-based heuristics, have been limited to high complexity (≥ quartic time).

Results: Breaking this barrier, we introduce the novel Sankoff-style algorithm ‘sparsified prediction and alignment of RNAs based on their structure ensembles (SPARSE)’, which runs in quadratic time without sequence-based heuristics. To achieve this low complexity, on par with sequence alignment algorithms, SPARSE features strong sparsification based on structural properties of the RNA ensembles. Following PMcomp, SPARSE gains further speed-up from lightweight energy computation. Although all existing lightweight Sankoff-style methods restrict Sankoff’s original model by disallowing loop deletions and insertions, SPARSE transfers the Sankoff algorithm to the lightweight energy model completely for the first time. Compared with LocARNA, SPARSE achieves similar alignment and better folding quality in significantly less time (speedup: 3.7). At similar run-time, it aligns low sequence identity instances substantially more accurate than RAF, which uses sequence-based heuristics.

Availability and implementation: SPARSE is freely available at http://www.bioinf.uni-freiburg.de/Software/SPARSE.

Contact:backofen@informatik.uni-freiburg.de

Supplementary information:Supplementary data are available at Bioinformatics online.

BlockClust: efficient clustering and classification of non-coding RNAs from short read RNA-seq profiles

Summary: Non-coding RNAs (ncRNAs) play a vital role in many cellular processes such as RNA splicing, translation, gene regulation. However the vast majority of ncRNAs still have no functional annotation. One prominent approach for putative function assignment is clustering of transcripts according to sequence and secondary structure. However sequence information is changed by post-transcriptional modifications, and secondary structure is only a proxy for the true 3D conformation of the RNA polymer. A different type of information that does not suffer from these issues and that can be used for the detection of RNA classes, is the pattern of processing and its traces in small RNA-seq reads data. Here we introduce BlockClust, an efficient approach to detect transcripts with similar processing patterns. We propose a novel way to encode expression profiles in compact discrete structures, which can then be processed using fast graph-kernel techniques. We perform both unsupervised clustering and develop family specific discriminative models; finally we show how the proposed approach is scalable, accurate and robust across different organisms, tissues and cell lines.

Availability: The whole BlockClust galaxy workflow including all tool dependencies is available at http://toolshed.g2.bx.psu.edu/view/rnateam/blockclust_workflow.

Contact:backofen@informatik.uni-freiburg.de; costa@informatik.uni-freiburg.de

Supplementary information:Supplementary data are available at Bioinformatics online.

A data parallel strategy for aligning multiple biological sequences on multi-core computers

In this paper, we address the large-scale biological sequence alignment problem, which has an increasing demand in computational biology. We employ data parallelism paradigm that is suitable for handling large-scale processing on multi-core computers to achieve a high degree of parallelism. Using the data parallelism paradigm, we propose a general strategy which can be used to speed up any multiple sequence alignment method. We applied five different clustering algorithms in our strategy and implemented rigorous tests on an 8-core computer using four traditional benchmarks and artificially generated sequences. The results show that our multi-core-based implementations can achieve up to 151-fold improvements in execution time while losing 2.19% accuracy on average. The source code of the proposed strategy, together with the test sets used in our analysis, is available on request.

GraphClust: alignment-free structural clustering of local RNA secondary structures

Motivation: Clustering according to sequence–structure similarity has now become a generally accepted scheme for ncRNA annotation. Its application to complete genomic sequences as well as whole transcriptomes is therefore desirable but hindered by extremely high computational costs.

Results: We present a novel linear-time, alignment-free method for comparing and clustering RNAs according to sequence and structure. The approach scales to datasets of hundreds of thousands of sequences. The quality of the retrieved clusters has been benchmarked against known ncRNA datasets and is comparable to state-of-the-art sequence–structure methods although achieving speedups of several orders of magnitude. A selection of applications aiming at the detection of novel structural ncRNAs are presented. Exemplarily, we predicted local structural elements specific to lincRNAs likely functionally associating involved transcripts to vital processes of the human nervous system. In total, we predicted 349 local structural RNA elements.

Availability: The GraphClust pipeline is available on request.

Contact:backofen@informatik.uni-freiburg.de

Supplementary information:Supplementary data are available at Bioinformatics online.

The H-factor as a novel quality metric for homology modeling

Background

Drug discovery typically starts with the identification of a potential target that is then tested and validated either through high-throughput screening against a library of drug compounds or by rational drug design. When the putative target is a protein, the latter approach requires the knowledge of its structure. Finding the structure of a protein is however a difficult task. Significant progress has come from high-resolution techniques such as X-ray crystallography and NMR; there are many proteins however whose structure have not yet been solved. Computational techniques for structure prediction are viable alternatives to experimental techniques for these cases. However, the proper validation of the structural models they generate remains an issue.

Findings

In this report, we focus on homology modeling techniques and introduce the H-factor, a new indicator for assessing the quality of protein structure models generated with these techniques. The H-factor is meant to mimic the R-factor used in X-ray crystallography. The method for computing the H-factor is fully described with a demonstration of its effectiveness on a test set of target proteins.

Conclusions

We have developed a web service for computing the H-factor for models of a protein structure. This service is freely accessible at http://koehllab.genomecenter.ucdavis.edu/toolkit/h-factor.