Splice Site Prediction Research Articles

Predicting DNA sequence splice site based on graph convolutional network and DNA graph construction

Identifying splice sites is essential for gene structure analysis and eukaryotic genome annotation. Recently, computational and deep learning approaches for splice site detection have advanced, focusing on reducing false positives by distinguishing true from pseudo splice sites. This paper introduces GraphSplice, a method using graph convolutional neural networks. It encodes DNA sequences into directed graphs to extract features and predict splice sites. Tested across multiple datasets, GraphSplice consistently achieved high accuracy (91%-94%) and F1Scores (92%-94%), outperforming state-of-the-art models by up to 9.16% for donors and 5.64% for acceptors. Cross-species experiments also show GraphSplice’s capability to annotate splice sites in under-trained genomic datasets, proving its wide applicability as a tool for DNA splice site analysis.

Self-supervised learning on millions of primary RNA sequences from 72 vertebrates improves sequence-based RNA splicing prediction.

Language models pretrained by self-supervised learning (SSL) have been widely utilized to study protein sequences, while few models were developed for genomic sequences and were limited to single species. Due to the lack of genomes from different species, these models cannot effectively leverage evolutionary information. In this study, we have developed SpliceBERT, a language model pretrained on primary ribonucleic acids (RNA) sequences from 72 vertebrates by masked language modeling, and applied it to sequence-based modeling of RNA splicing. Pretraining SpliceBERT on diverse species enables effective identification of evolutionarily conserved elements. Meanwhile, the learned hidden states and attention weights can characterize the biological properties of splice sites. As a result, SpliceBERT was shown effective on several downstream tasks: zero-shot prediction of variant effects on splicing, prediction of branchpoints in humans, and cross-species prediction of splice sites. Our study highlighted the importance of pretraining genomic language models on a diverse range of species and suggested that SSL is a promising approach to enhance our understanding of the regulatory logic underlying genomic sequences.

DRANetSplicer: A Splice Site Prediction Model Based on Deep Residual Attention Networks.

The precise identification of splice sites is essential for unraveling the structure and function of genes, constituting a pivotal step in the gene annotation process. In this study, we developed a novel deep learning model, DRANetSplicer, that integrates residual learning and attention mechanisms for enhanced accuracy in capturing the intricate features of splice sites. We constructed multiple datasets using the most recent versions of genomic data from three different organisms, Oryza sativa japonica, Arabidopsis thaliana and Homo sapiens. This approach allows us to train models with a richer set of high-quality data. DRANetSplicer outperformed benchmark methods on donor and acceptor splice site datasets, achieving an average accuracy of (96.57%, 95.82%) across the three organisms. Comparative analyses with benchmark methods, including SpliceFinder, Splice2Deep, Deep Splicer, EnsembleSplice, and DNABERT, revealed DRANetSplicer's superior predictive performance, resulting in at least a (4.2%, 11.6%) relative reduction in average error rate. We utilized the DRANetSplicer model trained on O. sativa japonica data to predict splice sites in A. thaliana, achieving accuracies for donor and acceptor sites of (94.89%, 94.25%). These results indicate that DRANetSplicer possesses excellent cross-organism predictive capabilities, with its performance in cross-organism predictions even surpassing that of benchmark methods in non-cross-organism predictions. Cross-organism validation showcased DRANetSplicer's excellence in predicting splice sites across similar organisms, supporting its applicability in gene annotation for understudied organisms. We employed multiple methods to visualize the decision-making process of the model. The visualization results indicate that DRANetSplicer can learn and interpret well-known biological features, further validating its overall performance. Our study systematically examined and confirmed the predictive ability of DRANetSplicer from various levels and perspectives, indicating that its practical application in gene annotation is justified.

Splice site prediction research based on location information

Reveal the mysteries of birth, death and so life has become one of the main purpose of bioinformatics, splice site prediction is one of the most important part, however, not been able to get this problem solved. Firstly, the third generation of genetic markers of single nucleotide polymorphisms had been used in that research to explore the influence of the SNP in splicing; Secondly, a modified hidden Markov model has been introduced; finally, experiments show that the SNP for the performance has a certain influence. In addition, location information based hidden Markov model designed also has positive effects. This method increases the effects dramatically than currently used methods.

TAGET: a toolkit for analyzing full-length transcripts from long-read sequencing

Single-molecule Real-time Isoform Sequencing (Iso-seq) of transcriptomes by PacBio can generate very long and accurate reads, thus providing an ideal platform for full-length transcriptome analysis. We present an integrated computational toolkit named TAGET for Iso-seq full-length transcript data analyses, including transcript alignment, annotation, gene fusion detection, and quantification analyses such as differential expression gene analysis and differential isoform usage analysis. We evaluate the performance of TAGET using a public Iso-seq dataset and newly sequenced Iso-seq datasets from tumor patients. TAGET gives significantly more precise novel splice site prediction and enables more accurate novel isoform and gene fusion discoveries, as validated by experimental validations and comparisons with RNA-seq data. We identify and experimentally validate a differential isoform usage gene ECM1, and further show that its isoform ECM1b may be a tumor-suppressor in laryngocarcinoma. Our results demonstrate that TAGET provides a valuable computational toolkit and can be applied to many full-length transcriptome studies.

Longitudinal Characterization of the Clinical Course of Intermediate-Severe Salla Disease

BackgroundBiallelic pathogenic variants in SLC17A5 cause three forms of free sialic acid storage disease categorized based on severity from least to most severe: Salla disease, intermediate-severe Salla disease, and infantile free sialic acid storage disease. Intermediate-severe Salla disease is the most recently described form. Here, we report a longitudinal characterization of intermediate-severe Salla disease progression in two sisters carrying the following biallelic variants in SLC17A5: c.406A>G (p.Lys136Glu) and c.819+1G>A. MethodsA retrospective review of medical records was performed. A developmental questionnaire was completed to obtain further clinical information. For functional characterization of the predicted splice site variant, RNA was extracted from patient blood samples and sequenced. ResultsDisease onset occurred within the first six months of life in both patients. Early childhood development was delayed with achievement of some milestones followed by a developmental plateau in late childhood. After this, both patients began a slow and progressive neurological regression in adolescence. Functional studies confirmed the pathogenicity of the c.819+1G>A variant, resulting in a frameshift and deletion of exon 6. ConclusionsWe present a detailed study describing the clinical course of intermediate-severe Salla disease with over 15 to 20 years of evolution and demonstrate the pathogenicity of the c.819+1G>A splice site variant.

An automated framework for evaluation of deep learning models for splice site predictions

A novel framework for the automated evaluation of various deep learning-based splice site detectors is presented. The framework eliminates time-consuming development and experimenting activities for different codebases, architectures, and configurations to obtain the best models for a given RNA splice site dataset. RNA splicing is a cellular process in which pre-mRNAs are processed into mature mRNAs and used to produce multiple mRNA transcripts from a single gene sequence. Since the advancement of sequencing technologies, many splice site variants have been identified and associated with the diseases. So, RNA splice site prediction is essential for gene finding, genome annotation, disease-causing variants, and identification of potential biomarkers. Recently, deep learning models performed highly accurately for classifying genomic signals. Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) and its bidirectional version (BLSTM), Gated Recurrent Unit (GRU), and its bidirectional version (BGRU) are promising models. During genomic data analysis, CNN’s locality feature helps where each nucleotide correlates with other bases in its vicinity. In contrast, BLSTM can be trained bidirectionally, allowing sequential data to be processed from forward and reverse directions. Therefore, it can process 1-D encoded genomic data effectively. Even though both methods have been used in the literature, a performance comparison was missing. To compare selected models under similar conditions, we have created a blueprint for a series of networks with five different levels. As a case study, we compared CNN and BLSTM models’ learning capabilities as building blocks for RNA splice site prediction in two different datasets. Overall, CNN performed better with 92% accuracy (6% improvement), 89% F1 score (8% improvement), and 96% AUC-PR (4% improvement) in human splice site prediction. Likewise, an outperforming performance with 96% accuracy (11% improvement), 94% F1 score (16% improvement), and 99% AUC-PR (7% improvement) is achieved in C. elegans splice site prediction. Overall, our results showed that CNN learns faster than BLSTM and BGRU. Moreover, CNN performs better at extracting sequence patterns than BLSTM and BGRU. To our knowledge, no other framework is developed explicitly for evaluating splice detection models to decide the best possible model in an automated manner. So, the proposed framework and the blueprint would help selecting different deep learning models, such as CNN vs. BLSTM and BGRU, for splice site analysis or similar classification tasks and in different problems.

Exploration of clinical and genetic findings in Ataxia-Telangiectasia (AT) patients from the Indian subcontinent

BackgroundAtaxia-Telangiectasia (AT) is a rare autosomal recessive neurodegenerative disorder. It is caused by mutations in the Ataxia-Telangiectasia mutated (ATM) gene, which codes for protein ATM serine/threonine kinase. ObjectiveWe aim to describe the clinical and radiological findings in children and adolescents of 20 molecularly confirmed cases of AT. We aim to correlate these findings with the genotype identified among them. MethodsThis retrospective study included 20 patients diagnosed clinically and genetically with AT over 10 years. The clinical, radiological and laboratory data were extracted from the hospital's electronic medical records. Molecular testing was done using next generation sequencing and Sanger sequencing. In silico predictions were performed for the variants identified by applying Cryp-Skip, Splice site prediction by Neural Network, Mutation Taster and Hope prediction tool. ResultsConsanguinity was documented in nearly half of the patients. Telangiectasia was absent in 10%. Microcephaly was seen in 40% cases. The incidence of malignancy in our study population was low. Molecular testing done in the 18 families (20 patients) identified 23 variants of which ten were novel. Biallelic homozygous variants were noted in 13 families and compound heterozygous in 5 families. Out of the 13 families who were homozygous, 8 families (61.5%) (9 patients) have history of consanguinity. In silico prediction of novel missense variants, NM_000051.4 (ATM_v201): c.2702T > C showed disruption of the α-helix of ATM protein and NM_000051.4 (ATM_v201): c.6679C > G is expected to disturb the rigidity of protein structure in the FAT domain. The four novel splice site variants and two intronic variants result in exon skipping as predicted by Cryp-Skip. ConclusionsAT should be confirmed by molecular testing in young-onset cerebellar ataxia, even without telangiectasia. Awareness of this rare disease will facilitate study of larger cohorts from Indian population to characterize variants and determine its prevalence in this population.

A deep intronic TCTN2 variant activating a cryptic exon predicted by SpliceRover in a patient with Joubert syndrome.

The recent introduction of genome sequencing in genetic analysis has led to the identification of pathogenic variants located in deep introns. Recently, several new tools have emerged to predict the impact of variants on splicing. Here, we present a Japanese boy of Joubert syndrome with biallelic TCTN2 variants. Exome sequencing identified only a heterozygous maternal nonsense TCTN2 variant (NM_024809.5:c.916C >T, p.(Gln306Ter)). Subsequent genome sequencing identified a deep intronic variant (c.1033+423G>A) inherited from his father. The machine learning algorithms SpliceAI, Squirls, and Pangolin were unable to predict alterations in splicing by the c.1033+423G>A variant. SpliceRover, a tool for splice site prediction using FASTA sequence, was able to detect a cryptic exon which was 85-bp away from the variant and within the inverted Alu sequence while SpliceRover scores for these splice sites showed slight increase (donor) or decrease (acceptor) between the reference and mutant sequences. RNA sequencing and RT-PCR using urinary cells confirmed inclusion of the cryptic exon. The patient showed major symptoms of TCTN2-related disorders such as developmental delay, dysmorphic facial features and polydactyly. He also showed uncommon features such as retinal dystrophy, exotropia, abnormal pattern of respiration, and periventricular heterotopia, confirming these as one of features of TCTN2-related disorders. Our study highlights usefulness of genome sequencing and RNA sequencing using urinary cells for molecular diagnosis of genetic disorders and suggests that database of cryptic splice sites predicted in introns by SpliceRover using the reference sequences can be helpful in extracting candidate variants from large numbers of intronic variants in genome sequencing.

AtLASS: A Scheme for End-to-End Prediction of Splice Sites Using Attention-based Bi-LSTM

AtLASS: A Scheme for End-to-End Prediction of Splice Sites Using Attention-based Bi-LSTM

CNNSplice: Robust models for splice site prediction using convolutional neural networks

The identification of splice site, or segments of an RNA gene where noncoding and coding sequences are connected in the 5′ and 3′ directions, is an essential post-transcriptional step for the annotation of functional genes and is required for the study and analysis of biological function in eukaryotic organisms through protein production and gene expression. Splice site detection tools have been proposed for this purpose; however, the models of these tools have a specific use case and are inefficiently or typically untransferable between organisms. Here, we present CNNSplice, a set of deep convolutional neural network models for splice site prediction. Using the five-fold cross-validation model selection technique, we explore several models based on typical machine learning applications and propose five high-performing models to efficiently predict the true and false SS in balanced and imbalanced datasets. Our evaluation results indicate that CNNSplice’s models achieve a better performance compared with existing methods across five organisms’ datasets. In addition, our generality test shows CNNSplice’s model ability to predict and annotate splice sites in new or poorly trained genome datasets indicating a broad application spectrum. CNNSplice demonstrates improved model prediction, interpretability, and generalizability on genomic datasets compared to existing splice site prediction tools. We have developed a web server for the CNNSplice algorithm which can be publicly accessed here: http://www.cnnsplice.online

Regulation of pre-mRNA splicing: roles in physiology and disease, and therapeutic prospects.

The removal of introns from mRNA precursors and its regulation by alternative splicing are key for eukaryotic gene expression and cellular function, as evidenced by the numerous pathologies induced or modified by splicing alterations. Major recent advances have been made in understanding the structures and functions of the splicing machinery, in the description and classification of physiological and pathological isoforms and in the development of the first therapies for genetic diseases based on modulation of splicing. Here, we review this progress and discuss important remaining challenges, including predicting splice sites from genomic sequences, understanding the variety of molecular mechanisms and logic of splicing regulation, and harnessing this knowledge for probing gene function and disease aetiology and for the design of novel therapeutic approaches.

Comparison of In Silico Tools for Splice-Altering Variant Prediction Using Established Spliceogenic Variants: An End-User's Point of View.

Assessing the impact of variants of unknown significance on splicing has become a critical issue and a bottleneck, especially with the widespread implementation of whole-genome or exome sequencing. Although multiple in silico tools are available, the interpretation and application of these tools are difficult and practical guidelines are still lacking. A streamlined decision-making process can facilitate the downstream RNA analysis in a more efficient manner. Therefore, we evaluated the performance of 8 in silico tools (Splice Site Finder, MaxEntScan, Splice-site prediction by neural network, GeneSplicer, Human Splicing Finder, SpliceAI, Splicing Predictions in Consensus Elements, and SpliceRover) using 114 NF1 spliceogenic variants, experimentally validated at the mRNA level. The change in the predicted score incurred by the variant of the nearest wild-type splice site was analyzed, and for type II, III, and IV splice variants, the change in the prediction score of de novo or cryptic splice site was also analyzed. SpliceAI and SpliceRover, tools based on deep learning, outperformed all other tools, with AUCs of 0.972 and 0.924, respectively. For de novo and cryptic splice sites, SpliceAI outperformed all other tools and showed a sensitivity of 95.7% at an optimal cut-off of 0.02 score change. Our results show that deep learning algorithms, especially those of SpliceAI, are validated at a significantly higher rate than other in silico tools for clinically relevant NF1 variants. This suggests that deep learning algorithms outperform traditional probabilistic approaches and classical machine learning tools in predicting the de novo and cryptic splice sites.

EnsembleSplice: ensemble deep learning model for splice site prediction

BackgroundIdentifying splice site regions is an important step in the genomic DNA sequencing pipelines of biomedical and pharmaceutical research. Within this research purview, efficient and accurate splice site detection is highly desirable, and a variety of computational models have been developed toward this end. Neural network architectures have recently been shown to outperform classical machine learning approaches for the task of splice site prediction. Despite these advances, there is still considerable potential for improvement, especially regarding model prediction accuracy, and error rate.ResultsGiven these deficits, we propose EnsembleSplice, an ensemble learning architecture made up of four (4) distinct convolutional neural networks (CNN) model architecture combination that outperform existing splice site detection methods in the experimental evaluation metrics considered including the accuracies and error rates. We trained and tested a variety of ensembles made up of CNNs and DNNs using the five-fold cross-validation method to identify the model that performed the best across the evaluation and diversity metrics. As a result, we developed our diverse and highly effective splice site (SS) detection model, which we evaluated using two (2) genomic Homo sapiens datasets and the Arabidopsis thaliana dataset. The results showed that for of the Homo sapiens EnsembleSplice achieved accuracies of 94.16% for one of the acceptor splice sites and 95.97% for donor splice sites, with an error rate for the same Homo sapiens dataset, 4.03% for the donor splice sites and 5.84% for the acceptor splice sites datasets.ConclusionsOur five-fold cross validation ensured the prediction accuracy of our models are consistent. For reproducibility, all the datasets used, models generated, and results in our work are publicly available in our GitHub repository here: https://github.com/OluwadareLab/EnsembleSplice

AttentionSplice: An Interpretable Multi‐Head Self‐Attention Based Hybrid Deep Learning Model in Splice Site Prediction

Pre-mRNA splicing is an essential procedure for gene transcription. Through the cutting of introns and exons, the DNA sequence can be decoded into different proteins related to different biological functions. The cutting boundaries are defined by the donor and acceptor splice sites. Characterizing the nucleotides patterns in detecting splice sites is sophisticated and challenges the conventional methods. Recently, the deep learning frame has been introduced in predicting splice sites and exhibits high performance. It extracts high dimension features from the DNA sequence automatically rather than infers the splice sites with prior knowledge of the relationships, dependencies, and characteristics of nucleotides in the DNA sequence. This paper proposes the AttentionSplice model, a hybrid construction combined with multi-head self-attention, convolutional neural network, bidirectional long short-term memory network. The performance of AttentionSplice is evaluated on the Homo sapiens (Human) and Caenorhabditis Elegans (Worm) datasets. Our model outperforms state-of-the-art models in the classification of splice sites. To provide interpretability of AttentionSplice models, we extract important positions and key motifs which could be essential for splice site detection through the attention learned by the model. Our result could offer novel insights into the underlying biological roles and molecular mechanisms of gene expression.

ANOS1 variants in a large cohort of Chinese patients with congenital hypogonadotropic hypogonadism.

Congenital hypogonadotropic hypogonadism (CHH) is a rare congenital gonadal dysplasia caused by defects in the synthesis, secretion or signal transduction of hypothalamic gonadotropin releasing hormone. The main manifestations of CHH are delayed or lack puberty, low levels of sex hormones and gonadotropins, and may be accompanied with other clinical phenotypes. Some patients with CHH are also accompanied with anosmia or hyposmia, which is called Kalman syndrome (KS). ANOS1, located on X chromosome, is the first gene associated with CHH in an X-linked recessive manner. This study aims to provide a basis for the genetic diagnosis of CHH by analyzing the gene variant spectrum of ANOS1 in CHH and the relationship between clinical phenotype and genotype. In this study, whole exome sequencing (WES) was used to screen rare sequencing variants (RSVs) of ANOS1 in a Chinese cohort of 165 male CHH patients. Four commonly used in silico tools were used to predict the function of the identified RSVs in coding region, including Polyphen2, Mutation Taster, SIFT, and Combined Annotation Dependent Depletion (CADD). Splice Site Prediction by Neural Network (NNSPLICE) was employed to predict possibilities of intronic RSVs to disrupt splicing. American College of Medical Genetics and Genomics (ACMG) guidelines was used to assess the pathogenicity of the detected RSVs. The ANOS1 genetic variant spectrum of CHH patients in Chinese population was established. The relationship between clinical phenotype and genotype was analyzed by collecting detailed clinical data. Through WES analysis for 165 CHH patients, ANOS1 RSVs were detected in 17 of them, with the frequency of 10.3%. A total of 13 RSVs were detected in the 17 probands, including 5 nonsense variants (p.T76X, p.R191X, p.W257X, p.R262X, and p.W589X), 2 splicing site variants (c.318+3A>C, c.1063-1G>C), and 6 missense variants (p.N402S, p.N155D, p.P504L, p.C157R, p.Q635P, and p.V560I). In these 17 CHH probands with ANOS1 RSVs, many were accompanied with other clinical phenotypes. The most common associated phenotype was cryptorchidism (10/17), followed by unilateral renal agenesis (3/17), dental agenesis (3/17), and synkinesia (3/17). Eight RSVs, including p.T76X, p.R191X, p.W257X, p.R262X, p.W589X, c.318+3A>C, c.1063-1G>C, and p.C157R, were predicted to be pathogenic or likely pathogenic ANOS1 RSVs by ACMG. Eight CHH patients with pathogenic or likely pathogenic ANOS1 variants had additional features. In contrast, only one out of nine CHH patients with non-pathogenic (likely benign or uncertain of significance) ANOS1 variants according to ACMG exhibited additional features. And function of the non-pathogenic ANOS1 variants accompanied with other CHH-associated RSVs. The ANOS1 genetic spectrum of CHH patients in Chinese population is established. Some of the correlations between clinical phenotype and genotype are also established. Our study indicates that CHH patients with pathogenic or likely pathogenic ANOS1 RSVs tend to exhibit additional phenotypes. Although non-pathogenic ANOS1 variants only may not be sufficient to cause CHH, they may function together with other CHH-associated RSVs to cause the disease.

Polymorphic differences within HLA-C alleles contribute to alternatively spliced transcripts lacking exon 5.

The HLA genes are amongst the most polymorphic in the human genome. Alternative splicing could add an extra layer of complexity, but has not been studied extensively. Here, we applied an RNA based approach to study the influence of allele polymorphism on alternative splicing of HLA‐C in peripheral blood. RNA was isolated from these peripheral cells, converted into cDNA and amplified specifically for 12 common HLA‐C allele groups. Through subsequent sequencing of HLA‐C, we observed alternative splicing variants of HLA‐C*04 and *16 that resulted in exon 5 skipping and were co‐expressed with the mature transcript. Investigation of intron 4 sequences of HLA‐C*04 and *16 compared with other HLA‐C alleles demonstrated no effect on predicted splice sites and branch point. To further investigate if the unique polymorphic positions in exon 5 of HLA‐C*04 or *16 may facilitate alternative splicing by acting on splicing regulatory elements (SRE), in‐silico splicing analysis was performed. While the HLA‐C*04 specific SNP in exon 5 had no effect on predicted exonic SRE, the HLA‐C*16 specific exon 5 SNP did alter exonic SRE. Our findings provide experimental and theoretical support for the concept that polymorphisms within the HLA‐C alleles influence the alternative splicing of HLA‐C.

Deep Splicer: A CNN Model for Splice Site Prediction in Genetic Sequences.

Many living organisms have DNA in their cells that is responsible for their biological features. DNA is an organic molecule of two complementary strands of four different nucleotides wound up in a double helix. These nucleotides are adenine (A), thymine (T), guanine (G), and cytosine (C). Genes are DNA sequences containing the information to synthesize proteins. The genes of higher eukaryotic organisms contain coding sequences, known as exons and non-coding sequences, known as introns, which are removed on splice sites after the DNA is transcribed into RNA. Genome annotation is the process of identifying the location of coding regions and determining their function. This process is fundamental for understanding gene structure; however, it is time-consuming and expensive when done by biochemical methods. With technological advances, splice site detection can be done computationally. Although various software tools have been developed to predict splice sites, they need to improve accuracy and reduce false-positive rates. The main goal of this research was to generate Deep Splicer, a deep learning model to identify splice sites in the genomes of humans and other species. This model has good performance metrics and a lower false-positive rate than the currently existing tools. Deep Splicer achieved an accuracy between 93.55% and 99.66% on the genetic sequences of different organisms, while Splice2Deep, another splice site detection tool, had an accuracy between 90.52% and 98.08%. Splice2Deep surpassed Deep Splicer on the accuracy obtained after evaluating C. elegans genomic sequences (97.88% vs. 93.62%) and A. thaliana (95.40% vs. 94.93%); however, Deep Splicer’s accuracy was better for H. sapiens (98.94% vs. 97.15%) and D. melanogaster (97.14% vs. 92.30%). The rate of false positives was 0.11% for human genetic sequences and 0.25% for other species’ genetic sequences. Another splice prediction tool, Splice Finder, had between 1% and 3% of false positives for human sequences, while other species’ sequences had around 4% and 10%.

A Bidirectional LSTM-RNN and GRU Method to Exon Prediction Using Splice-Site Mapping

Deep Learning techniques (DL) significantly improved the accuracy of predictions and classifications of deoxyribonucleic acid (DNA). On the other hand, identifying and predicting splice sites in eukaryotes is difficult due to many erroneous discoveries. To address this issue, we propose a deep learning model for recognizing and anticipating splice sites in eukaryotic DNA sequences based on a bidirectional Long Short-Term Memory (LSTM) Recurrent Neural Network (RNN) and Gated recurrent unit (GRU). The non-coding introns of the gene are spliced out, and the coding exons are joined during the splicing of the original mRNA transcript. This bidirectional LSTM-RNN-GRU model incorporates intron features in order of their length constraints, beginning with splice site donor (GT) and ending with splice site acceptor (AG). The performance of the model improves as the number of training epochs grows. The best level of accuracy for this model is 96.1 percent.

The Four mRNA Bases Have Quite Different (Un)folding Free Energies, Applications to RNA Splicing and Translation Initiation with BindOligoNet

Expression of mRNA is often regulated by the binding of a small RNA (miRNA, snoRNA, siRNA). While the pairing contribution to the net free energy is well parameterized and can be computed in O(N) time, the cost of removing pre-existing mRNA secondary structure has not received sufficient attention. Conventional methods for computing the unfolding free energy of a target mRNA are costly, scaling like the cube of the number of target bases O(N3). Here we introduce a model to describe the unfolding costs of the binding site, which features surprisingly big differences in the free energy parameters for the four bases. The model is implemented in our O(N) algorithm, BindOligoNet. Donor splice site prediction is more accurate when using our calculation of spliceosomal U1-snRNA to mRNA net binding free energy. Our base-dependent free energies also correlate with efficient ribosome docking near the start codon.