DNA sequences performs as natural language processing by exploiting deep learning algorithm for the identification of N4-methylcytosine

DeepSplice: a deep learning approach for accurate prediction of alternative splicing events in the human genome

Alternative splicing (AS) is a crucial process in genetic information processing that generates multiple mRNA molecules from a single gene, producing diverse proteins. Accurate prediction of AS events is essential for understanding various physiological aspects, including disease progression and prognosis. Machine learning (ML) techniques have been widely employed in bioinformatics to address this challenge. However, existing models have limitations in capturing AS events in the presence of mutations and achieving high prediction performance. To overcome these limitations, this research presents deep splicing code (DSC), a deep learning (DL)-based model for AS prediction. The proposed model aims to improve predictive ability by investigating state-of-the-art techniques in AS and developing a DL model specifically designed to predict AS events accurately. The performance of the DSC model is evaluated against existing techniques, revealing its potential to enhance the understanding and predictive power of DL algorithms in AS. It outperforms other models by achieving an average AUC score of 92%. The significance of this research lies in its contribution to identifying functional implications and potential therapeutic targets associated with AS, with applications in genomics, bioinformatics, and biomedical research. The findings of this study have the potential to advance the field and pave the way for more precise and reliable predictions of AS events, ultimately leading to a deeper understanding of genetic information processing and its impact on human physiology and disease.

Identification of individuals using functional near-infrared spectroscopy based on a one-dimensional convolutional neural network

In recent years, the application of functional near-infrared spectroscopy (fNIRS) and deep learning techniques has emerged as a promising method for personal identification. In this study, we innovatively utilized a deep learning framework and fNIRS data for personal identification. The framework is a one-dimensional convolutional neural network (Conv1D) trained on resting-state fNIRS signals collected from the frontal cortex of adults. In data preprocessing, we employed a sliding window-based data augmentation technique and high-pass filter, which could result in the highest identification accuracy through multiple experiments. Based on a data set consisting of 56 adult participants, the identification accuracy of 90.36% is achieved for training data with a window size of approximately 4.62 s; with the increase in training data window size, the identification accuracy can reach (97.65 ± 2.35)%. Our results suggest that deep learning is valuable for fNIRS-based personal identification, with potential applications in security, biometrics, and healthcare.

Discovery of a non-canonical GRHL1 binding site using deep convolutional and recurrent neural networks

BACKGROUND

Transcription factors regulate gene expression by binding to transcription factor binding sites (TFBSs). Most models for predicting TFBSs are based on position weight matrices (PWMs), which require a specific motif to be present in the DNA sequence and do not consider interdependencies of nucleotides. Novel approaches such as Transcription Factor Flexible Models or recurrent neural networks consequently provide higher accuracies. However, it is unclear whether such approaches can uncover novel non-canonical, hitherto unexpected TFBSs relevant to human transcriptional regulation.

RESULTS

In this study, we trained a convolutional recurrent neural network with HT-SELEX data for GRHL1 binding and applied it to a set of GRHL1 binding sites obtained from ChIP-Seq experiments from human cells. We identified 46 non-canonical GRHL1 binding sites, which were not found by a conventional PWM approach. Unexpectedly, some of the newly predicted binding sequences lacked the CNNG core motif, so far considered obligatory for GRHL1 binding. Using isothermal titration calorimetry, we experimentally confirmed binding between the GRHL1-DNA binding domain and predicted GRHL1 binding sites, including a non-canonical GRHL1 binding site. Mutagenesis of individual nucleotides revealed a correlation between predicted binding strength and experimentally validated binding affinity across representative sequences. This correlation was neither observed with a PWM-based nor another deep learning approach.

CONCLUSIONS

Our results show that convolutional recurrent neural networks may uncover unanticipated binding sites and facilitate quantitative transcription factor binding predictions.

4mC-CGRU: Identification of N4-Methylcytosine (4mC) sites using convolution gated recurrent unit in Rosaceae genome

An epigenetic modification is DNA N4-methylcytosine (4mC) that affects several biological functions without altering the DNA nucleotides, including DNA conformation, cell development, replication, stability, and DNA structural changes. To prevent restriction enzyme from damaging self-DNA, 4mC performs a critical role in restriction-modification functions. Existing studies mainly focused on finding hand-crafted features to identify 4mC locations, but these methods are inefficient due to high time consuming and high costs. In our research work, we propose a 4mC-CGRU which is a deep learning-based computational model with a standard encoding method to identify the 4mC sites from DNA sequences that learned autonomous feature selection in the Rosaceae genome, particularly in Rosa chinensis (R. chinensis) and Fragaria vesca (F. vesca). The proposed model consists of a convolutional neural network (CNN) and a gated recurrent unit network (GRU)-based model for identifying 4mC sites from Fragaria vesca and Rosa chinensis in the genomes. The CNN model extracts useful features from the datasets and the GRU classifies the DNA sequences. Thus, our approach can automatically extract important features to detect relative sites from DNA sequence. The performance analysis shows that the proposed model consistently outperforms over the state-of-the-art works in detecting 4mC sites.

DNABERT-based explainable lncRNA identification in plant genome assemblies

Long non-coding ribonucleic acids (lncRNAs) have been shown to play an important role in plant gene regulation, involving both epigenetic and transcript regulation. LncRNAs are transcripts longer than 200 nucleotides that are not translated into functional proteins but can be translated into small peptides. Machine learning models have predominantly used transcriptome data with manually defined features to detect lncRNAs, however, they often underrepresent the abundance of lncRNAs and can be biased in their detection. Here we present a study using Natural Language Processing (NLP) models to identify plant lncRNAs from genomic sequences rather than transcriptomic data. The NLP models were trained to predict lncRNAs for seven model and crop species (Zea mays, Arabidopsis thaliana, Brassica napus, Brassica oleracea, Brassica rapa, Glycine max and Oryza sativa) using publicly available genomic references. We demonstrated that lncRNAs can be accurately predicted from genomic sequences with the highest accuracy of 83.4% for Z. mays and the lowest accuracy of 57.9% for B. rapa, revealing that genome assembly quality might affect the accuracy of lncRNA identification. Furthermore, we demonstrated the potential of using NLP models for cross-species prediction with an average of 63.1% accuracy using target species not previously seen by the model. As more species are incorporated into the training datasets, we expect the accuracy to increase, becoming a more reliable tool for uncovering novel lncRNAs. Finally, we show that the models can be interpreted using explainable artificial intelligence to identify motifs important to lncRNA prediction and that these motifs frequently flanked the lncRNA sequence.

A Grid Search-Based Multilayer Dynamic Ensemble System to Identify DNA N4—Methylcytosine Using Deep Learning Approach

DNA (Deoxyribonucleic Acid) N4-methylcytosine (4mC), a kind of epigenetic modification of DNA, is important for modifying gene functions, such as protein interactions, conformation, and stability in DNA, as well as for the control of gene expression throughout cell development and genomic imprinting. This simply plays a crucial role in the restriction–modification system. To further understand the function and regulation mechanism of 4mC, it is essential to precisely locate the 4mC site and detect its chromosomal distribution. This research aims to design an efficient and high-throughput discriminative intelligent computational system using the natural language processing method “word2vec” and a multi-configured 1D convolution neural network (1D CNN) to predict 4mC sites. In this article, we propose a grid search-based multi-layer dynamic ensemble system (GS-MLDS) that can enhance existing knowledge of each level. Each layer uses a grid search-based weight searching approach to find the optimal accuracy while minimizing computation time and additional layers. We have used eight publicly available benchmark datasets collected from different sources to test the proposed model’s efficiency. Accuracy results in test operations were obtained as follows: 0.978, 0.954, 0.944, 0.961, 0.950, 0.973, 0.948, 0.952, 0.961, and 0.980. The proposed model has also been compared to 16 distinct models, indicating that it can accurately predict 4mC.

DNA-MP: a generalized DNA modifications predictor for multiple species based on powerful sequence encoding method

Accurate prediction of deoxyribonucleic acid (DNA) modifications is essential to explore and discern the process of cell differentiation, gene expression and epigenetic regulation. Several computational approaches have been proposed for particular type-specific DNA modification prediction. Two recent generalized computational predictors are capable of detecting three different types of DNA modifications; however, type-specific and generalized modifications predictors produce limited performance across multiple species mainly due to the use of ineffective sequence encoding methods. The paper in hand presents a generalized computational approach "DNA-MP" that is competent to more precisely predict three different DNA modifications across multiple species. Proposed DNA-MP approach makes use of a powerful encoding method "position specific nucleotides occurrence based 117 on modification and non-modification class densities normalized difference" (POCD-ND) to generate the statistical representations of DNA sequences and a deep forest classifier for modifications prediction. POCD-ND encoder generates statistical representations by extracting position specific distributional information of nucleotides in the DNA sequences. We perform a comprehensive intrinsic and extrinsic evaluation of the proposed encoder and compare its performance with 32 most widely used encoding methods on $17$ benchmark DNA modifications prediction datasets of $12$ different species using $10$ different machine learning classifiers. Overall, with all classifiers, the proposed POCD-ND encoder outperforms existing $32$ different encoders. Furthermore, combinedly over 5-fold cross validation benchmark datasets and independent test sets, proposed DNA-MP predictor outperforms state-of-the-art type-specific and generalized modifications predictors by an average accuracy of 7% across 4mc datasets, 1.35% across 5hmc datasets and 10% for 6ma datasets. To facilitate the scientific community, the DNA-MP web application is available at https://sds_genetic_analysis.opendfki.de/DNA_Modifications/.

A novel circRNA-miRNA association prediction model based on structural deep neural network embedding

A large amount of clinical evidence began to mount, showing that circular ribonucleic acids (RNAs; circRNAs) perform a very important function in complex diseases by participating in transcription and translation regulation of microRNA (miRNA) target genes. However, with strict high-throughput techniques based on traditional biological experiments and the conditions and environment, the association between circRNA and miRNA can be discovered to be labor-intensive, expensive, time-consuming, and inefficient. In this paper, we proposed a novel computational model based on Word2vec, Structural Deep Network Embedding (SDNE), Convolutional Neural Network and Deep Neural Network, which predicts the potential circRNA-miRNA associations, called Word2vec, SDNE, Convolutional Neural Network and Deep Neural Network (WSCD). Specifically, the WSCD model extracts attribute feature and behaviour feature by word embedding and graph embedding algorithm, respectively, and ultimately feed them into a feature fusion model constructed by combining Convolutional Neural Network and Deep Neural Network to deduce potential circRNA-miRNA interactions. The proposed method is proved on dataset and obtained a prediction accuracy and an area under the receiver operating characteristic curve of 81.61% and 0.8898, respectively, which is shown to have much higher accuracy than the state-of-the-art models and classifier models in prediction. In addition, 23 miRNA-related circular RNAs (circRNAs) from the top 30 were confirmed in relevant experiences. In these works, all results represent that WSCD would be a helpful supplementary reliable method for predicting potential miRNA-circRNA associations compared to wet laboratory experiments.

Rich learning representations for human activity recognition: How to empower deep feature learning for biological time series

Deep learning versus feature engineering has drawn significant attention specifically for applications where expertly crafted features have been used for decades. Human activity recognition is no exception where statistical and motion specific features have shown potential in detecting falls and other daily activities across a wide range of datasets. This paper provides an in-depth study and comparison of two fundamentally different approaches to HAR while introducing a novel way to harness the spectral properties of biological time series in addition to temporal features. A research group at Stanford recently proposed subject agnostic features as state-of-the-art when applied to a large dataset with many participants of different ages. In this paper, we demonstrate that implicit feature learning in the latent spaces of deep learning algorithms can be powerful alternatives to using finely tuned domain-specific features for HAR. In fact, when using a spectrotemporal representation of the raw sensor data in the form of spectrograms, a standard convolutional neural network without any prior conditioning on the features, statistically significantly outperforms the prior state-of-the-art using subject agnostic features in all the different partitions of the dataset with a significant  29.8% reduction in the overall average error rate.

BERT-PPII: The Polyproline Type II Helix Structure Prediction Model Based on BERT and Multichannel CNN

Predicting the polyproline type II (PPII) helix structure is crucial important in many research areas, such as the protein folding mechanisms, the drug targets, and the protein functions. However, many existing PPII helix prediction algorithms encode the protein sequence information in a single way, which causes the insufficient learning of protein sequence feature information. To improve the protein sequence encoding performance, this paper proposes a BERT-based PPII helix structure prediction algorithm (BERT-PPII), which learns the protein sequence information based on the BERT model. The BERT model's CLS vector can fairly fuse sample's each amino acid residue information. Thus, we utilize the CLS vector as the global feature to represent the sample's global contextual information. As the interactions among the protein chains' local amino acid residues have an important influence on the formation of PPII helix, we utilize the CNN to extract local amino acid residues' features which can further enhance the information expression of protein sequence samples. In this paper, we fuse the CLS vectors with CNN local features to improve the performance of predicting PPII structure. Compared to the state-of-the-art PPIIPRED method, the experimental results on the unbalanced dataset show that the proposed method improves the accuracy value by 1% on the strict dataset and 2% on the less strict dataset. Correspondingly, the results on the balanced dataset show that the AUCs of the proposed method are 0.826 on the strict dataset and 0.785 on less strict datasets, respectively. For the independent test set, the proposed method has the AUC value of 0.827 on the strict dataset and 0.783 on the less strict dataset. The above experimental results have proved that the proposed BERT-PPII method can achieve a superior performance of predicting the PPII helix.

Systematic Analysis and Accurate Identification of DNA N4-Methylcytosine Sites by Deep Learning

DNA N⁴-methylcytosine (4mC) is a pivotal epigenetic modification that plays an essential role in DNA replication, repair, expression and differentiation. To gain insight into the biological functions of 4mC, it is critical to identify their modification sites in the genomics. Recently, deep learning has become increasingly popular in recent years and frequently employed for the 4mC site identification. However, a systematic analysis of how to build predictive models using deep learning techniques is still lacking. In this work, we first summarized all existing deep learning-based predictors and systematically analyzed their models, features and datasets, etc. Then, using a typical standard dataset with three species (A. thaliana, C. elegans, and D. melanogaster), we assessed the contribution of different model architectures, encoding methods and the attention mechanism in establishing a deep learning-based model for the 4mC site prediction. After a series of optimizations, convolutional-recurrent neural network architecture using the one-hot encoding and attention mechanism achieved the best overall prediction performance. Extensive comparison experiments were conducted based on the same dataset. This work will be helpful for researchers who would like to build the 4mC prediction models using deep learning in the future.

Online Diagnosis and Classification of CT Images Collected by Internet of Things Using Deep Learning

Deep learning technology has recently played an important role in image, language processing, and feature extraction. In the past disease diagnosis, most medical staff fixed the images together for observation and then combined with their own work experience to judge. The diagnosis results are subjective, time-consuming, and inefficient. In order to improve the efficiency of diagnosis, this paper applies the deep learning algorithm to the online diagnosis and classification of CT images. Based on this, in this paper, the deep learning algorithm is applied to CT image online diagnosis and classification. Based on a brief analysis of the current situation of CT image classification, this paper proposes to use the Internet of things technology to collect CT image information and establishes the Internet of things to collect the CT image model. In view of image classification and diagnosis, the convolution neural network algorithm in the deep learning algorithm is proposed to diagnose and classify CT images, and several factors affecting the accuracy of classification are proposed, including the convolution number and network layer number. Using the CT image of the hospital brain for simulation analysis, the simulation results confirm the effectiveness of the deep learning algorithm. With the increase of convolution and network layer and the decrease of compensation, the accuracy of image classification will decline. Using the maximum pool method, reducing the step size can improve the classification effect. Using relu function as the activation function can improve the classification accuracy. In the process of large data set processing, appropriately adding a network layer can improve classification accuracy. In the diagnosis and analysis of brain CT images, the overall classification accuracy is close to 70%, and in the diagnosis of tumor diseases, the accuracy is higher, up to 80%.

BERT-m7G: A Transformer Architecture Based on BERT and Stacking Ensemble to Identify RNA N7-Methylguanosine Sites from Sequence Information

As one of the most prevalent posttranscriptional modifications of RNA, N7-methylguanosine (m7G) plays an essential role in the regulation of gene expression. Accurate identification of m7G sites in the transcriptome is invaluable for better revealing their potential functional mechanisms. Although high-throughput experimental methods can locate m7G sites precisely, they are overpriced and time-consuming. Hence, it is imperative to design an efficient computational method that can accurately identify the m7G sites. In this study, we propose a novel method via incorporating BERT-based multilingual model in bioinformatics to represent the information of RNA sequences. Firstly, we treat RNA sequences as natural sentences and then employ bidirectional encoder representations from transformers (BERT) model to transform them into fixed-length numerical matrices. Secondly, a feature selection scheme based on the elastic net method is constructed to eliminate redundant features and retain important features. Finally, the selected feature subset is input into a stacking ensemble classifier to predict m7G sites, and the hyperparameters of the classifier are tuned with tree-structured Parzen estimator (TPE) approach. By 10-fold cross-validation, the performance of BERT-m7G is measured with an ACC of 95.48% and an MCC of 0.9100. The experimental results indicate that the proposed method significantly outperforms state-of-the-art prediction methods in the identification of m7G modifications.

Construction and Validation of a Lung Cancer Diagnostic Model Based on 6-Gene Methylation Frequency in Blood, Clinical Features, and Serum Tumor Markers

Lung cancer has a high mortality rate. Promoting early diagnosis and screening of lung cancer is the most effective way to enhance the survival rate of lung cancer patients. Through computer technology, a comprehensive evaluation of genetic testing results and basic clinical information of lung cancer patients could effectively diagnose early lung cancer and indicate cancer risks. This study retrospectively collected 70 pairs of lung cancer tissue samples and normal human tissue samples. The methylation frequencies of 6 genes (FHIT, p16, MGMT, RASSF1A, APC, DAPK) in lung cancer patients, the basic clinical information, and tumor marker levels of these patients were analyzed. Then, the python package "sklearn" was employed to build a support vector machine (SVM) classifier which performed 10-fold cross-validation to construct diagnostic models that could identify lung cancer risk of suspected cases. Receiver operation characteristic (ROC) curves were drawn, and the performance of the combined diagnostic model based on several factors (clinical information, tumor marker level, and methylation frequency of 6 genes in blood) was shown to be better than that of models with only one pathological feature. The AUC value of the combined model was 0.963, and the sensitivity, specificity, and accuracy were 0.900, 0.971, and 0.936, respectively. The above results revealed that the diagnostic model based on these features was highly reliable, which could screen and diagnose suspected early lung cancer patients, contributing to increasing diagnosis rate and survival rate of lung cancer patients.

Representation learning applications in biological sequence analysis

Although remarkable advances have been reported in high-throughput sequencing, the ability to aptly analyze a substantial amount of rapidly generated biological (DNA/RNA/protein) sequencing data remains a critical hurdle. To tackle this issue, the application of natural language processing (NLP) to biological sequence analysis has received increased attention. In this method, biological sequences are regarded as sentences while the single nucleic acids/amino acids or k-mers in these sequences represent the words. Embedding is an essential step in NLP, which performs the conversion of these words into vectors. Specifically, representation learning is an approach used for this transformation process, which can be applied to biological sequences. Vectorized biological sequences can then be applied for function and structure estimation, or as input for other probabilistic models. Considering the importance and growing trend for the application of representation learning to biological research, in the present study, we have reviewed the existing knowledge in representation learning for biological sequence analysis.