Hyb4mC: a hybrid DNA2vec-based model for DNA N4-methylcytosine sites prediction

iDNA-OpenPrompt: OpenPrompt learning model for identifying DNA methylation

Introduction: DNA methylation is a critical epigenetic modification involving the addition of a methyl group to the DNA molecule, playing a key role in regulating gene expression without changing the DNA sequence. The main difficulty in identifying DNA methylation sites lies in the subtle and complex nature of methylation patterns, which may vary across different tissues, developmental stages, and environmental conditions. Traditional methods for methylation site identification, such as bisulfite sequencing, are typically labor-intensive, costly, and require large amounts of DNA, hindering high-throughput analysis. Moreover, these methods may not always provide the resolution needed to detect methylation at specific sites, especially in genomic regions that are rich in repetitive sequences or have low levels of methylation. Furthermore, current deep learning approaches generally lack sufficient accuracy. Methods: This study introduces the iDNA-OpenPrompt model, leveraging the novel OpenPrompt learning framework. The model combines a prompt template, prompt verbalizer, and Pre-trained Language Model (PLM) to construct the prompt-learning framework for DNA methylation sequences. Moreover, a DNA vocabulary library, BERT tokenizer, and specific label words are also introduced into the model to enable accurate identification of DNA methylation sites. Results and Discussion: An extensive analysis is conducted to evaluate the predictive, reliability, and consistency capabilities of the iDNA-OpenPrompt model. The experimental outcomes, covering 17 benchmark datasets that include various species and three DNA methylation modifications (4mC, 5hmC, 6mA), consistently indicate that our model surpasses outstanding performance and robustness approaches.

SDBA: Score Domain-Based Attention for DNA N4-Methylcytosine Site Prediction from Multiperspectives

In tasks related to DNA sequence classification, choosing the appropriate encoding methods is challenging. Some of the methods encode sequences based on prior knowledge that limits the ability of the model to obtain multiperspective information from the sequences. We introduced a new trainable ensemble method based on the attention mechanism SDBA, which stands for Score Domain-Based Attention. Unlike other methods, we fed the task-independent encoding results into the models and dynamically ensembled features from different perspectives using the SDBA mechanism. This approach allows the model to acquire and weight sequence features voluntarily. SDBA is conceptually general and empirically powerful. It has achieved new state-of-the-art results on the benchmark data sets associated with DNA N4-methylcytosine site prediction.

Time series-based hybrid ensemble learning model with multivariate multidimensional feature coding for DNA methylation prediction

BACKGROUND

DNA methylation is a form of epigenetic modification that impacts gene expression without modifying the DNA sequence, thereby exerting control over gene function and cellular development. The prediction of DNA methylation is vital for understanding and exploring gene regulatory mechanisms. Currently, machine learning algorithms are primarily used for model construction. However, several challenges remain to be addressed, including limited prediction accuracy, constrained generalization capability, and insufficient learning capacity.

RESULTS

In response to the aforementioned challenges, this paper leverages the similarities between DNA sequences and time series to introduce a time series-based hybrid ensemble learning model, called Multi2-Con-CAPSO-LSTM. The model utilizes multivariate and multidimensional encoding approach, combining three types of time series encodings with three kinds of genetic feature encodings, resulting in a total of nine types of feature encoding matrices. Convolutional Neural Networks are utilized to extract features from DNA sequences, including temporal, positional, physicochemical, and genetic information, thereby creating a comprehensive feature matrix. The Long Short-Term Memory model is then optimized using the Chaotic Accelerated Particle Swarm Optimization algorithm for predicting DNA methylation.

CONCLUSIONS

Through cross-validation experiments conducted on 17 species involving three types of DNA methylation (6 mA, 5hmC, and 4mC), the results demonstrate the robust predictive capabilities of the Multi2-Con-CAPSO-LSTM model in DNA methylation prediction across various types and species. Compared with other benchmark models, the Multi2-Con-CAPSO-LSTM model demonstrates significant advantages in sensitivity, specificity, accuracy, and correlation. The model proposed in this paper provides valuable insights and inspiration across various disciplines, including sequence alignment, genetic evolution, time series analysis, and structure-activity relationships.

CRBSP:Prediction of CircRNA-RBP Binding Sites Based on Multimodal Intermediate Fusion

Circular RNA (CircRNA) is widely expressed and has physiological and pathological significance, regulating post-transcriptional processes via its protein-binding activity. However, whereas much work has been done on linear RNA and RNA binding protein (RBP), little is known about the binding sites of CircRNA. The current report is on the development of a medium-term multimodal data fusion strategy, CRBSP, to predict CircRNA-RBP binding sites. CRBSP represents the CircRNA trinucleotide semantic, location, composition and frequency information as the corresponding coding methods of Word to vector (Word2vec), Position-specific trinucleotide propensity (PSTNP), Pseudo trinucleotide composition (PseTNC) and Trinucleotide nucleotide composition (TNC), respectively. CNN (Convolution Neural Networks) was used to extract global information and BiLSTM (bidirectional Long- and Short-Term Memory network) encoder and LSTM (Long- and Short-Term Memory network) decoder for local sequence information. Enhancement of the contributions of key features by the self-attention mechanism was followed by mid-term fusion of the four enhanced features. Logistic Regression (LR) classifier showed that CRBSP gives a mean AUC value of 0.9362 through 5-fold Cross Validation of all 37 datasets, a performance which is superior to five current state-of-the-art models. Similar evaluation of linear RNA-RBP binding sites gave an AUC value of 0.7615 which is also higher than other prediction methods, demonstrating the robustness of CRBSP.

GR-m6A: Prediction of N6-methyladenosine sites in mammals with molecular graph and residual network

RNA N6-methyladenine (m6A), which is produced by the methylation of the N6 position of eukaryotic adenine, is a relatively common post-transcriptional modification on the surface of the molecule, which frequently plays a crucial role in biological processes. Biological experimental methods to identify m6A have been studied and implemented in recent years, but they cannot be promoted widely due to drawbacks such as the time and cost of reagents and equipment. Therefore, researchers have proposed computational strategies for identifying m6A sites, but these strategies do not account for the mechanism of methylation occurrence or the structure of RNA molecules. This study, therefore, proposed a novel deep learning model for predicting m6A sites, GR-m6A, which predicts m6A sites by extracting features from the physicochemical properties and spatial structure of molecules via residual networks. In GR-m6A, each RNA base string is represented by SMILES as two matrices comprising topology structural information and node attributes with molecular physicochemical characteristics. The feature encoding matrix was then obtained by fusing the topology matrix and the node matrix in accordance with the graphical convolutional network principle. Correspondingly, the more discriminative features were extracted from the encoding matrix using the residual neural network and predicted using a multilayer perceptron. As evident from the 5-fold cross-validation and independent validation, the GR-m6A model outperformed other existing methods. Thus, we hope that GR-m6A can aid researchers in predicting mammalian m6A loci. The source code and database are available at https://github.com/YingLiangjxau/GR-m6A.

Prediction of DNA Methylation based on Multi-dimensional feature encoding and double convolutional fully connected convolutional neural network

DNA methylation takes on critical significance to the regulation of gene expression by affecting the stability of DNA and changing the structure of chromosomes. DNA methylation modification sites should be identified, which lays a solid basis for gaining more insights into their biological functions. Existing machine learning-based methods of predicting DNA methylation have not fully exploited the hidden multidimensional information in DNA gene sequences, such that the prediction accuracy of models is significantly limited. Besides, most models have been built in terms of a single methylation type. To address the above-mentioned issues, a deep learning-based method was proposed in this study for DNA methylation site prediction, termed the MEDCNN model. The MEDCNN model is capable of extracting feature information from gene sequences in three dimensions (i.e., positional information, biological information, and chemical information). Moreover, the proposed method employs a convolutional neural network model with double convolutional layers and double fully connected layers while iteratively updating the gradient descent algorithm using the cross-entropy loss function to increase the prediction accuracy of the model. Besides, the MEDCNN model can predict different types of DNA methylation sites. As indicated by the experimental results,the deep learning method based on coding from multiple dimensions outperformed single coding methods, and the MEDCNN model was highly applicable and outperformed existing models in predicting DNA methylation between different species. As revealed by the above-described findings, the MEDCNN model can be effective in predicting DNA methylation sites.

Identifying potential drug-target interactions based on ensemble deep learning

Introduction

Drug-target interaction prediction is one important step in drug research and development. Experimental methods are time consuming and laborious.

Methods

In this study, we developed a novel DTI prediction method called EnGDD by combining initial feature acquisition, dimensional reduction, and DTI classification based on Gradient boosting neural network, Deep neural network, and Deep Forest.

Results

EnGDD was compared with seven stat-of-the-art DTI prediction methods (BLM-NII, NRLMF, WNNGIP, NEDTP, DTi2Vec, RoFDT, and MolTrans) on the nuclear receptor, GPCR, ion channel, and enzyme datasets under cross validations on drugs, targets, and drug-target pairs, respectively. EnGDD computed the best recall, accuracy, F1-score, AUC, and AUPR under the majority of conditions, demonstrating its powerful DTI identification performance. EnGDD predicted that D00182 and hsa2099, D07871 and hsa1813, DB00599 and hsa2562, D00002 and hsa10935 have a higher interaction probabilities among unknown drug-target pairs and may be potential DTIs on the four datasets, respectively. In particular, D00002 (Nadide) was identified to interact with hsa10935 (Mitochondrial peroxiredoxin3) whose up-regulation might be used to treat neurodegenerative diseases. Finally, EnGDD was used to find possible drug targets for Parkinson's disease and Alzheimer's disease after confirming its DTI identification performance. The results show that D01277, D04641, and D08969 may be applied to the treatment of Parkinson's disease through targeting hsa1813 (dopamine receptor D2) and D02173, D02558, and D03822 may be the clues of treatment for patients with Alzheimer's disease through targeting hsa5743 (prostaglandinendoperoxide synthase 2). The above prediction results need further biomedical validation.

Discussion

We anticipate that our proposed EnGDD model can help discover potential therapeutic clues for various diseases including neurodegenerative diseases.

EpiTEAmDNA: Sequence feature representation via transfer learning and ensemble learning for identifying multiple DNA epigenetic modification types across species

Methylation is a major DNA epigenetic modification for regulating the biological processes without altering the DNA sequence, and multiple types of DNA methylations have been discovered, including 6mA, 5hmC, and 4mC. Multiple computational approaches were developed to automatically identify the DNA methylation residues using machine learning or deep learning algorithms. The machine learning (ML) based methods are difficult to be transferred to the other predicting tasks of the DNA methylation sites using additional knowledge. Deep learning (DL) may facilitate the transfer learning of knowledge from similar tasks, but they are often ineffective on small datasets. This study proposes an integrated feature representation framework EpiTEAmDNA based on the strategies of transfer learning and ensemble learning, which is evaluated on multiple DNA methylation types across 15 species. EpiTEAmDNA integrates convolutional neural network (CNN) and conventional machine learning methods, and shows improved performances than the existing DL-based methods on small datasets when no additional knowledge is available. The experimental data suggests that the EpiTEAmDNA models may be further improved via transfer learning based on additional knowledge. The evaluation experiments on the independent test datasets also suggest that the proposed EpiTEAmDNA framework outperforms the existing models in most prediction tasks of the 3 DNA methylation types across 15 species. The source code, pre-trained global model, and the EpiTEAmDNA feature representation framework are freely available at http://www.healthinformaticslab.org/supp/.

DRSN4mCPred: accurately predicting sites of DNA N4-methylcytosine using deep residual shrinkage network for diagnosis and treatment of gastrointestinal cancer in the precision medicine era

Introduction

The DNA N4-methylcytosine (4mC) site levels of those suffering from digestive system cancers were higher, and the pathogenesis of digestive system cancers may also be related to the changes in DNA 4mC levels. Identifying DNA 4mC sites is a very important step in studying the analysis of biological function and cancer prediction. Extracting accurate features from DNA sequences is the key to establishing a prediction model of effective DNA 4mC sites. This study sought to develop a new predictive model, DRSN4mCPred, which aimed to improve the performance of the predicting DNA 4mC sites.

Methods

The model adopted multi-scale channel attention to extract features and used attention feature fusion (AFF) to fuse features. In order to capture features information more accurately and effectively, this model utilized Deep Residual Shrinkage Network with Channel-Wise thresholds (DRSN-CW) to eliminate noise-related features and achieve a more precise feature representation, thereby, distinguishing the sites in DNA with 4mC and non-4mC. Additionally, the predictive model incorporated an inverted residual block, a Multi-scale Channel Attention Module (MS-CAM), a Bi-directional Long Short Term Memory Network (Bi-LSTM), AFF, and DRSN-CW.

Results and Discussion

The results indicated the predictive model DRSN4mCPred had extremely good performance in predicting the DNA 4mC sites across different species. This paper will potentially provide support for the diagnosis and treatment of gastrointestinal cancer based on artificial intelligence in the precise medical era.

4mCBERT: A computing tool for the identification of DNA N4-methylcytosine sites by sequence- and chemical-derived information based on ensemble learning strategies

N4-methylcytosine (4mC) is an important DNA chemical modification pattern which is a new methylation modification discovered in recent years and plays critical roles in gene expression regulation, defense against invading genetic elements, genomic imprinting, and so on. Identifying 4mC site from DNA sequence segment contributes to discovering more novel modification patterns. In this paper, we present a model called 4mCBERT that encodes DNA sequence segments by sequence characteristics including one-hot, electron-ion interaction pseudopotential, nucleotide chemical property, word2vec and chemical information containing physicochemical properties (PCP), chemical bidirectional encoder representations from transformers (chemical BERT) and employs ensemble learning framework to develop a prediction model. PCP and chemical BERT features are firstly constructed and applied to predict 4mC sites and show positive contributions to identifying 4mC. For the Matthew's Correlation Coefficient, 4mCBERT significantly outperformed other state-of-the-art models on six independent benchmark datasets including A. thaliana, C. elegans, D. melanogaster, E. coli, G. Pickering, and G. subterraneous by 4.32 % to 24.39 %, 2.52 % to 31.65 %, 2 % to 16.49 %, 6.63 % to 35.15, 8.59 % to 61.85 %, and 8.45 % to 34.45 %. Moreover, 4mCBERT is designed to allow users to predict 4mC sites and retrain 4mC prediction models. In brief, 4mCBERT shows higher performance on six benchmark datasets by incorporating sequence- and chemical-driven information and is available at http://cczubio.top/4mCBERT and https://github.com/abcair/4mCBERT.

Screening potential lncRNA biomarkers for breast cancer and colorectal cancer combining random walk and logistic matrix factorization

Breast cancer and colorectal cancer are two of the most common malignant tumors worldwide. They cause the leading causes of cancer mortality. Many researches have demonstrated that long noncoding RNAs (lncRNAs) have close linkages with the occurrence and development of the two cancers. Therefore, it is essential to design an effective way to identify potential lncRNA biomarkers for them. In this study, we developed a computational method (LDA-RWLMF) by integrating random walk with restart and Logistic Matrix Factorization to investigate the roles of lncRNA biomarkers in the prognosis and diagnosis of the two cancers. We first fuse disease semantic and Gaussian association profile similarities and lncRNA functional and Gaussian association profile similarities. Second, we design a negative selection algorithm to extract negative LncRNA-Disease Associations (LDA) based on random walk. Third, we develop a logistic matrix factorization model to predict possible LDAs. We compare our proposed LDA-RWLMF method with four classical LDA prediction methods, that is, LNCSIM1, LNCSIM2, ILNCSIM, and IDSSIM. The results from 5-fold cross validation on the MNDR dataset show that LDA-RWLMF computes the best AUC value of 0.9312, outperforming the above four LDA prediction methods. Finally, we rank all lncRNA biomarkers for the two cancers after determining the performance of LDA-RWLMF, respectively. We find that 48 and 50 lncRNAs have the highest association scores with breast cancer and colorectal cancer among all lncRNAs known to associate with them on the MNDR dataset, respectively. We predict that lncRNAs HULC and HAR1A could be separately potential biomarkers for breast cancer and colorectal cancer and need to biomedical experimental validation.

CapsNet-LDA: predicting lncRNA-disease associations using attention mechanism and capsule network based on multi-view data

Cumulative studies have shown that many long non-coding RNAs (lncRNAs) are crucial in a number of diseases. Predicting potential lncRNA-disease associations (LDAs) can facilitate disease prevention, diagnosis and treatment. Therefore, it is vital to develop practical computational methods for LDA prediction. In this study, we propose a novel predictor named capsule network (CapsNet)-LDA for LDA prediction. CapsNet-LDA first uses a stacked autoencoder for acquiring the informative low-dimensional representations of the lncRNA-disease pairs under multiple views, then the attention mechanism is leveraged to implement an adaptive allocation of importance weights to them, and they are subsequently processed using a CapsNet-based architecture for predicting LDAs. Different from the conventional convolutional neural networks (CNNs) that have some restrictions with the usage of scalar neurons and pooling operations. the CapsNets use vector neurons instead of scalar neurons that have better robustness for the complex combination of features and they use dynamic routing processes for updating parameters. CapsNet-LDA is superior to other five state-of-the-art models on four benchmark datasets, four perturbed datasets and an independent test set in the comparison experiments, demonstrating that CapsNet-LDA has excellent performance and robustness against perturbation, as well as good generalization ability. The ablation studies verify the effectiveness of some modules of CapsNet-LDA. Moreover, the ability of multi-view data to improve performance is proven. Case studies further indicate that CapsNet-LDA can accurately predict novel LDAs for specific diseases.

MuLan-Methyl-multiple transformer-based language models for accurate DNA methylation prediction

Transformer-based language models are successfully used to address massive text-related tasks. DNA methylation is an important epigenetic mechanism, and its analysis provides valuable insights into gene regulation and biomarker identification. Several deep learning-based methods have been proposed to identify DNA methylation, and each seeks to strike a balance between computational effort and accuracy. Here, we introduce MuLan-Methyl, a deep learning framework for predicting DNA methylation sites, which is based on 5 popular transformer-based language models. The framework identifies methylation sites for 3 different types of DNA methylation: N6-adenine, N4-cytosine, and 5-hydroxymethylcytosine. Each of the employed language models is adapted to the task using the "pretrain and fine-tune" paradigm. Pretraining is performed on a custom corpus of DNA fragments and taxonomy lineages using self-supervised learning. Fine-tuning aims at predicting the DNA methylation status of each type. The 5 models are used to collectively predict the DNA methylation status. We report excellent performance of MuLan-Methyl on a benchmark dataset. Moreover, we argue that the model captures characteristic differences between different species that are relevant for methylation. This work demonstrates that language models can be successfully adapted to applications in biological sequence analysis and that joint utilization of different language models improves model performance. Mulan-Methyl is open source, and we provide a web server that implements the approach.

Computational prediction and interpretation of druggable proteins using a stacked ensemble-learning framework

Discovery of potential drugs requires rapid and precise identification of drug targets. Although traditional experimental methodologies can accurately identify drug targets, they are time-consuming and inappropriate for high-throughput screening. Computational approaches based on machine learning (ML) algorithms can expedite the prediction of druggable proteins; however, the performance of the existing computational methods remains unsatisfactory. This study proposes a computational tool, SPIDER, to enhance the accurate prediction of druggable proteins. SPIDER employs various feature descriptors pertaining to several aspects, including physicochemical properties, compositional information, and composition-transition-distribution information, coupled with well-known ML algorithms to facilitate the construction of the final meta-predictor. The experimental results showed that SPIDER enabled more precise and robust prediction of druggable proteins than the baseline models and current existing methods in terms of the independent test dataset. An online web server was established and made freely available online.

•

Computational models can expedite the identification of potential druggable proteins

•

SPIDER represents the first stacked model proposed for druggable protein prediction

•

SPIDER enables more precise prediction of druggable proteins than existing methods

•

The SPIDER web server is available at http://pmlabstack.pythonanywhere.com/SPIDER.