Adapt-Kcr: a novel deep learning framework for accurate prediction of lysine crotonylation sites based on learning embedding features and attention architecture

N-GlycoPred: A hybrid deep learning model for accurate identification of N-glycosylation sites

Studies have shown that protein glycosylation in cells reflects the real-time dynamics of biological processes, and the occurrence and development of many diseases are closely related to protein glycosylation. Abnormal protein glycosylation can be used as a potential diagnostic and prognostic marker of a disease, as well as a therapeutic target and a new breakthrough point for exploring pathogenesis. To address the issue of significant differences in the prediction results of previous models for different species, we constructed a hybrid deep learning model N-GlycoPred on the basis of dual-layer convolution, a paired attention mechanism and BiLSTM for accurate identification of N-glycosylation sites. By adopting one-hot encoding or the AAindex, we specifically selected the optimum combination of features and deep learning frameworks for human and mouse to refine the models. Based on six independent test datasets, our N-GlycoPred model achieved an average AUC of 0.9553, which is 0.23% higher than MusiteDeep. The comparison results indicate that our model can serve as a powerful tool for N-glycosylation site prescreening for biological researchers.

LMCrot: an enhanced protein crotonylation site predictor by leveraging an interpretable window-level embedding from a transformer-based protein language model

MOTIVATION

Recent advancements in natural language processing have highlighted the effectiveness of global contextualized representations from protein language models (pLMs) in numerous downstream tasks. Nonetheless, strategies to encode the site-of-interest leveraging pLMs for per-residue prediction tasks, such as crotonylation (Kcr) prediction, remain largely uncharted.

RESULTS

Herein, we adopt a range of approaches for utilizing pLMs by experimenting with different input sequence types (full-length protein sequence versus window sequence), assessing the implications of utilizing per-residue embedding of the site-of-interest as well as embeddings of window residues centered around it. Building upon these insights, we developed a novel residual ConvBiLSTM network designed to process window-level embeddings of the site-of-interest generated by the ProtT5-XL-UniRef50 pLM using full-length sequences as input. This model, termed T5ResConvBiLSTM, surpasses existing state-of-the-art Kcr predictors in performance across three diverse datasets. To validate our approach of utilizing full sequence-based window-level embeddings, we also delved into the interpretability of ProtT5-derived embedding tensors in two ways: firstly, by scrutinizing the attention weights obtained from the transformer's encoder block; and secondly, by computing SHAP values for these tensors, providing a model-agnostic interpretation of the prediction results. Additionally, we enhance the latent representation of ProtT5 by incorporating two additional local representations, one derived from amino acid properties and the other from supervised embedding layer, through an intermediate fusion stacked generalization approach, using an n-mer window sequence (or, peptide/fragment). The resultant stacked model, dubbed LMCrot, exhibits a more pronounced improvement in predictive performance across the tested datasets.

AVAILABILITY AND IMPLEMENTATION

LMCrot is publicly available at https://github.com/KCLabMTU/LMCrot.

MVNN-HNHC:A multi-view neural network for identification of human non-histone crotonylation sites

Crotonylation on lysine sites in human non-histone proteins plays a crucial role in biology activities. However, because traditional experimental methods for crotonylation site identification are time-consuming and labor-intensive, computational prediction methods have become increasingly popular in recent years. Despite its significance, crotonylation site prediction has received less attention in non-histone proteins than in histones. In this study, we proposed a Multi-View Neural Network for identification of Human Non-Histone Crotonylation sites, named MVNN-HNHC. MVNN-HNHC integrated multi-view encoding features and adaptive encoding features through multi-channel neural network to deeply learn about attribute differences between crotonylation sites and non-crotonylation sites from various aspects. In MVNN-HNHC, convolutional neural networks can obtain local information from these features, and bidirectional long short term memory networks were utilized to extract sequence information. Then, we employ the attention mechanism to fuse the outputs of various feature extraction modules. Finally, the fully connection network acted as the classifier to predict whether a lysine site was crotonylation site or non-crotonylation site. Performance metrics on independent test set, including sensitivity, specificity, accuracy, Matthews correlation coefficient, and area under the curve (AUC) values reach 80.06 %, 75.77 %, 77.06 %, 0.5203, and 0.7792, respectively. To verify the effectiveness of this method, we carry out a series of experiments and the results show that MVNN-HNHC is an effective tool for predicting crotonylation sites in non-histone proteins. The data and code are available on https://github.com/xbbxhbc/junjun0612.git.

PlantNh-Kcr: a deep learning model for predicting non-histone crotonylation sites in plants

BACKGROUND

Lysine crotonylation (Kcr) is a crucial protein post-translational modification found in histone and non-histone proteins. It plays a pivotal role in regulating diverse biological processes in both animals and plants, including gene transcription and replication, cell metabolism and differentiation, as well as photosynthesis. Despite the significance of Kcr, detection of Kcr sites through biological experiments is often time-consuming, expensive, and only a fraction of crotonylated peptides can be identified. This reality highlights the need for efficient and rapid prediction of Kcr sites through computational methods. Currently, several machine learning models exist for predicting Kcr sites in humans, yet models tailored for plants are rare. Furthermore, no downloadable Kcr site predictors or datasets have been developed specifically for plants. To address this gap, it is imperative to integrate existing Kcr sites detected in plant experiments and establish a dedicated computational model for plants.

RESULTS

Most plant Kcr sites are located on non-histones. In this study, we collected non-histone Kcr sites from five plants, including wheat, tabacum, rice, peanut, and papaya. We then conducted a comprehensive analysis of the amino acid distribution surrounding these sites. To develop a predictive model for plant non-histone Kcr sites, we combined a convolutional neural network (CNN), a bidirectional long short-term memory network (BiLSTM), and attention mechanism to build a deep learning model called PlantNh-Kcr. On both five-fold cross-validation and independent tests, PlantNh-Kcr outperformed multiple conventional machine learning models and other deep learning models. Furthermore, we conducted an analysis of species-specific effect on the PlantNh-Kcr model and found that a general model trained using data from multiple species outperforms species-specific models.

CONCLUSION

PlantNh-Kcr represents a valuable tool for predicting plant non-histone Kcr sites. We expect that this model will aid in addressing key challenges and tasks in the study of plant crotonylation sites.

O-GlcNAcPRED-DL: Prediction of Protein O-GlcNAcylation Sites Based on an Ensemble Model of Deep Learning

O-linked β-N-acetylglucosamine (O-GlcNAc) is a post-translational modification (i.e., O-GlcNAcylation) on serine/threonine residues of proteins, regulating a plethora of physiological and pathological events. As a dynamic process, O-GlcNAc functions in a site-specific manner. However, the experimental identification of the O-GlcNAc sites remains challenging in many scenarios. Herein, by leveraging the recent progress in cataloguing experimentally identified O-GlcNAc sites and advanced deep learning approaches, we establish an ensemble model, O-GlcNAcPRED-DL, a deep learning-based tool, for the prediction of O-GlcNAc sites. In brief, to make a benchmark O-GlcNAc data set, we extracted the information on O-GlcNAc from the recently constructed database O-GlcNAcAtlas, which contains thousands of experimentally identified and curated O-GlcNAc sites on proteins from multiple species. To overcome the imbalance between positive and negative data sets, we selected five groups of negative data sets in humans and mice to construct an ensemble predictor based on connection of a convolutional neural network and bidirectional long short-term memory. By taking into account three types of sequence information, we constructed four network frameworks, with the systematically optimized parameters used for the models. The thorough comparison analysis on two independent data sets of humans and mice and six independent data sets from other species demonstrated remarkably increased sensitivity and accuracy of the O-GlcNAcPRED-DL models, outperforming other existing tools. Moreover, a user-friendly Web server for O-GlcNAcPRED-DL has been constructed, which is freely available at http://oglcnac.org/pred_dl.

LSA-ac4C: A hybrid neural network incorporating double-layer LSTM and self-attention mechanism for the prediction of N4-acetylcytidine sites in human mRNA

N4-acetylcytidine (ac4C) is a vital constituent of the epitranscriptome and plays a crucial role in the regulation of mRNA expression. Numerous studies have established correlations between ac4C and the incidence, progression and prognosis of various cancers. Therefore, accurately predicting ac4C sites is an important step towards comprehending the biological functions of this modification and devising effective therapeutic interventions. Wet experiments are primary methods for studying ac4C, but computational methods have emerged as a promising supplement due to their cost-effectiveness and shorter research cycles. However, current models still have inherent limitations in terms of predictive performance and generalization ability. Here, we utilized automated machine learning technology to establish a reliable baseline and constructed a deep hybrid neural network, LSA-ac4C, which combines double-layer Long Short-Term Memory (LSTM) and self-attention mechanism for accurate ac4C sites prediction. Benchmarking comparisons demonstrate that LSA-ac4C exhibits superior performance compared to the current state-of-the-art method, with ACC, MCC and AUROC improving by 2.89 %, 5.96 % and 1.53 %, respectively, on an independent test set. Overall, LSA-ac4C serves as a powerful tool for predicting ac4C sites in human mRNA, thus benefiting research on RNA modification. For the convenience of the research community, a web server has been established at http://tubic.org/ac4C.

RMTLysPTM: recognizing multiple types of lysine PTM sites by deep analysis on sequences

Post-translational modification (PTM) occurs after a protein is translated from ribonucleic acid. It is an important living creature life phenomenon because it is implicated in almost all cellular processes. Identification of PTM sites from a given protein sequence is a hot topic in bioinformatics. Lots of computational methods have been proposed, and they provide good performance. However, most previous methods can only tackle one PTM type. Few methods consider multiple PTM types. In this study, a multi-label classification model, named RMTLysPTM, was developed to recognize four types of lysine (K) PTM sites, including acetylation, crotonylation, methylation and succinylation. The surrounding sites of a lysine site were selected to constitute a peptide segment, representing the lysine at the center. Deep analysis was conducted to count the distribution of 2-residues with fixed location across the four types of lysine PTM sites. By aggregating the distribution information of 2-residues in one peptide segment, the peptide segment was encoded by informative features. Furthermore, a prediction engine that can precisely capture the traits of the above representations was designed to recognize the types of lysine PTM sites. The cross-validation results on two datasets (Qiu and CPLM training datasets) suggested that the model had extremely high performance and RMTLysPTM had strong generalization ability by testing it on protein Q16778 and CPLM testing datasets. The model was found to be generally superior to all previous models and those using popular methods and features. A web server was set up for RMTLysPTM, and it can be accessed at http://119.3.127.138/.

MSTL-Kace: Prediction of Prokaryotic Lysine Acetylation Sites Based on Multistage Transfer Learning Strategy

As one of the most important post-translational modifications (PTM), lysine acetylation (Kace) plays an important role in various biological activities. Traditional experimental methods for identifying Kace sites are inefficient and expensive. Instead, several machine learning methods have been developed for Kace site prediction, and hand-crafted features have been used to encode the protein sequences. However, there are still two challenges: the complex biological information may be under-represented by these manmade features and the small sample issue of some species needs to be addressed. We propose a novel model, MSTL-Kace, which was developed based on transfer learning strategy with pretrained bidirectional encoder representations from transformers (BERT) model. In this model, the high-level embeddings were extracted from species-specific BERT models, and a two-stage fine-tuning strategy was used to deal with small sample issue. Specifically, a domain-specific BERT model was pretrained using all of the sequences in our data sets, which was then fine-tuned, or two-stage fine-tuned based on the training data set of each species to obtain the species-specific BERT models. Afterward, the embeddings of residues were extracted from the fine-tuned model and fed to the different downstream learning algorithms. After comparison, the best model for the six prokaryotic species was built by using a random forest. The results for the independent test sets show that our model outperforms the state-of-the-art methods on all six species. The source codes and data for MSTL-Kace are available at https://github.com/leo97king/MSTL-Kace.

DapNet-HLA: Adaptive dual-attention mechanism network based on deep learning to predict non-classical HLA binding sites

Human leukocyte antigen (HLA) plays a vital role in immunomodulatory function. Studies have shown that immunotherapy based on non-classical HLA has essential applications in cancer, COVID-19, and allergic diseases. However, there are few deep learning methods to predict non-classical HLA alleles. In this work, an adaptive dual-attention network named DapNet-HLA is established based on existing datasets. Firstly, amino acid sequences are transformed into digital vectors by looking up the table. To overcome the feature sparsity problem caused by unique one-hot encoding, the fused word embedding method is used to map each amino acid to a low-dimensional word vector optimized with the training of the classifier. Then, we use the GCB (group convolution block), SENet attention (squeeze-and-excitation networks), BiLSTM (bidirectional long short-term memory network), and Bahdanau attention mechanism to construct the classifier. The use of SENet can make the weight of the effective feature map high, so that the model can be trained to achieve better results. Attention mechanism is an Encoder-Decoder model used to improve the effectiveness of RNN, LSTM or GRU (gated recurrent neural network). The ablation experiment shows that DapNet-HLA has the best adaptability for five datasets. On the five test datasets, the ACC index and MCC index of DapNet-HLA are 4.89% and 0.0933 higher than the comparison method, respectively. According to the ROC curve and PR curve verified by the 5-fold cross-validation, the AUC value of each fold has a slight fluctuation, which proves the robustness of the DapNet-HLA. The codes and datasets are accessible at https://github.com/JYY625/DapNet-HLA.

Deepro-Glu: combination of convolutional neural network and Bi-LSTM models using ProtBert and handcrafted features to identify lysine glutarylation sites

Lysine glutarylation (Kglu) is a newly discovered post-translational modification of proteins with important roles in mitochondrial functions, oxidative damage, etc. The established biological experimental methods to identify glutarylation sites are often time-consuming and costly. Therefore, there is an urgent need to develop computational methods for efficient and accurate identification of glutarylation sites. Most of the existing computational methods only utilize handcrafted features to construct the prediction model and do not consider the positive impact of the pre-trained protein language model on the prediction performance. Based on this, we develop an ensemble deep-learning predictor Deepro-Glu that combines convolutional neural network and bidirectional long short-term memory network using the deep learning features and traditional handcrafted features to predict lysine glutaryation sites. The deep learning features are generated from the pre-trained protein language model called ProtBert, and the handcrafted features consist of sequence-based features, physicochemical property-based features and evolution information-based features. Furthermore, the attention mechanism is used to efficiently integrate the deep learning features and the handcrafted features by learning the appropriate attention weights. 10-fold cross-validation and independent tests demonstrate that Deepro-Glu achieves competitive or superior performance than the state-of-the-art methods. The source codes and data are publicly available at https://github.com/xwanggroup/Deepro-Glu.

Identification of species-specific RNA N6-methyladinosine modification sites from RNA sequences

N6-methyladinosine (m6A) modification is the most abundant co-transcriptional modification in eukaryotic RNA and plays important roles in cellular regulation. Traditional high-throughput sequencing experiments used to explore functional mechanisms are time-consuming and labor-intensive, and most of the proposed methods focused on limited species types. To further understand the relevant biological mechanisms among different species with the same RNA modification, it is necessary to develop a computational scheme that can be applied to different species. To achieve this, we proposed an attention-based deep learning method, adaptive-m6A, which consists of convolutional neural network, bi-directional long short-term memory and an attention mechanism, to identify m6A sites in multiple species. In addition, three conventional machine learning (ML) methods, including support vector machine, random forest and logistic regression classifiers, were considered in this work. In addition to the performance of ML methods for multi-species prediction, the optimal performance of adaptive-m6A yielded an accuracy of 0.9832 and the area under the receiver operating characteristic curve of 0.98. Moreover, the motif analysis and cross-validation among different species were conducted to test the robustness of one model towards multiple species, which helped improve our understanding about the sequence characteristics and biological functions of RNA modifications in different species.

FCCCSR_Glu: a semi-supervised learning model based on FCCCSR algorithm for prediction of glutarylation sites

Glutarylation is a post-translational modification which plays an irreplaceable role in various functions of the cell. Therefore, it is very important to accurately identify the glutarylation substrates and its corresponding glutarylation sites. In recent years, many computational methods of glutarylation sites have emerged one after another, but there are still many limitations, among which noisy data and the class imbalance problem caused by the uncertainty of non-glutarylation sites are great challenges. In this study, we propose a new semi-supervised learning algorithm, named FCCCSR, to identify reliable non-glutarylation lysine sites from unlabeled samples as negative samples. FCCCSR first finds core objects from positive samples according to reverse nearest neighbor information, and then clusters core objects based on natural neighbor structure. Finally, reliable negative samples are selected according to clustering result. With FCCCSR algorithm, we propose a new method named FCCCSR_Glu for glutarylation sites identification. In this study, multi-view features are extracted and fused to describe peptides, including amino acid composition, BLOSUM62, amino acid factors and composition of k-spaced amino acid pairs. Then, reliable negative samples selected by FCCCSR and positive samples are combined to establish models and XGBoost optimized by differential evolution algorithm is used as the classifier. On the independent testing dataset, FCCCSR_Glu achieves 85.18%, 98.36%, 94.31% and 0.8651 in sensitivity, specificity, accuracy and Matthew's Correlation Coefficient, respectively, which is superior to state-of-the-art methods in predicting glutarylation sites. Therefore, FCCCSR_Glu can be a useful tool for glutarylation sites prediction and FCCCSR algorithm can effectively select reliable negative samples from unlabeled samples. The data and code are available on https://github.com/xbbxhbc/FCCCSR_Glu.git.

Antimicrobial Peptides Prediction method based on sequence multidimensional feature embedding

Antimicrobial peptides (AMPs) are alkaline substances with efficient bactericidal activity produced in living organisms. As the best substitute for antibiotics, they have been paid more and more attention in scientific research and clinical application. AMPs can be produced from almost all organisms and are capable of killing a wide variety of pathogenic microorganisms. In addition to being antibacterial, natural AMPs have many other therapeutically important activities, such as wound healing, antioxidant and immunomodulatory effects. To discover new AMPs, the use of wet experimental methods is expensive and difficult, and bioinformatics technology can effectively solve this problem. Recently, some deep learning methods have been applied to the prediction of AMPs and achieved good results. To further improve the prediction accuracy of AMPs, this paper designs a new deep learning method based on sequence multidimensional representation. By encoding and embedding sequence features, and then inputting the model to identify AMPs, high-precision classification of AMPs and Non-AMPs with lengths of 10–200 is achieved. The results show that our method improved accuracy by 1.05% compared to the most advanced model in independent data validation without decreasing other indicators.

PD-BertEDL: An Ensemble Deep Learning Method Using BERT and Multivariate Representation to Predict Peptide Detectability

Peptide detectability is defined as the probability of identifying a peptide from a mixture of standard samples, which is a key step in protein identification and analysis. Exploring effective methods for predicting peptide detectability is helpful for disease treatment and clinical research. However, most existing computational methods for predicting peptide detectability rely on a single information. With the increasing complexity of feature representation, it is necessary to explore the influence of multivariate information on peptide detectability. Thus, we propose an ensemble deep learning method, PD-BertEDL. Bidirectional encoder representations from transformers (BERT) is introduced to capture the context information of peptides. Context information, sequence information, and physicochemical information of peptides were combined to construct the multivariate feature space of peptides. We use different deep learning methods to capture the high-quality features of different categories of peptides information and use the average fusion strategy to integrate three model prediction results to solve the heterogeneity problem and to enhance the robustness and adaptability of the model. The experimental results show that PD-BertEDL is superior to the existing prediction methods, which can effectively predict peptide detectability and provide strong support for protein identification and quantitative analysis, as well as disease treatment.

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.