A deep learning-based method for the prediction of DNA interacting residues in a protein

A comprehensive review of computational methods for Protein-DNA binding site prediction

Accurately identifying protein-DNA binding sites is essential for understanding the molecular mechanisms underlying biological processes, which in turn facilitates advancements in drug discovery and design. While biochemical experiments provide the most accurate way to locate DNA-binding sites, they are generally time-consuming, resource-intensive, and expensive. There is a pressing need to develop computational methods that are both efficient and accurate for DNA-binding site prediction. This study thoroughly reviews and categorizes major computational approaches for predicting DNA-binding sites, including template detection, statistical machine learning, and deep learning-based methods. The 14 state-of-the-art DNA-binding site prediction models have been benchmarked on 136 non-redundant proteins, where the deep learning-based, especially pre-trained large language model-based, methods achieve superior performance over the other two categories. Applications of these DNA-binding site prediction methods are also involved.

iProtDNA-SMOTE: Enhancing protein-DNA binding sites prediction through imbalanced graph neural networks

Protein-DNA interactions play a crucial role in cellular biology, essential for maintaining life processes and regulating cellular functions. We propose a method called iProtDNA-SMOTE, which utilizes non-equilibrium graph neural networks along with pre-trained protein language models to predict DNA binding residues. This approach effectively addresses the class imbalance issue in predicting protein-DNA binding sites by leveraging unbalanced graph data, thus enhancing model's generalization and specificity. We trained the model on two datasets, TR646 and TR573, and conducted a series of experiments to evaluate its performance. The model achieved AUC values of 0.850, 0.896, and 0.858 on the independent test datasets TE46, TE129, and TE181, respectively. These results indicate that iProtDNA-SMOTE outperforms existing methods in terms of accuracy and generalization for predicting DNA binding sites, offering reliable and effective predictions to minimize errors. The model has been thoroughly validated for its ability to predict protein-DNA binding sites with high reliability and precision. For the convenience of the scientific community, the benchmark datasets and codes are publicly available at https://github.com/primrosehry/iProtDNA-SMOTE.

Integrating Deep Learning Models with Genome-Wide Association Study-Based Identification Enhanced Phenotype Predictions in Group A Streptococcus

Group A Streptococcus (GAS) is a major pathogen with diverse clinical outcomes linked to its genetic variability, making accurate phenotype prediction essential. While previous studies have identified many GAS-associated genetic factors, translating these findings into predictive models remains challenging due to data complexity. The current study aimed to integrate deep learning models with genome-wide association study-derived genetic variants to predict pathogenic phenotypes in GAS. We evaluated the performance of several deep neural network models, including CNN, ResNet18, LSTM, and their ensemble approach in predicting GAS phenotypes. It was found that the ensemble model consistently achieved the highest prediction accuracy across phenotypes. Models trained on the full 4722-genotype set outperformed those trained on a reduced 175-genotype set, underscoring the importance of comprehensive variant data in capturing complex genotype-phenotype interactions. Performance changes in the reduced 175-genotype set compared to the full-set genotype scenarios revealed the impact of data dimensionality on model effectiveness, with CNN remaining robust, while ResNet18 and LSTM underperformed. Our findings emphasized the potential of deep learning in phenotype prediction and the critical role of data-model compatibility.

Deep-HPI-pred: An R-Shiny applet for network-based classification and prediction of Host-Pathogen protein-protein interactions

Host-pathogen interactions (HPIs) are vital in numerous biological activities and are intrinsically linked to the onset and progression of infectious diseases. HPIs are pivotal in the entire lifecycle of diseases: from the onset of pathogen introduction, navigating through the mechanisms that bypass host cellular defenses, to its subsequent proliferation inside the host. At the heart of these stages lies the synergy of proteins from both the host and the pathogen. By understanding these interlinking protein dynamics, we can gain crucial insights into how diseases progress and pave the way for stronger plant defenses and the swift formulation of countermeasures. In the framework of current study, we developed a web-based R/Shiny app, Deep-HPI-pred, that uses network-driven feature learning method to predict the yet unmapped interactions between pathogen and host proteins. Leveraging citrus and CLas bacteria training datasets as case study, we spotlight the effectiveness of Deep-HPI-pred in discerning Protein-protein interaction (PPIs) between them. Deep-HPI-pred use Multilayer Perceptron (MLP) models for HPI prediction, which is based on a comprehensive evaluation of topological features and neural network architectures. When subjected to independent validation datasets, the predicted models consistently surpassed a Matthews correlation coefficient (MCC) of 0.80 in host-pathogen interactions. Remarkably, the use of Eigenvector Centrality as the leading topological feature further enhanced this performance. Further, Deep-HPI-pred also offers relevant gene ontology (GO) term information for each pathogen and host protein within the system. This protein annotation data contributes an additional layer to our understanding of the intricate dynamics within host-pathogen interactions. In the additional benchmarking studies, the Deep-HPI-pred model has proven its robustness by consistently delivering reliable results across different host-pathogen systems, including plant-pathogens (accuracy of 98.4% and 97.9%), human-virus (accuracy of 94.3%), and animal-bacteria (accuracy of 96.6%) interactomes. These results not only demonstrate the model's versatility but also pave the way for gaining comprehensive insights into the molecular underpinnings of complex host-pathogen interactions. Taken together, the Deep-HPI-pred applet offers a unified web service for both identifying and illustrating interaction networks. Deep-HPI-pred applet is freely accessible at its homepage: https://cbi.gxu.edu.cn/shiny-apps/Deep-HPI-pred/ and at github: https://github.com/tahirulqamar/Deep-HPI-pred.

Twenty years of advances in prediction of nucleic acid-binding residues in protein sequences

Computational prediction of nucleic acid-binding residues in protein sequences is an active field of research, with over 80 methods that were released in the past 2 decades. We identify and discuss 87 sequence-based predictors that include dozens of recently published methods that are surveyed for the first time. We overview historical progress and examine multiple practical issues that include availability and impact of predictors, key features of their predictive models, and important aspects related to their training and assessment. We observe that the past decade has brought increased use of deep neural networks and protein language models, which contributed to substantial gains in the predictive performance. We also highlight advancements in vital and challenging issues that include cross-predictions between deoxyribonucleic acid (DNA)-binding and ribonucleic acid (RNA)-binding residues and targeting the two distinct sources of binding annotations, structure-based versus intrinsic disorder-based. The methods trained on the structure-annotated interactions tend to perform poorly on the disorder-annotated binding and vice versa, with only a few methods that target and perform well across both annotation types. The cross-predictions are a significant problem, with some predictors of DNA-binding or RNA-binding residues indiscriminately predicting interactions with both nucleic acid types. Moreover, we show that methods with web servers are cited substantially more than tools without implementation or with no longer working implementations, motivating the development and long-term maintenance of the web servers. We close by discussing future research directions that aim to drive further progress in this area.

Deciphering the Language of Protein-DNA Interactions: A Deep Learning Approach Combining Contextual Embeddings and Multi-Scale Sequence Modeling

Deciphering the mechanisms governing protein-DNA interactions is crucial for understanding key cellular processes and disease pathways. In this work, we present a powerful deep learning approach that significantly advances the computational prediction of DNA-interacting residues from protein sequences. Our method leverages the rich contextual representations learned by pre-trained protein language models, such as ProtTrans, to capture intrinsic biochemical properties and sequence motifs indicative of DNA binding sites. We then integrate these contextual embeddings with a multi-window convolutional neural network architecture, which scans across the sequence at varying window sizes to effectively identify both local and global binding patterns. Comprehensive evaluation on curated benchmark datasets demonstrates the remarkable performance of our approach, achieving an area under the ROC curve (AUC) of 0.89 - a substantial improvement over previous state-of-the-art sequence-based predictors. This showcases the immense potential of pairing advanced representation learning and deep neural network designs for uncovering the complex syntax governing protein-DNA interactions directly from primary sequences. Our work not only provides a robust computational tool for characterizing DNA-binding mechanisms, but also highlights the transformative opportunities at the intersection of language modeling, deep learning, and protein sequence analysis. The publicly available code and data further facilitate broader adoption and continued development of these techniques for accelerating mechanistic insights into vital biological processes and disease pathways. In addition, the code and data for this work are available at https://github.com/B1607/DIRP.

A hybrid method for discovering interferon-gamma inducing peptides in human and mouse

Interferon-gamma (IFN-γ) is a versatile pleiotropic cytokine essential for both innate and adaptive immune responses. It exhibits both pro-inflammatory and anti-inflammatory properties, making it a promising therapeutic candidate for treating various infectious diseases and cancers. We present IFNepitope2, a host-specific technique to annotate IFN-γ inducing peptides, it is an updated version of IFNepitope introduced by Dhanda et al. In this study, dataset used for developing prediction method contain experimentally validated 25,492 and 7983 IFN-γ inducing peptides in human and mouse host, respectively. In initial phase, machine learning techniques have been exploited to develop classification model using wide range of peptide features. Further, to improve machine learning based models or alignment free models, we explore potential of similarity-based technique BLAST. Finally, a hybrid model has been developed that combine best machine learning based model with BLAST. In most of the case, models based on extra tree perform better than other machine learning techniques. In case of peptide features, compositional feature particularly dipeptide composition performs better than one-hot encoding or binary profile. Our best machine learning based models achieved AUROC 0.89 and 0.83 for human and mouse host, respectively. The hybrid model achieved the AUROC 0.90 and 0.85 for human and mouse host, respectively. All models have been evaluated on an independent/validation dataset not used for training or testing these models. Newly developed method performs better than existing method on independent dataset. The major objective of this study is to predict, design and scan IFN-γ inducing peptides, thus server/software have been developed ( https://webs.iiitd.edu.in/raghava/ifnepitope2/ ). This method is also available as standalone at https://github.com/raghavagps/ifnepitope2 and python package index at https://pypi.org/project/ifnepitope2/ .

PDNAPred: Interpretable prediction of protein-DNA binding sites based on pre-trained protein language models

Protein-DNA interactions play critical roles in various biological processes and are essential for drug discovery. However, traditional experimental methods are labor-intensive and unable to keep pace with the increasing volume of protein sequences, leading to a substantial number of proteins lacking DNA-binding annotations. Therefore, developing an efficient computational method to identify protein-DNA binding sites is crucial. Unfortunately, most existing computational methods rely on manually selected features or protein structure information, making these methods inapplicable to large-scale prediction tasks. In this study, we introduced PDNAPred, a sequence-based method that combines two pre-trained protein language models with a designed CNN-GRU network to identify DNA-binding sites. Additionally, to tackle the issue of imbalanced dataset samples, we employed focal loss. Our comprehensive experiments demonstrated that PDNAPred significantly improved the accuracy of DNA-binding site prediction, outperforming existing state-of-the-art sequence-based methods. Remarkably, PDNAPred also achieved results comparable to advanced structure-based methods. The designed CNN-GRU network enhances its capability to detect DNA-binding sites accurately. Furthermore, we validated the versatility of PDNAPred by training it on RNA-binding site datasets, showing its potential as a general framework for amino acid binding site prediction. Finally, we conducted model interpretability analysis to elucidate the reasons behind PDNAPred's outstanding performance.

Advances in the Application of Protein Language Modeling for Nucleic Acid Protein Binding Site Prediction

Protein and nucleic acid binding site prediction is a critical computational task that benefits a wide range of biological processes. Previous studies have shown that feature selection holds particular significance for this prediction task, making the generation of more discriminative features a key area of interest for many researchers. Recent progress has shown the power of protein language models in handling protein sequences, in leveraging the strengths of attention networks, and in successful applications to tasks such as protein structure prediction. This naturally raises the question of the applicability of protein language models in predicting protein and nucleic acid binding sites. Various approaches have explored this potential. This paper first describes the development of protein language models. Then, a systematic review of the latest methods for predicting protein and nucleic acid binding sites is conducted by covering benchmark sets, feature generation methods, performance comparisons, and feature ablation studies. These comparisons demonstrate the importance of protein language models for the prediction task. Finally, the paper discusses the challenges of protein and nucleic acid binding site prediction and proposes possible research directions and future trends. The purpose of this survey is to furnish researchers with actionable suggestions for comprehending the methodologies used in predicting protein-nucleic acid binding sites, fostering the creation of protein-centric language models, and tackling real-world obstacles encountered in this field.

A hybrid approach for predicting transcription factors

Transcription factors are essential DNA-binding proteins that regulate the transcription rate of several genes and control the expression of genes inside a cell. The prediction of transcription factors with high precision is important for understanding biological processes such as cell differentiation, intracellular signaling, and cell-cycle control. In this study, we developed a hybrid method that combines alignment-based and alignment-free methods for predicting transcription factors with higher accuracy. All models have been trained, tested, and evaluated on a large dataset that contains 19,406 transcription factors and 523,560 non-transcription factor protein sequences. To avoid biases in evaluation, the datasets were divided into training and validation/independent datasets, where 80% of the data was used for training, and the remaining 20% was used for external validation. In the case of alignment-free methods, models were developed using machine learning techniques and the composition-based features of a protein. Our best alignment-free model obtained an AUC of 0.97 on an independent dataset. In the case of the alignment-based method, we used BLAST at different cut-offs to predict the transcription factors. Although the alignment-based method demonstrated excellent performance, it was unable to cover all transcription factors due to instances of no hits. To combine the strengths of both methods, we developed a hybrid method that combines alignment-free and alignment-based methods. In the hybrid method, we added the scores of the alignment-free and alignment-based methods and achieved a maximum AUC of 0.99 on the independent dataset. The method proposed in this study performs better than existing methods. We incorporated the best models in the webserver/Python Package Index/standalone package of "TransFacPred" (https://webs.iiitd.edu.in/raghava/transfacpred).

MucLiPred: Multi-Level Contrastive Learning for Predicting Nucleic Acid Binding Residues of Proteins

Protein-molecule interactions play a crucial role in various biological functions, with their accurate prediction being pivotal for drug discovery and design processes. Traditional methods for predicting protein-molecule interactions are limited. Some can only predict interactions with a specific molecule, restricting their applicability, while others target multiple molecule types but fail to efficiently process diverse interaction information, leading to complexity and inefficiency. This study presents a novel deep learning model, MucLiPred, equipped with a dual contrastive learning mechanism aimed at improving the prediction of multiple molecule-protein interactions and the identification of potential molecule-binding residues. The residue-level paradigm focuses on differentiating binding from non-binding residues, illuminating detailed local interactions. The type-level paradigm, meanwhile, analyzes overarching contexts of molecule types, like DNA or RNA, ensuring that representations of identical molecule types gravitate closer in the representational space, bolstering the model's proficiency in discerning interaction motifs. This dual approach enables comprehensive multi-molecule predictions, elucidating the relationships among different molecule types and strengthening precise protein-molecule interaction predictions. Empirical evidence demonstrates MucLiPred's superiority over existing models in robustness and prediction accuracy. The integration of dual contrastive learning techniques amplifies its capability to detect potential molecule-binding residues with precision. Further optimization, separating representational and classification tasks, has markedly improved its performance. MucLiPred thus represents a significant advancement in protein-molecule interaction prediction, setting a new precedent for future research in this field.

Protein-DNA binding sites prediction based on pre-trained protein language model and contrastive learning

Protein-DNA interaction is critical for life activities such as replication, transcription and splicing. Identifying protein-DNA binding residues is essential for modeling their interaction and downstream studies. However, developing accurate and efficient computational methods for this task remains challenging. Improvements in this area have the potential to drive novel applications in biotechnology and drug design. In this study, we propose a novel approach called Contrastive Learning And Pre-trained Encoder (CLAPE), which combines a pre-trained protein language model and the contrastive learning method to predict DNA binding residues. We trained the CLAPE-DB model on the protein-DNA binding sites dataset and evaluated the model performance and generalization ability through various experiments. The results showed that the area under ROC curve values of the CLAPE-DB model on the two benchmark datasets reached 0.871 and 0.881, respectively, indicating superior performance compared to other existing models. CLAPE-DB showed better generalization ability and was specific to DNA-binding sites. In addition, we trained CLAPE on different protein-ligand binding sites datasets, demonstrating that CLAPE is a general framework for binding sites prediction. To facilitate the scientific community, the benchmark datasets and codes are freely available at https://github.com/YAndrewL/clape.

Prediction of RNA-interacting residues in a protein using CNN and evolutionary profile

This paper describes a method Pprint2, which is an improved version of Pprint developed for predicting RNA-interacting residues in a protein. Training and independent/validation datasets used in this study comprises of 545 and 161 non-redundant RNA-binding proteins, respectively. All models were trained on training dataset and evaluated on the validation dataset. The preliminary analysis reveals that positively charged amino acids such as H, R and K, are more prominent in the RNA-interacting residues. Initially, machine learning based models have been developed using binary profile and obtain maximum area under curve (AUC) 0.68 on validation dataset. The performance of this model improved significantly from AUC 0.68 to 0.76, when evolutionary profile is used instead of binary profile. The performance of our evolutionary profile-based model improved further from AUC 0.76 to 0.82, when convolutional neural network has been used for developing model. Our final model based on convolutional neural network using evolutionary information achieved AUC 0.82 with Matthews correlation coefficient of 0.49 on the validation dataset. Our best model outperforms existing methods when evaluated on the independent/validation dataset. A user-friendly standalone software and web-based server named 'Pprint2' has been developed for predicting RNA-interacting residues (https://webs.iiitd.edu.in/raghava/pprint2 and https://github.com/raghavagps/pprint2).

Prediction of celiac disease associated epitopes and motifs in a protein

Introduction

Celiac disease (CD) is an autoimmune gastrointestinal disorder causes immune-mediated enteropathy against gluten. Gluten immunogenic peptides have the potential to trigger immune responses which leads to damage the small intestine. HLA-DQ2/DQ8 are major alleles that bind to epitope/antigenic region of gluten and induce celiac disease. There is a need to identify CD associated epitopes in protein-based foods and therapeutics.

Methods

In this study, computational tools have been developed to predict CD associated epitopes and motifs. Dataset used for training, testing and evaluation contain experimentally validated CD associated and non-CD associate peptides. We perform positional analysis to identify the most significant position of an amino acid residue in the peptide and checked the frequency of HLA alleles. We also compute amino acid composition to develop machine learning based models. We also developed ensemble method that combines motif-based approach and machine learning based models.

Results and Discussion

Our analysis support existing hypothesis that proline (P) and glutamine (Q) are highly abundant in CD associated peptides. A model based on density of P&Q in peptides has been developed for predicting CD associated peptides which achieve maximum AUROC 0.98 on independent data. We discovered motifs (e.g., QPF, QPQ, PYP) which occurs specifically in CD associated peptides. We also developed machine learning based models using peptide composition and achieved maximum AUROC 0.99. Finally, we developed ensemble method that combines motif-based approach and machine learning based models. The ensemble model-predict CD associated motifs with 100% accuracy on an independent dataset, not used for training. Finally, the best models and motifs has been integrated in a web server and standalone software package "CDpred". We hope this server anticipate the scientific community for the prediction, designing and scanning of CD associated peptides as well as CD associated motifs in a protein/peptide sequence (https://webs.iiitd.edu.in/raghava/cdpred/).