Interleukin-6 (IL-6) is a key immunomodulatory cytokine implicated in diverse physiological processes and pathological conditions, including autoimmune diseases, cancers, and cytokine storms. Immunogenic peptides capable of inducing IL-6 expression are key modulators of host immune responses and represent promising candidates for therapeutic design and epitope-based vaccine development. However, experimental identification of IL-6-inducing peptides remains laborious and unsuitable for large-scale screening. Although existing computational approaches show promise, many often struggle to capture both global contextual semantics and local motif-level features essential for peptide immunogenicity. To address these limitations, we present CONTRA-IL6, a novel deep learning framework that integrates Transformer fusion and convolutional localization modules with stacked pretrained protein language model embeddings to predict IL-6-inducing peptides. Comprehensive benchmarking on an independent dataset demonstrates that CONTRA-IL6 achieves superior predictive performance over six state-of-the-art predictors. Notably, it achieves the highest Matthews correlation coefficient (MCC, 0.504) and F1 (0.549) and improves over the best-performing existing method by 3.2% in MCC and 4.3% in F1, demonstrating balanced and robust performance. Feature space visualizations (uniform manifold approximation and projection, kernel density estimation) showed clear class separation, while 1D gradient-weighted class activation mapping++ highlighted strong attention to specific C-terminal regions. Crucially, we moved beyond these attribution methods by employing in silico mutagenesis, which causally confirmed the functional importance and physicochemical constraints. Ablation studies further confirmed the synergistic contribution of global and local modules to model performance. CONTRA-IL6 offers a robust, scalable, and interpretable solution for immunoinformatics research. The standalone package is freely available at https://pypi.org/project/contra-il6/ to facilitate broader community use.

Threonine proteases (TPs) are enzymes vital for several biological processes and diseases including Alzheimer's disease and cancer. Their potential to target and degrade proteins intracellularly makes them valuable for various therapeutic and industrial applications. However, traditional experimental methods for identifying and characterizing novel TPs are exhaustive, time-consuming, and expensive. To address this, we developed TP-ML, a support vector machine-based prediction tool that can differentiate TP from non-TP sequences. We generated a benchmark dataset and calculated the physicochemical and compositional features using primary amino acid sequences. Subsequently, a comparison was made between the two feature selection approaches to identify the optimal feature sets from the original encodings. These optimal features were then used to train five different machine-learning classifiers, each assessed independently. TP-ML was selected as the best model showing consistent performance during cross-validation and independent evaluation, and achieved an accuracy of 0.934 and 0.888, respectively. We anticipate TP-ML to be a powerful tool for identifying TPs, aiding in their experimental characterization and industrial application exploration.

Large language models (LLMs) have significantly advanced computational biology by enabling the integration of molecular, protein, and natural language data to accelerate drug discovery. However, existing molecular captioning approaches often underutilize diverse molecular modalities and lack interpretability. In this study, we introduce XMolCap, a novel explainable molecular captioning framework that integrates molecular images, SMILES strings, and graph-based structures through a stacked multimodal fusion mechanism. The framework is built upon a BioT5-based encoder-decoder architecture, which serves as the backbone for extracting feature representations from SELFIES. By leveraging specialized models such as SwinOCSR, SciBERT, and GIN-MoMu, XMolCap effectively captures complementary information from each modality. Our model not only achieves state-of-the-art performance on two benchmark datasets (L+M-24 and ChEBI-20), outperforming several strong baselines, but also provides detailed, functional group-aware, and property-specific explanations through graph-based interpretation. XMolCap is publicly available at https://github.com/cbbl-skku-org/XMolCap/ for reproducibility and local deployment. We believe it holds strong potential for clinical and pharmaceutical applications by generating accurate, interpretable molecular descriptions that deepen our understanding of molecular properties and interactions.

The worldwide appearance of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has generated significant concern and posed a considerable challenge to global health. Phosphorylation is a common post-translational modification that affects many vital cellular functions and is closely associated with SARS-CoV-2 infection. Precise identification of phosphorylation sites could provide more in-depth insight into the processes underlying SARS-CoV-2 infection and help alleviate the continuing COVID-19 crisis. Currently, available computational tools for predicting these sites lack accuracy and effectiveness. In this study, we designed an innovative meta-learning model, Meta-Learning for Serine/Threonine Phosphorylation (MeL-STPhos), to precisely identify protein phosphorylation sites. We initially performed a comprehensive assessment of 29 unique sequence-derived features, establishing prediction models for each using 14 renowned machine learning methods, ranging from traditional classifiers to advanced deep learning algorithms. We then selected the most effective model for each feature by integrating the predicted values. Rigorous feature selection strategies were employed to identify the optimal base models and classifier(s) for each cell-specific dataset. To the best of our knowledge, this is the first study to report two cell-specific models and a generic model for phosphorylation site prediction by utilizing an extensive range of sequence-derived features and machine learning algorithms. Extensive cross-validation and independent testing revealed that MeL-STPhos surpasses existing state-of-the-art tools for phosphorylation site prediction. We also developed a publicly accessible platform at https://balalab-skku.org/MeL-STPhos. We believe that MeL-STPhos will serve as a valuable tool for accelerating the discovery of serine/threonine phosphorylation sites and elucidating their role in post-translational regulation.

2'-O-methylation (2OM) is the most common post-transcriptional modification of RNA. It plays a crucial role in RNA splicing, RNA stability and innate immunity. Despite advances in high-throughput detection, the chemical stability of 2OM makes it difficult to detect and map in messenger RNA. Therefore, bioinformatics tools have been developed using machine learning (ML) algorithms to identify 2OM sites. These tools have made significant progress, but their performances remain unsatisfactory and need further improvement. In this study, we introduced H2Opred, a novel hybrid deep learning (HDL) model for accurately identifying 2OM sites in human RNA. Notably, this is the first application of HDL in developing four nucleotide-specific models [adenine (A2OM), cytosine (C2OM), guanine (G2OM) and uracil (U2OM)] as well as a generic model (N2OM). H2Opred incorporated both stacked 1D convolutional neural network (1D-CNN) blocks and stacked attention-based bidirectional gated recurrent unit (Bi-GRU-Att) blocks. 1D-CNN blocks learned effective feature representations from 14 conventional descriptors, while Bi-GRU-Att blocks learned feature representations from five natural language processing-based embeddings extracted from RNA sequences. H2Opred integrated these feature representations to make the final prediction. Rigorous cross-validation analysis demonstrated that H2Opred consistently outperforms conventional ML-based single-feature models on five different datasets. Moreover, the generic model of H2Opred demonstrated a remarkable performance on both training and testing datasets, significantly outperforming the existing predictor and other four nucleotide-specific H2Opred models. To enhance accessibility and usability, we have deployed a user-friendly web server for H2Opred, accessible at https://balalab-skku.org/H2Opred/. This platform will serve as an invaluable tool for accurately predicting 2OM sites within human RNA, thereby facilitating broader applications in relevant research endeavors.

BACKGROUND: Cell-penetrating peptides (CPPs) have received considerable attention as a means of transporting pharmacologically active molecules into living cells without damaging the cell membrane, and thus hold great promise as future therapeutics. Recently, several machine learning-based algorithms have been proposed for predicting CPPs. However, most existing predictive methods do not consider the agreement (disagreement) between similar (dissimilar) CPPs and depend heavily on expert knowledge-based handcrafted features. RESULTS: In this study, we present SiameseCPP, a novel deep learning framework for automated CPPs prediction. SiameseCPP learns discriminative representations of CPPs based on a well-pretrained model and a Siamese neural network consisting of a transformer and gated recurrent units. Contrastive learning is used for the first time to build a CPP predictive model. Comprehensive experiments demonstrate that our proposed SiameseCPP is superior to existing baseline models for predicting CPPs. Moreover, SiameseCPP also achieves good performance on other functional peptide datasets, exhibiting satisfactory generalization ability.

Detection and analysis of viral genomes with Nanopore sequencing has shown great promise in the surveillance of pathogen outbreaks. However, the number of virus detection pipelines supporting Nanopore sequencing is very limited. Here, we present VirPipe, a new pipeline for the detection of viral genomes from Nanopore or Illumina sequencing input featuring streamlined installation and customization. AVAILABILITY AND IMPLEMENTATION: VirPipe source code and documentation are freely available for download at https://github.com/KijinKims/VirPipe, implemented in Python and Nextflow.

Recently, speech emotion recognition (SER) has become an active research area in speech processing, particularly with the advent of deep learning (DL). Numerous DL-based methods have been proposed for SER. However, most of the existing DL-based models are complex and require a large amounts of data to achieve a good performance. In this study, a new framework of deep attention-based dilated convolutional-recurrent neural networks coupled with a hybrid data augmentation method was proposed for addressing SER tasks. The hybrid data augmentation method constitutes an upsampling technique for generating more speech data samples based on the traditional and generative adversarial network approaches. By leveraging both convolutional and recurrent neural networks in a dilated form along with an attention mechanism, the proposed DL framework can extract high-level representations from three-dimensional log Mel spectrogram features. Dilated convolutional neural networks acquire larger receptive fields, whereas dilated recurrent neural networks overcome complex dependencies as well as the vanishing and exploding gradient issues. Furthermore, the loss functions are reconfigured by combining the SoftMax loss and the center-based losses to classify various emotional states. The proposed framework was implemented using the Python programming language and the TensorFlow deep learning library. To validate the proposed framework, the EmoDB and ERC benchmark datasets, which are imbalanced and/or small datasets, were employed. The experimental results indicate that the proposed framework outperforms other related state-of-the-art methods, yielding the highest unweighted recall rates of 88.03 ± 1.39 (%) and 66.56 ± 0.67 (%) for the EmoDB and ERC datasets, respectively.

Protein post-translational modification (PTM) is an important regulatory mechanism that plays a key role in both normal and disease states. Acetylation on lysine residues is one of the most potent PTMs owing to its critical role in cellular metabolism and regulatory processes. Identifying protein lysine acetylation (Kace) sites is a challenging task in bioinformatics. To date, several machine learning-based methods for the in silico identification of Kace sites have been developed. Of those, a few are prokaryotic species-specific. Despite their attractive advantages and performances, these methods have certain limitations. Therefore, this study proposes a novel predictor STALLION (STacking-based Predictor for ProkAryotic Lysine AcetyLatION), containing six prokaryotic species-specific models to identify Kace sites accurately. To extract crucial patterns around Kace sites, we employed 11 different encodings representing three different characteristics. Subsequently, a systematic and rigorous feature selection approach was employed to identify the optimal feature set independently for five tree-based ensemble algorithms and built their respective baseline model for each species. Finally, the predicted values from baseline models were utilized and trained with an appropriate classifier using the stacking strategy to develop STALLION. Comparative benchmarking experiments showed that STALLION significantly outperformed existing predictor on independent tests. To expedite direct accessibility to the STALLION models, a user-friendly online predictor was implemented, which is available at: http://thegleelab.org/STALLION.

Recently, machine learning methods have been developed to identify various peptide bio-activities. However, due to the lack of experimentally validated peptides, machine learning methods cannot provide a sufficiently trained model, easily resulting in poor generalizability. Furthermore, there is no generic computational framework to predict the bioactivities of different peptides. Thus, a natural question is whether we can use limited samples to build an effective predictive model for different kinds of peptides. To address this question, we propose Mutual Information Maximization Meta-Learning (MIMML), a novel meta-learning-based predictive model for bioactive peptide discovery. Using few samples from various functional peptides, MIMML can sufficiently learn the discriminative information amongst various functions and characterize functional differences. Experimental results show excellent performance of MIMML though using far fewer training samples as compared to the state-of-the-art methods. We also decipher the latent relationships among different kinds of functions to understand what meta-model learned to improve a specific task. In summary, this study is a pioneering work in the field of functional peptide mining and provides the first-of-its-kind solution for few-sample learning problems in biological sequence analysis, accelerating the new functional peptide discovery. The source codes and datasets are available on https://github.com/TearsWaiting/MIMML.

Long noncoding RNAs (lncRNAs) are primarily regulated by their cellular localization, which is responsible for their molecular functions, including cell cycle regulation and genome rearrangements. Accurately identifying the subcellular location of lncRNAs from sequence information is crucial for a better understanding of their biological functions and mechanisms. In contrast to traditional experimental methods, bioinformatics or computational methods can be applied for the annotation of lncRNA subcellular locations in humans more effectively. In the past, several machine learning-based methods have been developed to identify lncRNA subcellular localization, but relevant work for identifying cell-specific localization of human lncRNA remains limited. In this study, we present the first application of the tree-based stacking approach, TACOS, which allows users to identify the subcellular localization of human lncRNA in 10 different cell types. Specifically, we conducted comprehensive evaluations of six tree-based classifiers with 10 different feature descriptors, using a newly constructed balanced training dataset for each cell type. Subsequently, the strengths of the AdaBoost baseline models were integrated via a stacking approach, with an appropriate tree-based classifier for the final prediction. TACOS displayed consistent performance in both the cross-validation and independent assessments compared with the other two approaches employed in this study. The user-friendly online TACOS web server can be accessed at https://balalab-skku.org/TACOS.

The COVID-19 pandemic caused several million deaths worldwide. Development of anti-coronavirus drugs is thus urgent. Unlike conventional non-peptide drugs, antiviral peptide drugs are highly specific, easy to synthesize and modify, and not highly susceptible to drug resistance. To reduce the time and expense involved in screening thousands of peptides and assaying their antiviral activity, computational predictors for identifying anti-coronavirus peptides (ACVPs) are needed. However, few experimentally verified ACVP samples are available, even though a relatively large number of antiviral peptides (AVPs) have been discovered. In this study, we attempted to predict ACVPs using an AVP dataset and a small collection of ACVPs. Using conventional features, a binary profile and a word-embedding word2vec (W2V), we systematically explored five different machine learning methods: Transformer, Convolutional Neural Network, bidirectional Long Short-Term Memory, Random Forest (RF) and Support Vector Machine. Via exhaustive searches, we found that the RF classifier with W2V consistently achieved better performance on different datasets. The two main controlling factors were: (i) the dataset-specific W2V dictionary was generated from the training and independent test datasets instead of the widely used general UniProt proteome and (ii) a systematic search was conducted and determined the optimal k-mer value in W2V, which provides greater discrimination between positive and negative samples. Therefore, our proposed method, named iACVP, consistently provides better prediction performance compared with existing state-of-the-art methods. To assist experimentalists in identifying putative ACVPs, we implemented our model as a web server accessible via the following link: http://kurata35.bio.kyutech.ac.jp/iACVP.

Coronavirus disease 2019 (COVID-19) has impacted public health as well as societal and economic well-being. In the last two decades, various prediction algorithms and tools have been developed for predicting antiviral peptides (AVPs). The current COVID-19 pandemic has underscored the need to develop more efficient and accurate machine learning (ML)-based prediction algorithms for the rapid identification of therapeutic peptides against severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). Several peptide-based ML approaches, including anti-coronavirus peptides (ACVPs), IL-6 inducing epitopes and other epitopes targeting SARS-CoV-2, have been implemented in COVID-19 therapeutics. Owing to the growing interest in the COVID-19 field, it is crucial to systematically compare the existing ML algorithms based on their performances. Accordingly, we comprehensively evaluated the state-of-the-art IL-6 and AVP predictors against coronaviruses in terms of core algorithms, feature encoding schemes, performance evaluation metrics and software usability. A comprehensive performance assessment was then conducted to evaluate the robustness and scalability of the existing predictors using well-constructed independent validation datasets. Additionally, we discussed the advantages and disadvantages of the existing methods, providing useful insights into the development of novel computational tools for characterizing and identifying epitopes or ACVPs. The insights gained from this review are anticipated to provide critical guidance to the scientific community in the rapid design and development of accurate and efficient next-generation in silico tools against SARS-CoV-2.

Neuropeptides (NPs) are the most versatile neurotransmitters in the immune systems that regulate various central anxious hormones. An efficient and effective bioinformatics tool for rapid and accurate large-scale identification of NPs is critical in immunoinformatics, which is indispensable for basic research and drug development. Although a few NP prediction tools have been developed, it is mandatory to improve their NPs' prediction performances. In this study, we have developed a machine learning-based meta-predictor called NeuroPred-FRL by employing the feature representation learning approach. First, we generated 66 optimal baseline models by employing 11 different encodings, six different classifiers and a two-step feature selection approach. The predicted probability scores of NPs based on the 66 baseline models were combined to be deemed as the input feature vector. Second, in order to enhance the feature representation ability, we applied the two-step feature selection approach to optimize the 66-D probability feature vector and then inputted the optimal one into a random forest classifier for the final meta-model (NeuroPred-FRL) construction. Benchmarking experiments based on both cross-validation and independent tests indicate that the NeuroPred-FRL achieves a superior prediction performance of NPs compared with the other state-of-the-art predictors. We believe that the proposed NeuroPred-FRL can serve as a powerful tool for large-scale identification of NPs, facilitating the characterization of their functional mechanisms and expediting their applications in clinical therapy. Moreover, we interpreted some model mechanisms of NeuroPred-FRL by leveraging the robust SHapley Additive exPlanation algorithm.

Enhancers are deoxyribonucleic acid (DNA) fragments which when bound by transcription factors enhance the transcription of related genes. Due to its sporadic distribution and similar fractions, identification of enhancers from the human genome seems a daunting task. Compared to the traditional experimental approaches, computational methods with easy-to-use platforms could be efficiently applied to annotate enhancers' functions and physiological roles. In this aspect, several bioinformatics tools have been developed to identify enhancers. Despite their spectacular performances, existing methods have certain drawbacks and limitations, including fixed length of sequences being utilized for model development and cell-specificity negligence. A novel predictor would be beneficial in the context of genome-wide enhancer prediction by addressing the above-mentioned issues. In this study, we constructed new datasets for eight different cell types. Utilizing these data, we proposed an integrative machine learning (ML)-based framework called Enhancer-IF for identifying cell-specific enhancers. Enhancer-IF comprehensively explores a wide range of heterogeneous features with five commonly used ML methods (random forest, extremely randomized tree, multilayer perceptron, support vector machine and extreme gradient boosting). Specifically, these five classifiers were trained with seven encodings and obtained 35 baseline models. The output of these baseline models was integrated and again inputted to five classifiers for the construction of five meta-models. Finally, the integration of five meta-models through ensemble learning improved the model robustness. Our proposed approach showed an excellent prediction performance compared to the baseline models on both training and independent datasets in different cell types, thus highlighting the superiority of our approach in the identification of the enhancers. We assume that Enhancer-IF will be a valuable tool for screening and identifying potential enhancers from the human DNA sequences.

The release of interleukin (IL)-6 is stimulated by antigenic peptides from pathogens as well as by immune cells for activating aggressive inflammation. IL-6 inducing peptides are derived from pathogens and can be used as diagnostic biomarkers for predicting various stages of disease severity as well as being used as IL-6 inhibitors for the suppression of aggressive multi-signaling immune responses. Thus, the accurate identification of IL-6 inducing peptides is of great importance for investigating their mechanism of action as well as for developing diagnostic and immunotherapeutic applications. This study proposes a novel stacking ensemble model (termed StackIL6) for accurately identifying IL-6 inducing peptides. More specifically, StackIL6 was constructed from twelve different feature descriptors derived from three major groups of features (composition-based features, composition-transition-distribution-based features and physicochemical properties-based features) and five popular machine learning algorithms (extremely randomized trees, logistic regression, multi-layer perceptron, support vector machine and random forest). To enhance the utility of baseline models, they were effectively and systematically integrated through a stacking strategy to build the final meta-based model. Extensive benchmarking experiments demonstrated that StackIL6 could achieve significantly better performance than the existing method (IL6PRED) and outperformed its constituent baseline models on both training and independent test datasets, which thereby support its excellent discrimination and generalization abilities. To facilitate easy access to the StackIL6 model, it was established as a freely available web server accessible at http://camt.pythonanywhere.com/StackIL6. It is anticipated that StackIL6 can help to facilitate rapid screening of promising IL-6 inducing peptides for the development of diagnostic and immunotherapeutic applications in the future.

DNA N6-methyladenine (6mA) represents important epigenetic modifications, which are responsible for various cellular processes. The accurate identification of 6mA sites is one of the challenging tasks in genome analysis, which leads to an understanding of their biological functions. To date, several species-specific machine learning (ML)-based models have been proposed, but majority of them did not test their model to other species. Hence, their practical application to other plant species is quite limited. In this study, we explored 10 different feature encoding schemes, with the goal of capturing key characteristics around 6mA sites. We selected five feature encoding schemes based on physicochemical and position-specific information that possesses high discriminative capability. The resultant feature sets were inputted to six commonly used ML methods (random forest, support vector machine, extremely randomized tree, logistic regression, naïve Bayes and AdaBoost). The Rosaceae genome was employed to train the above classifiers, which generated 30 baseline models. To integrate their individual strength, Meta-i6mA was proposed that combined the baseline models using the meta-predictor approach. In extensive independent test, Meta-i6mA showed high Matthews correlation coefficient values of 0.918, 0.827 and 0.635 on Rosaceae, rice and Arabidopsis thaliana, respectively and outperformed the existing predictors. We anticipate that the Meta-i6mA can be applied across different plant species. Furthermore, we developed an online user-friendly web server, which is available at http://kurata14.bio.kyutech.ac.jp/Meta-i6mA/.

Deoxyribonucleic acid replication is one of the most crucial tasks taking place in the cell, and it has to be precisely regulated. This process is initiated in the replication origins (ORIs), and thus it is essential to identify such sites for a deeper understanding of the cellular processes and functions related to the regulation of gene expression. Considering the important tasks performed by ORIs, several experimental and computational approaches have been developed in the prediction of such sites. However, existing computational predictors for ORIs have certain curbs, such as building only single-feature encoding models, limited systematic feature engineering efforts and failure to validate model robustness. Hence, we developed a novel species-specific yeast predictor called yORIpred that accurately identify ORIs in the yeast genomes. To develop yORIpred, we first constructed optimal 40 baseline models by exploring eight different sequence-based encodings and five different machine learning classifiers. Subsequently, the predicted probability of 40 models was considered as the novel feature vector and carried out iterative feature learning approach independently using five different classifiers. Our systematic analysis revealed that the feature representation learned by the support vector machine algorithm (yORIpred) could well discriminate the distribution characteristics between ORIs and non-ORIs when compared with the other four algorithms. Comprehensive benchmarking experiments showed that yORIpred achieved superior and stable performance when compared with the existing predictors on the same training datasets. Furthermore, independent evaluation showcased the best and accurate performance of yORIpred thus underscoring the significance of iterative feature representation. To facilitate the users in obtaining their desired results without undergoing any mathematical, statistical or computational hassles, we developed a web server for the yORIpred predictor, which is available at: http://thegleelab.org/yORIpred.

Origins of replication sites (ORIs), which refers to the initiative locations of genomic DNA replication, play essential roles in DNA replication process. Detection of ORIs' distribution in genome scale is one of key steps to in-depth understanding their regulation mechanisms. In this study, we presented a novel machine learning-based approach called Stack-ORI encompassing 10 cell-specific prediction models for identifying ORIs from four different eukaryotic species (Homo sapiens, Mus musculus, Drosophila melanogaster and Arabidopsis thaliana). For each cell-specific model, we employed 12 feature encoding schemes that cover nucleic acid composition, position-specific and physicochemical properties information. The optimal feature set was identified from each encoding individually and developed their respective baseline models using the eXtreme Gradient Boosting (XGBoost) classifier. Subsequently, the predicted scores of 12 baseline models are integrated as a novel feature vector to train XGBoost and develop the final model. Extensive experimental results show that Stack-ORI achieves significantly better performance as compared with their baseline models on both training and independent datasets. Interestingly, Stack-ORI consistently outperforms existing predictor in all cell-specific models, not only on training but also on independent test. Moreover, our novel approach provides necessary interpretations that help understanding model success by leveraging the powerful SHapley Additive exPlanation algorithm, thus underlining the most important feature encoding schemes significant for predicting cell-specific ORIs.

MOTIVATION: The identification of bitter peptides through experimental approaches is an expensive and time-consuming endeavor. Due to the huge number of newly available peptide sequences in the post-genomic era, the development of automated computational models for the identification of novel bitter peptides is highly desirable. RESULTS: In this work, we present BERT4Bitter, a bidirectional encoder representation from transformers (BERT)-based model for predicting bitter peptides directly from their amino acid sequence without using any structural information. To the best of our knowledge, this is the first time a BERT-based model has been employed to identify bitter peptides. Compared to widely used machine learning models, BERT4Bitter achieved the best performance with an accuracy of 0.861 and 0.922 for cross-validation and independent tests, respectively. Furthermore, extensive empirical benchmarking experiments on the independent dataset demonstrated that BERT4Bitter clearly outperformed the existing method with improvements of 8.0% accuracy and 16.0% Matthews coefficient correlation, highlighting the effectiveness and robustness of BERT4Bitter. We believe that the BERT4Bitter method proposed herein will be a useful tool for rapidly screening and identifying novel bitter peptides for drug development and nutritional research. AVAILABILITYAND IMPLEMENTATION: The user-friendly web server of the proposed BERT4Bitter is freely accessible at http://pmlab.pythonanywhere.com/BERT4Bitter. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Cell-penetrating peptides (CPPs) facilitate the delivery of therapeutically relevant molecules, including DNA, proteins and oligonucleotides, into cells both in vitro and in vivo. This unique ability explores the possibility of CPPs as therapeutic delivery and its potential applications in clinical therapy. Over the last few decades, a number of machine learning (ML)-based prediction tools have been developed, and some of them are freely available as web portals. However, the predictions produced by various tools are difficult to quantify and compare. In particular, there is no systematic comparison of the web-based prediction tools in performance, especially in practical applications. In this work, we provide a comprehensive review on the biological importance of CPPs, CPP database and existing ML-based methods for CPP prediction. To evaluate current prediction tools, we conducted a comparative study and analyzed a total of 12 models from 6 publicly available CPP prediction tools on 2 benchmark validation sets of CPPs and non-CPPs. Our benchmarking results demonstrated that a model from the KELM-CPPpred, namely KELM-hybrid-AAC, showed a significant improvement in overall performance, when compared to the other 11 prediction models. Moreover, through a length-dependency analysis, we find that existing prediction tools tend to more accurately predict CPPs and non-CPPs with the length of 20-25 residues long than peptides in other length ranges.

MOTIVATION: Therapeutic peptides failing at clinical trials could be attributed to their toxicity profiles like hemolytic activity, which hamper further progress of peptides as drug candidates. The accurate prediction of hemolytic peptides (HLPs) and its activity from the given peptides is one of the challenging tasks in immunoinformatics, which is essential for drug development and basic research. Although there are a few computational methods that have been proposed for this aspect, none of them are able to identify HLPs and their activities simultaneously. RESULTS: In this study, we proposed a two-layer prediction framework, called HLPpred-Fuse, that can accurately and automatically predict both hemolytic peptides (HLPs or non-HLPs) as well as HLPs activity (high and low). More specifically, feature representation learning scheme was utilized to generate 54 probabilistic features by integrating six different machine learning classifiers and nine different sequence-based encodings. Consequently, the 54 probabilistic features were fused to provide sufficiently converged sequence information which was used as an input to extremely randomized tree for the development of two final prediction models which independently identify HLP and its activity. Performance comparisons over empirical cross-validation analysis, independent test and case study against state-of-the-art methods demonstrate that HLPpred-Fuse consistently outperformed these methods in the identification of hemolytic activity. AVAILABILITY AND IMPLEMENTATION: For the convenience of experimental scientists, a web-based tool has been established at http://thegleelab.org/HLPpred-Fuse. CONTACT: [email protected] or [email protected] or [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Cardiovascular disease is the primary cause of death globally accounting for approximately 17.7 million deaths per year. One of the stakes linked with cardiovascular diseases and other complications is hypertension. Naturally derived bioactive peptides with antihypertensive activities serve as promising alternatives to pharmaceutical drugs. So far, there is no comprehensive analysis, assessment of diverse features and implementation of various machine-learning (ML) algorithms applied for antihypertensive peptide (AHTP) model construction. RESULTS: In this study, we utilized six different ML algorithms, namely, Adaboost, extremely randomized tree (ERT), gradient boosting (GB), k-nearest neighbor, random forest (RF) and support vector machine (SVM) using 51 feature descriptors derived from eight different feature encodings for the prediction of AHTPs. While ERT-based trained models performed consistently better than other algorithms regardless of various feature descriptors, we treated them as baseline predictors, whose predicted probability of AHTPs was further used as input features separately for four different ML-algorithms (ERT, GB, RF and SVM) and developed their corresponding meta-predictors using a two-step feature selection protocol. Subsequently, the integration of four meta-predictors through an ensemble learning approach improved the balanced prediction performance and model robustness on the independent dataset. Upon comparison with existing methods, mAHTPred showed superior performance with an overall improvement of approximately 6-7% in both benchmarking and independent datasets. AVAILABILITY AND IMPLEMENTATION: The user-friendly online prediction tool, mAHTPred is freely accessible at http://thegleelab.org/mAHTPred. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Accurate identification of N4-methylcytosine (4mC) modifications in a genome wide can provide insights into their biological functions and mechanisms. Machine learning recently have become effective approaches for computational identification of 4mC sites in genome. Unfortunately, existing methods cannot achieve satisfactory performance, owing to the lack of effective DNA feature representations that are capable to capture the characteristics of 4mC modifications. RESULTS: In this work, we developed a new predictor named 4mcPred-IFL, aiming to identify 4mC sites. To represent and capture discriminative features, we proposed an iterative feature representation algorithm that enables to learn informative features from several sequential models in a supervised iterative mode. Our analysis results showed that the feature representations learnt by our algorithm can capture the discriminative distribution characteristics between 4mC sites and non-4mC sites, enlarging the decision margin between the positives and negatives in feature space. Additionally, by evaluating and comparing our predictor with the state-of-the-art predictors on benchmark datasets, we demonstrate that our predictor can identify 4mC sites more accurately. AVAILABILITY AND IMPLEMENTATION: The user-friendly webserver that implements the proposed 4mcPred-IFL is well established, and is freely accessible at http://server.malab.cn/4mcPred-IFL. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The accurate ranking of predicted structural models and selecting the best model from a given candidate pool remain as open problems in the field of structural bioinformatics. The quality assessment (QA) methods used to address these problems can be grouped into two categories: consensus methods and single-model methods. Consensus methods in general perform better and attain higher correlation between predicted and true quality measures. However, these methods frequently fail to generate proper quality scores for native-like structures which are distinct from the rest of the pool. Conversely, single-model methods do not suffer from this drawback and are better suited for real-life applications where many models from various sources may not be readily available. RESULTS: In this study, we developed a support-vector-machine-based single-model global quality assessment (SVMQA) method. For a given protein model, the SVMQA method predicts TM-score and GDT_TS score based on a feature vector containing statistical potential energy terms and consistency-based terms between the actual structural features (extracted from the three-dimensional coordinates) and predicted values (from primary sequence). We trained SVMQA using CASP8, CASP9 and CASP10 targets and determined the machine parameters by 10-fold cross-validation. We evaluated the performance of our SVMQA method on various benchmarking datasets. Results show that SVMQA outperformed the existing best single-model QA methods both in ranking provided protein models and in selecting the best model from the pool. According to the CASP12 assessment, SVMQA was the best method in selecting good-quality models from decoys in terms of GDTloss. AVAILABILITY AND IMPLEMENTATION: SVMQA method can be freely downloaded from http://lee.kias.re.kr/SVMQA/SVMQA_eval.tar.gz. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Protein folding rate is one of the important properties of a protein. Protein folding rate prediction is useful for understanding protein folding process and guiding protein design. In this study, we developed a support vector machine (SVM) based method to predict protein folding kinetic types (two-state or non-two-state) and the real-value folding rate using the features calculated from the three-dimensional structure such as contact order, various properties from the non-local contact clusters, secondary structural information and sequence length. We systematically studied the contributions of individual features to folding rate prediction. Based on the highest contributions of individual features, we trained our machine using leave one out cross-validation and tested on a testing dataset. The Pearson correlation coefficient, mean absolute difference and root mean square error between the predicted and experimental folding rates (base-10 logarithmic scale) are 0.814, 0.752 and 0.910 for two-state proteins, and 0.860, 0.687 and 0.876 for non-two-state proteins. Moreover, our method predicts whether a protein of known atomic structure folds according to two-state or non-two-state kinetics and correctly classifies 80% of the folding mechanism on a testing dataset. Finally, we evaluated the performance of our method along with the other eight existing protein folding rate prediction tools on non-overlapping benchmarking dataset. The prediction performance will also be reported and discussed.