HLA-II immunopeptidome profiling and deep learning reveal features of antigenicity to inform antigen discovery.

CD4+ T cell responses are exquisitely antigen specific and directed toward peptide epitopes displayed by human leukocyte antigen class II (HLA-II) on antigen-presenting cells. Underrepresentation of diverse alleles in ligand databases and an incomplete understanding of factors affecting antigen presentation in vivo have limited progress in defining principles of peptide immunogenicity. Here, we employed monoallelic immunopeptidomics to identify 358,024 HLA-II binders, with a particular focus on HLA-DQ and HLA-DP. We uncovered peptide-binding patterns across a spectrum of binding affinities and enrichment of structural antigen features. These aspects underpinned the development of context-aware predictor of T cell antigens (CAPTAn), a deep learning model that predicts peptide antigens based on their affinity to HLA-II and full sequence of their source proteins. CAPTAn was instrumental in discovering prevalent T cell epitopes from bacteria in the human microbiome and a pan-variant epitope from SARS-CoV-2. Together CAPTAn and associated datasets present a resource for antigen discovery and the unraveling genetic associations of HLA alleles with immunopathologies.

AFMC: An alignment framework for multiple computing services and providers

The Hirschberg algorithm is commonly used for protein sequence alignment, which is a very important task in bioinformatics. This article presents the AFMC framework for using the Hirschberg method to perform sequence alignment in multiple cloud computing services of different models, such as Infrastructure-as-a-Service and Function-as-a-Service (FaaS). Experiments were carried out in which several instances of AWS EC2, Azure VMs and Google Compute Engine as well as varied configurations of AWS Lambda, Azure Function, and Google Cloud Function were used to pairwise align COVID-19 spike proteins. The services were submitted to different levels of simultaneity to align the genetic sequences. The findings reveal that there is a tradeoff between predicted execution time and cost for this application, for example, FaaS-oriented cloud service models generally took less time to process the workloads. On the other hand, it was observed that, as the level of concurrence increased, there was a marked augmentation in cost. In this context, a framework that provides multi cloud solutions for bioinformatics such as AFMC is essential. © 2023 John Wiley & Sons, Ltd.

Hamlet-Pattern-Based Automated COVID-19 and Influenza Detection Model Using Protein Sequences.

SARS-CoV-2 and Influenza-A can present similar symptoms. Computer-aided diagnosis can help facilitate screening for the two conditions, and may be especially relevant and useful in the current COVID-19 pandemic because seasonal Influenza-A infection can still occur. We have developed a novel text-based classification model for discriminating between the two conditions using protein sequences of varying lengths. We downloaded viral protein sequences of SARS-CoV-2 and Influenza-A with varying lengths (all 100 or greater) from the NCBI database and randomly selected 16,901 SARS-CoV-2 and 19,523 Influenza-A sequences to form a two-class study dataset. We used a new feature extraction function based on a unique pattern, HamletPat, generated from the text of Shakespeare's Hamlet, and a signum function to extract local binary pattern-like bits from overlapping fixed-length (27) blocks of the protein sequences. The bits were converted to decimal map signals from which histograms were extracted and concatenated to form a final feature vector of length 1280. The iterative Chi-square function selected the 340 most discriminative features to feed to an SVM with a Gaussian kernel for classification. The model attained 99.92% and 99.87% classification accuracy rates using hold-out (75:25 split ratio) and five-fold cross-validations, respectively. The excellent performance of the lightweight, handcrafted HamletPat-based classification model suggests that it can be a valuable tool for screening protein sequences to discriminate between SARS-CoV-2 and Influenza-A infections.

ML-DTD: Machine Learning-Based Drug Target Discovery for the Potential Treatment of COVID-19.

Recent research has highlighted that a large section of druggable protein targets in the Human interactome remains unexplored for various diseases. It might lead to the drug repurposing study and help in the in-silico prediction of new drug-human protein target interactions. The same applies to the current pandemic of COVID-19 disease in global health issues. It is highly desirable to identify potential human drug targets for COVID-19 using a machine learning approach since it saves time and labor compared to traditional experimental methods. Structure-based drug discovery where druggability is determined by molecular docking is only appropriate for the protein whose three-dimensional structures are available. With machine learning algorithms, differentiating relevant features for predicting targets and non-targets can be used for the proteins whose 3-D structures are unavailable. In this research, a Machine Learning-based Drug Target Discovery (ML-DTD) approach is proposed where a machine learning model is initially built up and tested on the curated dataset consisting of COVID-19 human drug targets and non-targets formed by using the Therapeutic Target Database (TTD) and human interactome using several classifiers like XGBBoost Classifier, AdaBoost Classifier, Logistic Regression, Support Vector Classification, Decision Tree Classifier, Random Forest Classifier, Naive Bayes Classifier, and K-Nearest Neighbour Classifier (KNN). In this method, protein features include Gene Set Enrichment Analysis (GSEA) ranking, properties derived from the protein sequence, and encoded protein network centrality-based measures. Among all these, XGBBoost, KNN, and Random Forest models are satisfactory and consistent. This model is further used to predict novel COVID-19 human drug targets, which are further validated by target pathway analysis, the emergence of allied repurposed drugs, and their subsequent docking study.

SARS-Arena: Sequence and Structure-Guided Selection of Conserved Peptides from SARS-related Coronaviruses for Novel Vaccine Development.

The pandemic caused by the SARS-CoV-2 virus, the agent responsible for the COVID-19 disease, has affected millions of people worldwide. There is constant search for new therapies to either prevent or mitigate the disease. Fortunately, we have observed the successful development of multiple vaccines. Most of them are focused on one viral envelope protein, the spike protein. However, such focused approaches may contribute for the rise of new variants, fueled by the constant selection pressure on envelope proteins, and the widespread dispersion of coronaviruses in nature. Therefore, it is important to examine other proteins, preferentially those that are less susceptible to selection pressure, such as the nucleocapsid (N) protein. Even though the N protein is less accessible to humoral response, peptides from its conserved regions can be presented by class I Human Leukocyte Antigen (HLA) molecules, eliciting an immune response mediated by T-cells. Given the increased number of protein sequences deposited in biological databases daily and the N protein conservation among viral strains, computational methods can be leveraged to discover potential new targets for SARS-CoV-2 and SARS-CoV-related viruses. Here we developed SARS-Arena, a user-friendly computational pipeline that can be used by practitioners of different levels of expertise for novel vaccine development. SARS-Arena combines sequence-based methods and structure-based analyses to (i) perform multiple sequence alignment (MSA) of SARS-CoV-related N protein sequences, (ii) recover candidate peptides of different lengths from conserved protein regions, and (iii) model the 3D structure of the conserved peptides in the context of different HLAs. We present two main Jupyter Notebook workflows that can help in the identification of new T-cell targets against SARS-CoV viruses. In fact, in a cross-reactive case study, our workflows identified a conserved N protein peptide (SPRWYFYYL) recognized by CD8+ T-cells in the context of HLA-B7+. SARS-Arena is available at https://github.com/KavrakiLab/SARS-Arena.

Accurate prediction of virus-host protein-protein interactions via a Siamese neural network using deep protein sequence embeddings.

Prediction and understanding of virus-host protein-protein interactions (PPIs) have relevance for the development of novel therapeutic interventions. In addition, virus-like particles open novel opportunities to deliver therapeutics to targeted cell types and tissues. Given our incomplete knowledge of PPIs on the one hand and the cost and time associated with experimental procedures on the other, we here propose a deep learning approach to predict virus-host PPIs. Our method (Siamese Tailored deep sequence Embedding of Proteins [STEP]) is based on recent deep protein sequence embedding techniques, which we integrate into a Siamese neural network. After showing the state-of-the-art performance of STEP on external datasets, we apply it to two use cases, severe acute respiratory syndrome coronavirus 2 and John Cunningham polyomavirus, to predict virus-host PPIs. Altogether our work highlights the potential of deep sequence embedding techniques originating from the field of NLP as well as explainable artificial intelligence methods for the analysis of biological sequences.

NGS data vectorization, clustering, and finding key codons in SARS-CoV-2 variations.

The rapid global spread and dissemination of SARS-CoV-2 has provided the virus with numerous opportunities to develop several variants. Thus, it is critical to determine the degree of the variations and in which part of the virus those variations occurred. Therefore, in this study, methods that could be used to vectorize the sequence data, perform clustering analysis, and visualize the results were proposed using machine learning methods. To conduct this study, a total of 224,073 cases of SARS-CoV-2 sequence data were collected through NCBI and GISAID, and the data were visualized using dimensionality reduction and clustering analysis models such as T-SNE and DBSCAN. The SARS-CoV-2 virus, which was first detected, was distinguished from different variations, including Omicron and Delta, in the cluster results. Furthermore, it was possible to examine which codon changes in the spike protein caused the variants to be distinguished using feature importance extraction models such as Random Forest or Shapely Value. The proposed method has the advantage of being able to analyse and visualize a large amount of data at once compared to the existing tree-based sequence data analysis. The proposed method was able to identify and visualize significant changes between the SARS-CoV-2 virus, which was first detected in Wuhan, China, in December 2019, and the newly formed mutant virus group. As a result of clustering analysis using sequence data, it was possible to confirm the formation of clusters among various variants in a two-dimensional graph, and by extracting the importance of variables, it was possible to confirm which codon changes played a major role in distinguishing variants. Furthermore, since the proposed method can handle a variety of data sequences, it can be used for all kinds of diseases, including influenza and SARS-CoV-2. Therefore, the proposed method has the potential to become widely used for the effective analysis of disease variations.

A Novel Scalable Feature Extraction Approach for COVID-19 Protein Sequences and their Cluster Analysis with Kernelized Fuzzy Algorithm

COVID-19 (Coronavirus Disease-19), a disease caused by the SARS-CoV-2 virus, was declared a pandemic by the World Health Organization on March 11, 2020. To solve the global problem of analysis of different variants of COVID-19 genome sequences, there is a need to develop intel-ligent, scalable machine learning techniques that can process and analyze important COVID-19 protein data by utilizing the Big Data framework. For this, we have first proposed a feature extraction approach for COVID-19 protein data named Scalable Distributed Co-occurrence-based Probability-Specific Feature extraction approach (SDCPSF). The proposed SDCPSF approach is executed on the Apache Spark cluster to preprocess the massive COVID-19 protein sequences. The proposed SDCPSF represents each variable-length COVID-19 protein sequence with fixed length six dimensions numeric feature vectors. Then the extracted features are used as input to the kernelized fuzzy clustering algorithms, i.e., KSRSIO-FCM and KSLFCM, which efficiently performs clustering of big data due to its in-memory cluster computing technique and thus forms clusters of COVID-19 genome sequences. Furthermore, the performance of KSRSIO-FCM is compared with another scalable clustering algorithm, i.e., KSLFCM, in terms of the Silhouette index (SI) and Davies-Bouldin index (DBI). © 2022 IEEE.

Classifying COVID-19 based on amino acids encoding with machine learning algorithms.

COVID-19 disease causes serious respiratory illnesses. Therefore, accurate identification of the viral infection cycle plays a key role in designing appropriate vaccines. The risk of this disease depends on proteins that interact with human receptors. In this paper, we formulate a novel model for COVID-19 named "amino acid encoding based prediction" (AAPred). This model is accurate, classifies the various coronavirus types, and distinguishes SARS-CoV-2 from other coronaviruses. With the AAPred model, we reduce the number of features to enhance its performance by selecting the most important ones employing statistical criteria. The protein sequence of SARS-CoV-2 for understanding the viral infection cycle is analyzed. Six machine learning classifiers related to decision trees, k-nearest neighbors, random forest, support vector machine, bagging ensemble, and gradient boosting are used to evaluate the model in terms of accuracy, precision, sensitivity, and specificity. We implement the obtained results computationally and apply them to real data from the National Genomics Data Center. The experimental results report that the AAPred model reduces the features to seven of them. The average accuracy of the 10-fold cross-validation is 98.69%, precision is 98.72%, sensitivity is 96.81%, and specificity is 97.72%. The features are selected utilizing information gain and classified with random forest. The proposed model predicts the type of Coronavirus and reduces the number of extracted features. We identify that SARS-CoV-2 has similar physicochemical characteristics in some regions of SARS-CoV. Also, we report that SARS-CoV-2 has similar infection cycles and sequences in some regions of SARS CoV indicating the affectedness of vaccines on SARS-CoV-2. A comparison with deep learning shows similar results with our method.

Underlying selection for the diversity of spike protein sequences of SARS-CoV-2.

The global spread of SARS-CoV-2 is fast moving and has caused a worldwide public health crisis. In the present article, we analyzed spike protein sequences of SARS-CoV-2 genomes to assess the impact of mutational diversity. We observed from amino acid usage patterns that spike proteins are associated with a diversity of mutational changes and most important underlying cause of variation of amino acid usage is the changes in hydrophobicity of spike proteins. The changing patterns of hydrophobicity of spike proteins over time and its influence on the receptor binding affinity provides crucial information on the SARS-CoV-2 interaction with human receptor. Our results also show that spike proteins have evolved to prefer more hydrophobic residues over time. The present study provides a comprehensive analysis of molecular sequence data to consider that mutational variants might play a crucial role in modulating the virulence and spread of the virus and has immediate implications for therapeutic strategies.

AC2: An Efficient Protein Sequence Compression Tool Using Artificial Neural Networks and Cache-Hash Models.

Recently, the scientific community has witnessed a substantial increase in the generation of protein sequence data, triggering emergent challenges of increasing importance, namely efficient storage and improved data analysis. For both applications, data compression is a straightforward solution. However, in the literature, the number of specific protein sequence compressors is relatively low. Moreover, these specialized compressors marginally improve the compression ratio over the best general-purpose compressors. In this paper, we present AC2, a new lossless data compressor for protein (or amino acid) sequences. AC2 uses a neural network to mix experts with a stacked generalization approach and individual cache-hash memory models to the highest-context orders. Compared to the previous compressor (AC), we show gains of 2-9% and 6-7% in reference-free and reference-based modes, respectively. These gains come at the cost of three times slower computations. AC2 also improves memory usage against AC, with requirements about seven times lower, without being affected by the sequences' input size. As an analysis application, we use AC2 to measure the similarity between each SARS-CoV-2 protein sequence with each viral protein sequence from the whole UniProt database. The results consistently show higher similarity to the pangolin coronavirus, followed by the bat and human coronaviruses, contributing with critical results to a current controversial subject. AC2 is available for free download under GPLv3 license.

Semicovariance Coefficient Analysis of Spike Proteins from SARS-CoV-2 and Other Coronaviruses for Viral Evolution and Characteristics Associated with Fatality.

Complex modeling has received significant attention in recent years and is increasingly used to explain statistical phenomena with increasing and decreasing fluctuations, such as the similarity or difference of spike protein charge patterns of coronaviruses. Different from the existing covariance or correlation coefficient methods in traditional integer dimension construction, this study proposes a simplified novel fractional dimension derivation with the exact Excel tool algorithm. It involves the fractional center moment extension to covariance, which results in a complex covariance coefficient that is better than the Pearson correlation coefficient, in the sense that the nonlinearity relationship can be further depicted. The spike protein sequences of coronaviruses were obtained from the GenBank and GISAID databases, including the coronaviruses from pangolin, bat, canine, swine (three variants), feline, tiger, SARS-CoV-1, MERS, and SARS-CoV-2 (including the strains from Wuhan, Beijing, New York, German, and the UK variant B.1.1.7) which were used as the representative examples in this study. By examining the values above and below the average/mean based on the positive and negative charge patterns of the amino acid residues of the spike proteins from coronaviruses, the proposed algorithm provides deep insights into the nonlinear evolving trends of spike proteins for understanding the viral evolution and identifying the protein characteristics associated with viral fatality. The calculation results demonstrate that the complex covariance coefficient analyzed by this algorithm is capable of distinguishing the subtle nonlinear differences in the spike protein charge patterns with reference to Wuhan strain SARS-CoV-2, which the Pearson correlation coefficient may overlook. Our analysis reveals the unique convergent (positive correlative) to divergent (negative correlative) domain center positions of each virus. The convergent or conserved region may be critical to the viral stability or viability; while the divergent region is highly variable between coronaviruses, suggesting high frequency of mutations in this region. The analyses show that the conserved center region of SARS-CoV-1 spike protein is located at amino acid residues 900, but shifted to the amino acid residues 700 in MERS spike protein, and then to amino acid residues 600 in SARS-COV-2 spike protein, indicating the evolution of the coronaviruses. Interestingly, the conserved center region of the spike protein in SARS-COV-2 variant B.1.1.7 shifted back to amino acid residues 700, suggesting this variant is more virulent than the original SARS-COV-2 strain. Another important characteristic our study reveals is that the distance between the divergent mean and the maximal divergent point in each of the viruses (MERS > SARS-CoV-1 > SARS-CoV-2) is proportional to viral fatality rate. This algorithm may help to understand and analyze the evolving trends and critical characteristics of SARS-COV-2 variants, other coronaviral proteins and viruses.

PreDTIs: prediction of drug-target interactions based on multiple feature information using gradient boosting framework with data balancing and feature selection techniques.

Discovering drug-target (protein) interactions (DTIs) is of great significance for researching and developing novel drugs, having a tremendous advantage to pharmaceutical industries and patients. However, the prediction of DTIs using wet-lab experimental methods is generally expensive and time-consuming. Therefore, different machine learning-based methods have been developed for this purpose, but there are still substantial unknown interactions needed to discover. Furthermore, data imbalance and feature dimensionality problems are a critical challenge in drug-target datasets, which can decrease the classifier performances that have not been significantly addressed yet. This paper proposed a novel drug-target interaction prediction method called PreDTIs. First, the feature vectors of the protein sequence are extracted by the pseudo-position-specific scoring matrix (PsePSSM), dipeptide composition (DC) and pseudo amino acid composition (PseAAC); and the drug is encoded with MACCS substructure fingerings. Besides, we propose a FastUS algorithm to handle the class imbalance problem and also develop a MoIFS algorithm to remove the irrelevant and redundant features for getting the best optimal features. Finally, balanced and optimal features are provided to the LightGBM Classifier to identify DTIs, and the 5-fold CV validation test method was applied to evaluate the prediction ability of the proposed method. Prediction results indicate that the proposed model PreDTIs is significantly superior to other existing methods in predicting DTIs, and our model could be used to discover new drugs for unknown disorders or infections, such as for the coronavirus disease 2019 using existing drugs compounds and severe acute respiratory syndrome coronavirus 2 protein sequences.

Deep Learning-Based Potential Ligand Prediction Framework for COVID-19 with Drug-Target Interaction Model.

To fight against the present pandemic scenario of COVID-19 outbreak, medication with drugs and vaccines is extremely essential other than ventilation support. In this paper, we present a list of ligands which are expected to have the highest binding affinity with the S-glycoprotein of 2019-nCoV and thus can be used to make the drug for the novel coronavirus. Here, we implemented an architecture using 1D convolutional networks to predict drug-target interaction (DTI) values. The network was trained on the KIBA (Kinase Inhibitor Bioactivity) dataset. With this network, we predicted the KIBA scores (which gives a measure of binding affinity) of a list of ligands against the S-glycoprotein of 2019-nCoV. Based on these KIBA scores, we are proposing a list of ligands (33 top ligands based on best interactions) which have a high binding affinity with the S-glycoprotein of 2019-nCoV and thus can be used for the formation of drugs.