Drug Repositioning for Coronavirus (COVID-19) using Different Deep Learning Architectures

In December 2019, the COVID-19 epidemic was found in Wuhan, China, and soon hundreds of millions were infected. Therefore, several efforts were made to identify commercially available drugs to repurpose them against COVID-19. Inferring potential drug indications through computational drug repositioning is an efficient method. The drug repositioning problem is a top-K recommendation function that presents the most likely drugs for specific diseases based on drug and disease-related data. The accurate prediction of drug-target interactions (DTI) is very important for drug repositioning. Deep learning (DL) models were recently exploited for promising DTI prediction performance. To build deep learning models for DTI prediction, encoder-decoder architectures can be utilized. In this paper, a deep learning-based drug repositioning approach is proposed, which is composed of two experimental phases. Firstly, training and evaluating different deep learning encoder-decoder architecture models using the benchmark DAVIS Dataset. The trained deep learning models have been evaluated using two evaluation metrics;mean square error and the concordance index. Secondly, predicting antiviral drugs for Covid-19 using the trained deep learning models created during the first phase. In this phase, these models have been experimented to predict different antiviral drug lists, which then have been compared with a recently published antiviral drug list for Covid-19 using the concordance index metric. The overall experimental results of both phases showed that the most accurate three deep learning compound-encoder/protein-encoder architectures are Morgan/AAC, CNN/AAC, and CNN/CNN with best values for the mean square error, the first phase concordance index, and the second phase concordance index. © 2023,International Journal of Advanced Computer Science and Applications. All Rights Reserved.

CNN-OSBO Encoder-Decoder Architecture for Drug-Target Interaction (DTI) Prediction of Covid-19 Targets

Drug Target Interaction (DTI) prediction is an important factor is drug discovery and repositioning (DDR) since it detects the response of a drug over a target protein. The Coronavirus disease 2019 (COVID-19) disease created groups of deadly pneumonia with clinical appearance mostly similar to SARS-CoV. The precise diagnosis of COVID-19 clinical outcome is more challenging, since the diseases has various forms with varying structures. So predicting the interactions between various drugs with the SARS-CoV target protein is very crucial need in these days, which may leads to discovery of new drugs for the deadly disease. Recently, Deep learning (DL) techniques have been applied by the researches for DTI prediction. Since CNN is one of the major DL models which has the ability to create predictive feature vectors or embeddings, CNN-OSBO encoder-decoder architecture for DTI prediction of Covid-19 targets has been designed Given the input drug and Covid-19 target pair of data, they are fed into the Convolution Neural Networks (CNN) with Opposition based Satin Bowerbird Optimizer (OSBO) encoder modules, separately. Here OSBO is utilized for regulating the hyper parameters (HPs) of CNN layers. Both the encoded data are then embedded to create a binding module. Finally the CNN Decoder module predicts the interaction of drugs over the Covid-19 targets by returning an affinity or interaction score. Experimental results state that DTI prediction using CNN+OSBO achieves better accuracy results when compared with the existing techniques. © 2022 IEEE.

Towards De Novo Drug Design for the Coronavirus: A Drug-Target Interaction Prediction Approach using Atom-enhanced Graph Neural Network with Multi-hop Gating Mechanism

For humans, the COVID-19 pandemic and Coronavirus have undeniably been a nightmare. Although there are effective vaccines, specific drugs are still urgent. Normally, to identify potential drugs, one needs to design and then test interactions between the drug and the virus in an in silico manner for determining candidates. This Drug-Target Interaction (DTI) process, can be done by molecular docking, which is too complicated and time-consuming for manual works. Therefore, it opens room for applying Artificial Intelligence (AI) techniques. In particular, Graph Neural Network (GNN) attracts recent attention since its high suitability for the nature of drug compounds and virus proteins. However, to introduce such a representation well-reflecting biological structures of biological compounds is not a trivial task. Moreover, since available datasets of Coronavirus are still not highly popular, the recently developed GNNs have been suffering from overfitting on them. We then address those issues by proposing a novel model known as Atom-enhanced Graph Neural Network with Multi-hop Gating Mechanism. On one hand, our model can learn more precise features of compounds and proteins. On the other hand, we introduce a new gating mechanism to create better atom representation from non-neighbor information. Once applying transfer learning from very large databanks, our model enjoys promising performance, especially when experimenting with Coronavirus. © 2022 IEEE.

Forces Driving a Magic Bullet to Its Target: Revisiting the Role of Thermodynamics in Drug Design, Development, and Optimization.

Drug discovery strategies have advanced significantly towards prioritizing target selectivity to achieve the longstanding goal of identifying "magic bullets" amongst thousands of chemical molecules screened for therapeutic efficacy. A myriad of emerging and existing health threats, including the SARS-CoV-2 pandemic, alarming increase in bacterial resistance, and potentially fatal chronic ailments, such as cancer, cardiovascular disease, and neurodegeneration, have incentivized the discovery of novel therapeutics in treatment regimens. The design, development, and optimization of lead compounds represent an arduous and time-consuming process that necessitates the assessment of specific criteria and metrics derived via multidisciplinary approaches incorporating functional, structural, and energetic properties. The present review focuses on specific methodologies and technologies aimed at advancing drug development with particular emphasis on the role of thermodynamics in elucidating the underlying forces governing ligand-target interaction selectivity and specificity. In the pursuit of novel therapeutics, isothermal titration calorimetry (ITC) has been utilized extensively over the past two decades to bolster drug discovery efforts, yielding information-rich thermodynamic binding signatures. A wealth of studies recognizes the need for mining thermodynamic databases to critically examine and evaluate prospective drug candidates on the basis of available metrics. The ultimate power and utility of thermodynamics within drug discovery strategies reside in the characterization and comparison of intrinsic binding signatures that facilitate the elucidation of structural-energetic correlations which assist in lead compound identification and optimization to improve overall therapeutic efficacy.

Potential SARS-CoV-2 nonstructural proteins inhibitors: drugs repurposing with drug-target networks and deep learning.

BACKGROUND: In the current COVID-19 pandemic, with an absence of approved drugs and widely accessible vaccines, repurposing existing drugs is vital to quickly developing a treatment for the disease. METHODS: In this study, we used a dataset consisting of sequences of viral proteins and chemical structures of pharmaceutical drugs for known drug-target interactions (DTIs) and artificially generated non-interacting DTIs to train a binary classifier with the ability to predict new DTIs. Random Forest (RF), deep neural network (DNN), and convolutional neural networks (CNN) were tested. The CNN and RF models were selected for the classification task. RESULTS: The models generalized well to the given DTI data and were used to predict DTIs involving SARS-CoV-2 nonstructural proteins (NSPs). We elucidated (with the CNN) 29 drugs involved in 82 DTIs with a 97% probability of interaction, 44 DTIs of which had a 99% probability of interaction, to treat COVID-19. The RF elucidated 6 drugs involved in 17 DTIs with a 90% probability of interacting. CONCLUSIONS: These results give new insight into possible inhibitors of the viral proteins beyond pharmacophore models and molecular docking procedures used in recent studies.

Combining biomedical knowledge graphs and text to improve predictions for drug-target interactions and drug-indications.

Biomedical knowledge is represented in structured databases and published in biomedical literature, and different computational approaches have been developed to exploit each type of information in predictive models. However, the information in structured databases and literature is often complementary. We developed a machine learning method that combines information from literature and databases to predict drug targets and indications. To effectively utilize information in published literature, we integrate knowledge graphs and published literature using named entity recognition and normalization before applying a machine learning model that utilizes the combination of graph and literature. We then use supervised machine learning to show the effects of combining features from biomedical knowledge and published literature on the prediction of drug targets and drug indications. We demonstrate that our approach using datasets for drug-target interactions and drug indications is scalable to large graphs and can be used to improve the ranking of targets and indications by exploiting features from either structure or unstructured information alone.

Discovering missing edges in drug-protein networks: Repurposing drugs for SARS-CoV-2

The COVID-19 pandemic, caused by the SARS-CoV-2 virus, led to a global health crisis, with more than 157 million cases confirmed infected by May 2021. Effective medication is desperately needed. Predicting drug-target interaction (DTI) is an important step to discover novel uses of chemical structures. Here, we develop a pipeline to predict novel DTIs based on the proteins of the coronavirus. Different datasets (human/SARSCoV-2 Protein-Protein interaction (PPI), Drug-Drug similarity (DD sim), and DTIs) are used and combined. After mapping all datasets onto a heterogeneous graph, path-related features are extracted. We then applied various machine learning (ML) algorithms to model our dataset and predict novel DTIs among unlabeled pairs. Possible drugs identified by the models with a high frequency are reported. In addition, evidence of the efficiency of the predicted medicines by the models against COVID-19 are presented. The proposed model can then be generalized to contain other features that provide a context to predict medicine for different diseases. © 2021 IEEE.

Prediction of Drug-Target Interactions Using Molecular Graph and GDNet-DTI Model

The prediction of drug-target interactions (DTIs) is of great significance to the fields of drug design and drug development. However, traditional biological experiments are time-consuming and cost-effective, which has prompted more people to turn their attention to the use of computers to assist in predicting DTIs. This paper proposes an improved prediction model based on multiple graph representation methods, which is GDNet-DTI that combined GCN and DeepWalk. First, a molecular map with atoms as nodes and chemical bonds as edges is generated using the SMILE sequence of drugs, and then GIN is used to extract the features of molecular map for better obtaining the complex interactions between atoms. For target proteins, the protein sequence is first represented by a word vector, and then the one-dimensional convolution is used to extract features for extracting the different levels of features. Then, based on obtained drug features and target features, a DTI-graph is generated, in which drugs and targets are represented as nodes and interactions are represented as edges. Finally, GDNet-DTI are used to obtain node neighborhood information and graph topology information of the DTI-graph. Compared with other advanced models, the results show that GDNet-DTI combined with multiple graph features can predict DTIs more accurately and effectively with DrugBank and four benchmark datasets. In addition, a case study with COVID-19 data is presented, which shows that the proposed method has the potential to predict the actual DTIs and can contribute to the development of drug discovery. © 2021 IEEE.

Leveraging Drug-Target Interaction Data for the Translation of Computational Models into Clinically Actionable Interventions

Computational modeling is an effective tool for studying complex disease. However, solutions to many models are purely mathematical and cannot immediately provide clinical insights. To overcome this barrier, we propose a series of quantitative scoring metrics that can be used in combination with drug-target interaction data to identify solutions that are readily clinically actionable. Furthermore, we introduce methods for the prediction and ranking of pharmaceutical interventions that closely align with these high-scoring solutions, with an emphasis on robustness across multiple solutions. We demonstrate these methods on a previously-described model of COVID-19 induced cytokine storm. These scoring methods ultimately identify multiple pharmaceutical candidates that have been shown to be effective in reducing mortality rates in COVID-19 patients. © 2021 IEEE.

Discovering DTI and DDI by Knowledge Graph with MHRW and Improved Neural Network

Drug discovery is of great significance in medical and biological research, while the study of Drug-Target Interaction (DTI) and Drug-Drug Interaction (DDI) can help accelerate drug discovery progress. This paper proposes a new hybrid method for DTI prediction and DDI prediction, which is called MHRW2Vec-TBAN that combines graph representation learning and neural network. MHRW2VecTBAN first constructs knowledge graph KG-DTI and KG-DDI that integrate data related to drugs and targets. Then, an improved graph representation learning model, MHRW2Vec model, is used to obtain feature vectors of reflecting the network structure information for improving the performance of representation learning. Finally, the feature vectors obtained are input to the improved neural network model TextCNN-BiLSTM-Attention Network (TBAN). The experimental results show that, compared with other existing methods, our method could discover more deeper the relationship between drugs and their potential neighborhoods, and has great improvements in DTI prediction and DDI prediction. In addition, the case study of prediction COVID-19 DTI also shows that the proposed model has the potential for actual drug discovery. © 2021 IEEE.

In Silico Prediction of Indonesian Herbs Compounds as Covid-19 Supportive Therapy using Support Vector Machine

Many kinds of research on drug discovery using computational or in silico methods have been carried out. In this era of the Covid-19 pandemic, this research was also carried out by utilizing a commonly used technique, namely using machine learning to predict the interaction of compounds and proteins. This technique is known as Drug Target Interaction (DTI). The compounds used are herbal originating from Indonesia, and the protein used is a potential Covid-19 protein, one of which is SARS-CoV-2. The prediction process with machine learning can only be done on structured data. The data on herbal and protein were processed in this research using the Fingerprint as a descriptor compound and Pseudo Amino Acid Composition (PseAAC) as a protein descriptor technique. The result is structured data processed with the Support Vector Machine algorithm to create an interaction prediction model. The result is that the prediction accuracy is 95.96%. Furthermore, this model can predict Indonesian herbal compounds as drug candidates for Covid-19 supportive therapy. © 2021 IEEE.

Incorporating FPConv-DTI deep learning network and borderline-SMOTE algorithm for predicting drug-target interactions

Drug-target interactions prediction is of great significance in medical and biological research, but traditional laboratory methods have disadvantages such as high cost and time-consuming. Therefore, in recent years, deep learning, similarity calculation methods and other methods are becoming more and more widely applied to related research. This paper proposes an improved deep learning model, named as FPConv-DTI, which uses the fingerprint information of drug and the evolution feature information of protein based on a convolutional neural network. The Borderline-SMOTE algorithm is also used to generate new positive examples for training to solve the imbalance problem, and combines the number of sample data to process the input differently. Experiments have been carried out with four standard datasets and Drugbank dataset. The results show that compared with other methods, our method has greatly improvement for predicting drug-target interactions. In addition, some COVID-19 drugs are also predicted with the best-performing model, which shows that FPConv-DTI model is the potential for practical drug prediction. © 2021 Association for Computing Machinery.

A Maximum Flow-Based Approach to Prioritize Drugs for Drug Repurposing of Chronic Diseases.

The discovery of new drugs is required in the time of global aging and increasing populations. Traditional drug development strategies are expensive, time-consuming, and have high risks. Thus, drug repurposing, which treats new/other diseases using existing drugs, has become a very admired tactic. It can also be referred to as the re-investigation of the existing drugs that failed to indicate the usefulness for the new diseases. Previously published literature used maximum flow approaches to identify new drug targets for drug-resistant infectious diseases but not for drug repurposing. Therefore, we are proposing a maximum flow-based protein-protein interactions (PPIs) network analysis approach to identify new drug targets (proteins) from the targets of the FDA (Food and Drug Administration) drugs and their associated drugs for chronic diseases (such as breast cancer, inflammatory bowel disease (IBD), and chronic obstructive pulmonary disease (COPD)) treatment. Experimental results showed that we have successfully turned the drug repurposing into a maximum flow problem. Our top candidates of drug repurposing, Guanidine, Dasatinib, and Phenethyl Isothiocyanate for breast cancer, IBD, and COPD were experimentally validated by other independent research as the potential candidate drugs for these diseases, respectively. This shows the usefulness of the proposed maximum flow approach for drug repurposing.

Efficient machine learning model for predicting drug-target interactions with case study for Covid-19.

BACKGROUND: Discover possible Drug Target Interactions (DTIs) is a decisive step in the detection of the effects of drugs as well as drug repositioning. There is a strong incentive to develop effective computational methods that can effectively predict potential DTIs, as traditional DTI laboratory experiments are expensive, time-consuming, and labor-intensive. Some technologies have been developed for this purpose, however large numbers of interactions have not yet been detected, the accuracy of their prediction still low, and protein sequences and structured data are rarely used together in the prediction process. METHODS: This paper presents DTIs prediction model that takes advantage of the special capacity of the structured form of proteins and drugs. Our model obtains features from protein amino-acid sequences using physical and chemical properties, and from drugs smiles (Simplified Molecular Input Line Entry System) strings using encoding techniques. Comparing the proposed model with different existing methods under K-fold cross validation, empirical results show that our model based on ensemble learning algorithms for DTI prediction provide more accurate results from both structures and features data. RESULTS: The proposed model is applied on two datasets:Benchmark (feature only) datasets and DrugBank (Structure data) datasets. Experimental results obtained by Light-Boost and ExtraTree using structures and feature data results in 98 % accuracy and 0.97 f-score comparing to 94 % and 0.92 achieved by the existing methods. Moreover, our model can successfully predict more yet undiscovered interactions, and hence can be used as a practical tool to drug repositioning. A case study of applying our prediction model on the proteins that are known to be affected by Corona viruses in order to predict the possible interactions among these proteins and existing drugs is performed. Also, our model is applied on Covid-19 related drugs announced on DrugBank. The results show that some drugs like DB00691 and DB05203 are predicted with 100 % accuracy to interact with ACE2 protein. This protein is a self-membrane protein that enables Covid-19 infection. Hence, our model can be used as an effective tool in drug reposition to predict possible drug treatments for Covid-19.

Deep Learning in the Quest for Compound Nomination for Fighting COVID-19.

The current COVID-19 pandemic initiated an unprecedented response from clinicians and the scientific community in all relevant biomedical fields. It created an incredible multidimensional data-rich framework in which deep learning proved instrumental to make sense of the data and build models used in prediction-validation workflows that in a matter of months have already produced results in assessing the spread of the outbreak, its taxonomy, population susceptibility, diagnostics or drug discovery and repurposing. More is expected to come in the near future by using such advanced machine learning techniques to combat this pandemic. This review aims to unravel just a small fraction of the large global endeavors by focusing on the research performed on the main COVID-19 targets, on the computational weaponry used in identifying drugs to combat the disease, and on some of the most important directions found to contain COVID-19 or alleviating its symptoms in the absence of specific medication.