Identification of oral therapeutics using an AI platform against the virus responsible for COVID-19, SARS-CoV-2

Abstract

Purpose: This study introduces a sophisticated computational pipeline, eVir, designed for the discovery of antiviral drugs based on their interactions within the human protein network. There is a pressing need for cost-effective therapeutics for infectious diseases (e.g., COVID-19), particularly in resource-limited countries. Therefore, our team devised an Artificial Intelligence (AI) system to explore repurposing opportunities for currently used oral therapies. The eVir system operates by identifying pharmaceutical compounds that mirror the effects of antiviral peptides (AVPs)-fragments of human proteins known to interfere with fundamental phases of the viral life cycle: entry, fusion, and replication. eVir extrapolates the probable antiviral efficacy of a given compound by analyzing its established and predicted impacts on the human protein-protein interaction network. This innovative approach provides a promising platform for drug repurposing against SARS-CoV-2 or any virus for which peptide data is available. Methods: The eVir AI software pipeline processes drug-protein and protein-protein interaction networks generated from open-source datasets. eVir uses Node2Vec, a graph embedding technique, to understand the nuanced connections among drugs and proteins. The embeddings are input a Siamese Network (SNet) and MLPs, each tailored for the specific mechanisms of entry, fusion, and replication, to evaluate the similarity between drugs and AVPs. Scores generated from the SNet and MLPs undergo a Platt probability calibration and are combined into a unified score that gauges the potential antiviral efficacy of a drug. This integrated approach seeks to boost drug identification confidence, offering a potential solution for detecting therapeutic candidates with pronounced antiviral potency. Once identified a number of compounds were tested for efficacy and toxicity in lung carcinoma cells (Calu-3) infected with SARS-CoV-2. A lead compound was further identified to determine its efficacy and toxicity in K18-hACE2 mice infected with SARS-CoV-2. Computational Predictions: The SNet confidently differentiated between similar and dissimilar drug pairs with an accuracy of 97.28% and AUC of 99.47%. Key compounds identified through these networks included Zinc, Mebendazole, Levomenol, Gefitinib, Niclosamide, and Imatinib. Notably, Mebendazole and Zinc showcased the highest similarity scores, while Imatinib, Levemenol, and Gefitinib also ranked within the top 20, suggesting their significant pharmacological potentials. Further examination of protein binding analysis using explainable AI focused on reverse engineering the causality of the networks. Protein interaction scores for Mebendazole and Imatinib revealed their effects on notable proteins such as CDPK1, VEGF2, ABL1, and several tyrosine protein kinases. Laboratory Studies: This study determined that Mebendazole, Gefitinib, Topotecan and to some extent Carfilzomib showed conventional drug-response curves, with IC50 values near or below that of Remdesivir with excellent confidence all above R2>0.91, and no cytotoxicity at the IC50 concentration in Calu-3 cells. Cyclosporine A showed antiviral activity, but also unconventional drug-response curves and low R2 which are explained by the non-dose dependent toxicity of the compound. Additionally, Niclosamide demonstrated a conventional drug-response curve with high confidence; however, its inherent cytotoxicity may be a confounding element that misrepresents true antiviral efficacy, by reflecting cellular damage rather than a genuine antiviral action. Remdesivir was used as a control compound and was evaluated in parallel with the submitted test article and had conventional drug-response curves validating the overall results of the assay. Mebendazole was identified from the cell studies to have efficacy at non-toxic concentrations and were further evaluated in mice infected with SARS-CoV-2. Mebendazole administered to K18-hACE2 mice infected with SARS-CoV-2, resulted in a 44.2% reduction in lung viral load compared to non-treated placebo control respectively. There were no significant differences in body weight and all clinical chemistry determinations evaluated (i.e., kidney and liver enzymes) between the different treatment groups. Conclusion: This research underscores the potential of repurposing existing compounds for treating COVID-19. Our preliminary findings underscore the therapeutic promise of several compounds, notably Mebendazole, in both in vitro and in vivo settings against SARS-CoV-2. Several of the drugs explored, especially Mebendazole, are off-label medication; their cost-effectiveness position them as economical therapies against SARS-CoV-2.

Keywords: COVID-19; SARS-CoV-2; antiviral activity; antiviral peptides; artificial intelligence; infectious diseases; oral therapeutics; siamese networks.

FIGURE 1

The input to the eVir software pipeline is a combination drug-protein and protein-protein interaction network. We use theNode2Vectool to generate embeddings for the drugs and AVPs. We then use four distinct multi-layer perceptrons (MLPs) to compute probabilities for likelihood of a drug distinctly having an impact on entry, replication, and fusion. We simultaneously use the AVP embeddings and drug embeddings as input into a Siamese Neural Network (SNet) to acquire probabilities of a specific drug effecting each mechanism. Last, we score the input drug by combining MLP and SNet probabilities.

FIGURE 2

The CDT function simulates the Gaussian distribution of expected drug concentrations, centered around a defined mean with a specific variance. This approach is typically used to model the distribution patterns of a substance’s concentration, assuming it follows a normal distribution. For established drugs, the function utilizes their known dosing and lethal dose (LD50) means and variance to project the expected distribution and influence the outcome. Conversely, for drugs without established parameters, it defaults to using the median concentration value from the entire dataset to estimate the distribution.

FIGURE 3

The neural network architecture of the Siamese Network (SNet). The training process utilizes the Adam optimizer with hyperparameters, including a learning rate of 0.00001, epsilon set to 1e-7, and momentum parameters (beta_1 = 0.9, beta_2 = 0.999). Adam’s adaptive learning rate and momentum help accelerate convergence and improve the model’s ability to learn complex patterns from the data.

FIGURE 4

Architecture of the MLP Networks. The model consists of an input layer with 512 neurons, followed by two hidden layers with 256 and 128 neurons, respectively. Each hidden layer utilizes the Rectified Linear Unit (ReLU) activation function, promoting non-linearity and feature extraction. To enhance the model’s learning ability, weight masking is incorporated to selectively emphasize important connections and reduce noise.

FIGURE 5

Histogram of predictions from the SNet on a test set of drug-drug pairs. Green peaks signify similar drug pairs and the red peaks display the drug pairs that are dissimilar. The optimal threshold is based on a J-statistic to find the best separation between the two populations. Embedded figure: a Receiver Operating Characteristic (ROC) curve verification of the SNet performance (97.28% Accuracy, 99.47% AUC). We have high confidence with known data that SNet can discriminate between similar and dissimilar drug pairs.

FIGURE 6

In vitro IC-50 drug-response curves for Calu-3 cells.