A deep learning method for repurposing antiviral drugs against new viruses via multi-view nonnegative matrix factorization and its application to SARS-CoV-2

Abstract

The outbreak of COVID-19 caused by SARS-coronavirus (CoV)-2 has made millions of deaths since 2019. Although a variety of computational methods have been proposed to repurpose drugs for treating SARS-CoV-2 infections, it is still a challenging task for new viruses, as there are no verified virus-drug associations (VDAs) between them and existing drugs. To efficiently solve the cold-start problem posed by new viruses, a novel constrained multi-view nonnegative matrix factorization (CMNMF) model is designed by jointly utilizing multiple sources of biological information. With the CMNMF model, the similarities of drugs and viruses can be preserved from their own perspectives when they are projected onto a unified latent feature space. Based on the CMNMF model, we propose a deep learning method, namely VDA-DLCMNMF, for repurposing drugs against new viruses. VDA-DLCMNMF first initializes the node representations of drugs and viruses with their corresponding latent feature vectors to avoid a random initialization and then applies graph convolutional network to optimize their representations. Given an arbitrary drug, its probability of being associated with a new virus is computed according to their representations. To evaluate the performance of VDA-DLCMNMF, we have conducted a series of experiments on three VDA datasets created for SARS-CoV-2. Experimental results demonstrate that the promising prediction accuracy of VDA-DLCMNMF. Moreover, incorporating the CMNMF model into deep learning gains new insight into the drug repurposing for SARS-CoV-2, as the results of molecular docking experiments reveal that four antiviral drugs identified by VDA-DLCMNMF have the potential ability to treat SARS-CoV-2 infections.

Acceptance of clinical artificial intelligence among physicians and medical students: A systematic review with cross-sectional survey

“…For example, DL-enabled AI could be used to address current dilemmas such as the workforce shortage and could ensure there is consistency by reducing variability in medical practice and by standardizing the quality of care ( 5 ). Some have suggested that the increasing use of AI will fundamentally change the nature of healthcare provision and clinical practice ( 6 – 8 ). However, this gradual transition could also cause concerns within the medical profession because adopting new technologies requires changes to medical practice.…”

Role of natural products towards the SARS-CoV-2: A critical review

“…Remdesivir, favipiravir, lopinavir, ritonavir, and arbidol are some of the antiviral therapy options that have been found as prospective alternatives to the medicine that is currently used to treat COVID-19. These drugs work through a variety of different methods of action, including the prevention of virus replication within host cells and the inhibition of viral entry into host cells [ [68] , [69] , [70] , [71] ]. In addition, immunotherapeutic approaches have emerged as one of the most prominent therapeutic modalities in recent years.…”

“…The use of convalescent plasma treatment and monoclonal antibodies has been characterized as a potential therapeutic strategy that is both effective and reliable [ 69 , 73 ]. This method seeks to strengthen patients' immune systems and prevent them from contracting viral infection [ [69] , [70] , [71] , [72] , [73] , [74] ]. Nevertheless, medicinal plants and NPs are still seen as viable options for preventing and treating various ailments.…”

Multi-view heterogeneous molecular network representation learning for protein–protein interaction prediction

“…During this process, it should be noted that all the parameters used in these baseline algorithms are the same as their original works. Moreover, we randomly divide all approved PPIs as positive samples and then the same number of negative samples are randomly selected from the complement set of positive samples [42].…”