GPCNDTA: Prediction of drug-target binding affinity through cross-attention networks augmented with graph features and pharmacophores

Drug-target affinity prediction is a challenging task in drug discovery. The latest computational models have limitations in mining edge information in molecule graphs, accessing to knowledge in pharmacophores, integrating multimodal data of the same biomolecule and realizing effective interactions between two different biomolecules. To solve these problems, we proposed a method called Graph features and Pharmacophores augmented Cross-attention Networks based Drug-Target binding Affinity prediction (GPCNDTA). First, we utilized the GNN module, the linear projection unit and self-attention layer to correspondingly extract features of drugs and proteins. Second, we devised intramolecular and intermolecular cross-attention to respectively fuse and interact features of drugs and proteins. Finally, the linear projection unit was applied to gain final features of drugs and proteins, and the Multi-Layer Perceptron was employed to predict drug-target binding affinity. Three major innovations of GPCNDTA are as follows: (i) developing the residual CensNet and the residual EW-GCN to correspondingly extract features of drug and protein graphs, (ii) regarding pharmacophores as a new type of priors to heighten drug-target affinity prediction performance, and (iii) devising intramolecular and intermolecular cross-attention, in which the intramolecular cross-attention realizes the effective fusion of different modal data related to the same biomolecule, and the intermolecular cross-attention fulfills the information interaction between two different biomolecules in attention space. The test results on five benchmark datasets imply that GPCNDTA achieves the best performance compared with state-of-the-art computational models. Besides, relying on ablation experiments, we proved effectiveness of GNN modules, pharmacophores and two cross-attention strategies in improving the prediction accuracy, stability and reliability of GPCNDA. In case studies, we applied GPCNDTA to predict binding affinities between 3C-like proteinase and 185 drugs, and observed that most binding affinities predicted by GPCNDTA are close to corresponding experimental measurements.

Withasomniferol C, a new potential SARS-CoV-2 main protease inhibitor from the Withania somnifera plant proposed by in silico approaches

Exploring potent herbal medicine candidates is a promising strategy for combating a pandemic in the present global health crisis. In Ayurveda (a traditional medicine system in India), Withania somnifera (WS) is one of the most important herbs and it has been used for millennia as Rasayana (a type of juice) for its wide-ranging health benefits. WS phytocompounds display a broad spectrum of biological activities (such as antioxidant, anticancer and antimicrobial) modulate detoxifying enzymes, and enhance immunity. Inspired by the numerous biological actions of WS phytocompounds, the present investigation explored the potential of the WS phytocompounds against the SARS-CoV-2 main protease (3CL^pro). We selected 11 specific withanolide compounds, such as withaphysalin, withasomniferol, and withafastuosin, through manual literature curation against 3CL^pro. A molecular similarity analysis showed their similarity with compounds that have an established inhibitory activity against the SARS-CoV-2. In silico molecular docking and molecular dynamics simulations elucidated withasomniferol C (WS11) as a potential candidate against SARS-CoV-2 3CL^pro. Additionally, the present work also presents a new method of validating docking poses using the AlteQ method.

Targeting SARS-CoV-2: a systematic drug repurposing approach to identify promising inhibitors against 3C-like proteinase and 2'-O-ribose methyltransferase

The recent pandemic associated with SARS-CoV-2, a virus of the Coronaviridae family, has resulted in an unprecedented number of infected people. The highly contagious nature of this virus makes it imperative for us to identify promising inhibitors from pre-existing antiviral drugs. Two druggable targets, namely 3C-like proteinase (3CLpro) and 2'-O-ribose methyltransferase (2'-O-MTase) were selected in this study due to their indispensable nature in the viral life cycle. 3CLpro is a cysteine protease responsible for the proteolysis of replicase polyproteins resulting in the formation of various functional proteins, whereas 2'-O-MTase methylates the ribose 2'-O position of the first and second nucleotide of viral mRNA, which sequesters it from the host immune system. The selected drug target proteins were screened against an in-house library of 123 antiviral drugs. Two promising drug molecules were identified for each protein based on their estimated free energy of binding (ΔG), the orientation of drug molecules in the active site and the interacting residues. The selected protein-drug complexes were then subjected to MD simulation, which consists of various structural parameters to equivalently reflect their physiological state. From the virtual screening results, two drug molecules were selected for each drug target protein [Paritaprevir (ΔG = -9.8 kcal/mol) & Raltegravir (ΔG = -7.8 kcal/mol) for 3CLpro and Dolutegravir (ΔG = -9.4 kcal/mol) and Bictegravir (ΔG = -8.4 kcal/mol) for 2'-OMTase]. After the extensive computational analysis, we proposed that Raltegravir, Paritaprevir, Bictegravir and Dolutegravir are excellent lead candidates for these crucial proteins and they could become potential therapeutic drugs against SARS-CoV-2. Communicated by Ramaswamy H. Sarma.

Design of a Novel Multi Epitope-Based Vaccine for Pandemic Coronavirus Disease (COVID-19) by Vaccinomics and Probable Prevention Strategy against Avenging Zoonotics

The emergence and rapid expansion of the coronavirus disease (COVID-19) require the development of effective countermeasures especially a vaccine to provide active acquired immunity against the virus. This study presented a comprehensive vaccinomics approach applied to the complete protein data published so far in the National Center for Biotechnological Information (NCBI) coronavirus data hub. We identified non-structural protein 8 (Nsp8), 3C-like proteinase, and spike glycoprotein as potential targets for immune responses to COVID-19. Epitopes prediction illustrated both B-cell and T-cell epitopes associated with the mentioned proteins. The shared B and T-cell epitopes: DRDAAMQRK and QARSEDKRA of Nsp8, EDMLNPNYEDL and EFTPFDVVR of 3C-like proteinase, and VNNSYECDIPI of the spike glycoprotein are regions of high potential interest and have a high likelihood of being recognized by the human immune system. The vaccine construct of the epitopes shows stimulation of robust primary immune responses and high level of interferon gamma. Also, the construct has the best conformation with respect to the tested innate immune receptors involving vigorous molecular mechanics and solvation energy. Designing of vaccination strategies that target immune response focusing on these conserved epitopes could generate immunity that not only provide cross protection across Betacoronaviruses but additionally resistant to virus evolution.