This study demonstrates the advantages of integrating AI with FBDD to streamline molecular design, providing a data-driven framework for discovering new Mac1 inhibitors and guiding future antiviral drug development.
Abstract
Fragment-based drug discovery (FBDD) is an effective approach for exploring chemical space using small, low-affinity fragments as starting points to facilitate development of lead compounds. Strategies to improve fragment potency include fragment merging and linking to generate higher-affinity inhibitors. Recently, artificial intelligence (AI) and machine learning (ML) have accelerated this process through structure-based optimization and generative compound design. Here, we present an AI-assisted FBDD workflow applied to the SARS-CoV-2 macrodomain (Mac1), a conserved viral protein involved in immune evasion and ADP-ribose metabolism. Using available structural data and previously identified fragments, we combined deep learning with molecular docking to design novel Mac1 binders. Selected compounds were synthesized and validated by NMR spectroscopy and X-ray crystallography, demonstrating improved binding relative to the original fragment hits with KD values in the range of 299-990 µM. This study demonstrates the advantages of integrating AI with FBDD to streamline molecular design, providing a data-driven framework for discovering new Mac1 inhibitors and guiding future antiviral drug development.
The design and validation of a non-proprietary 19 F NMR screening library (OpenFL600) capable of probing diverse binding pockets, as demonstrated for RNA, G protein-coupled receptors, kinases, protein-protein interactions, and proteases is presented.
Simon H. Rüdisser, Gabriela R. Stadler, G. Padroni et al.· Angewandte Chemie· 0 citations
A fragment-based drug discovery strategy combined with crystallographic screening and structural similarities among the fragments provide a robust structural framework for the rational design of PLpro inhibitors and support the development of novel antiviral therapeutics.
Tingting Wu, Zheng Zhou, Xingyu Li et al.· Journal of Biological Chemis...· 0 citations
The rapid evolution of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and the emergence of drug resistance necessitate the development of multi-target antiviral therapeutics. This study aimed to computationally design a novel triple-target inhibitor derived from the natural flavonoid Baicalein, targeting the essential SARS-CoV-2 enzymes 3CLpro, PLpro, and RdRp to combat viral resistance. An integrated computational approach was employed, including virtual screening of compound libraries, molecular docking using Molecular Operating Environment software, fragment-based drug design to optimize Baicalein, and molecular dynamics simulations over 200 ns with AMBER 18 to assess binding stability. Binding free energies were calculated via MM/GBSA methods, and pharmacokinetic properties were evaluated using ADME/T predictions. Retrosynthetic analysis was performed to confirm synthetic feasibility. The novel compound BD02 demonstrated stable binding to all three targets, with significantly improved binding free energies compared to Baicalein: 3CLpro (-50.68 kcal/mol), PLpro (-59.05 kcal/mol), and RdRp (-51.98 kcal/mol). Molecular dynamics simulations showed low root mean square deviation values, indicating high structural stability. ADME/T predictions revealed favorable drug-like properties, including good absorption and low toxicity risks. BD02 is a computationally promising synthetic lead scaffold for broad-spectrum anti-coronavirus design. This multi-target design theoretically helps mitigate viral resistance and provides structural design references.
Di Han, Hongkun Yang, Yifan Wang et al.· Bioorganic chemistry (Print)· 0 citations
We report the discovery of 31 SARS-CoV-2 inhibitors identified across two computational-to-experimental screening campaigns, with average cell-based antiviral potency satisfying 〈EC50〉 ≤ CC50/3 in A549-hACE2 cells. These compounds were selected from 60 successfully tested molecules evaluated using nanoluciferase reporter assay (nLuc), cytopathic effect (CPE), and host-cell cytotoxicity (CC50) assays. We used a funnel-like computational framework where candidate compounds were progressively prioritized through different scores, improving robustness against the limitations of individual metrics. However, none of the docking-based scoring metrics, including MM/GBSA and Glide scores, showed statistically significant correlation with experimental activity, illustrating the limitations of docking-only approaches and supporting a major contribution of the machine-learning predictions to the high hit-identification rate. Although direct target validation is still lacking, the machine-learning (ML) workflow was grounded in experimental nsp13 inhibition labels, and molecular dynamics (MD) simulations suggest that these compounds can act by stabilizing apo-like open conformations of nsp13, with differential interactions involving the 1B domain contributing to potency differences. Among the compounds evaluated, the phenoxypropanol (PP) and bipiperidine (BPP) series stood out as the most promising SARS-CoV-2 antivirals. Integrated analysis of structure–activity relationships, physicochemical parameters, metabolic stability, and interdomain dynamics delineated structural optimization paths for both series.
Alma C. Castañeda-Leautaud, Thomas D. Bannister, Eunjung Kim et al.· Chemical Science· 0 citations
Histone deacetylase 2 (HDAC2) is a promising epigenetic target for cancer therapy; however, the clinical utility of Vorinostat (SAHA) is limited by poor selectivity, toxicity, and inadequate blood–brain barrier (BBB) permeability. A structure–activity relationship (SAR)-guided strategy was employed to identify novel coumarin-based HDAC2 inhibitors. A total of 2555 PubChem coumarin derivatives were filtered using Lipinski's criteria, yielding 1879 drug-like molecules for docking against HDAC2 (PDB ID: 7ZZT). Seventeen lead compounds guided the design of 46 novel derivatives. Docking identified compounds with binding affinities of −7.0 to −9.3 kcal/mol, exceeding SAHA (−7.3 kcal/mol). SAR analysis highlighted the coumarin scaffold, an additional aromatic ring, a 3–6 atom linker, a furan bridge, and terminal hydroxyl or methoxy groups as key structural features. Compound C5 exhibited high functional similarity with SAHA (R
2
= 0.898), favorable BBB permeability, superior predicted HDAC2 inhibition (IC
50
= 0.560 µM vs. 1.116 µM), enhanced predicted antiproliferative activity against MCF7, MDA-MB-231, HL-60, and U87MG cell lines, and stable binding confirmed by steered and 100 ns molecular dynamics simulations, identifying Asp104 as a key stabilizing residue.
Unknown authors· Main group chemistry (Print)· 0 citations
This study applies MPS2IT-DTI (Molecule and Protein Sequence to Image Transformer for Drug-Target Interaction), a deep learning framework that represents molecular and protein sequences as images as images using k-mer frequency encoding, enabling convolutional neural networks to capture spatial compositional patterns associated with biochemical interactions.
Jackson G. de Souza, Marcelo A. C. Fernandes, Raquel de Melo Barbosa· Computational biology and ch...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
