The rapid emergence of metallo-b-lactamase-mediated antibiotic resistance has created an urgent need for new inhibitor discovery strategies. In this work, a machine-learning-guided workflow was developed to generate and prioritize potential inhibitors targeting NDM-1. A SMILES-based variational autoencoder was first pretrained on a broad molecular dataset to learn general chemical syntax and latent molecular representations. The model was then fine-tuned on an 8-hydroxyquinoline-enriched dataset to bias molecular generation toward zinc-binding chemical space relevant to metallo-β-lactamase inhibition. Generated compounds were processed through structural filtering and docking-based evaluation to create training data for downstream predictive modeling. Molecular fingerprints and physicochemical descriptors were then used to train XGBoost models for docking score prediction and classification of potential binders. Classification proved especially useful for prescreening because it avoided overinterpreting small differences in noisy docking scores while still enriching for compounds likely to perform well in docking. The resulting workflow demonstrates how generative modeling and supervised machine learning can be combined to reduce chemical search space, prioritize candidate inhibitors, and guide computational drug discovery. Although experimental validation remains necessary, this approach provides a scalable framework for identifying promising zinc-binding compounds for further molecular simulation and inhibitor development that can be expanded in future studies.
Breast cancer is one of the most prevalent and lethal malignancies affecting women globally. The increasing resistance to current therapeutic strategies highlights the need for novel molecular targets. Inositol-requiring enzyme 1 alpha (IRE1α), a key sensor in the unfolded protein response (UPR), has emerged as a promising therapeutic target due to its role in tumour progression and survival. This study employed an integrative in silico approach combining machine learning, molecular docking, and molecular dynamics simulations to identify potent, non-toxic IRE1α inhibitors for breast cancer treatment. An initial library of 115 compounds retrieved from ChEMBL and MedChemExpress was used for machine learning-based toxicity modelling. Literature curation identified 44 reported IRE1α inhibitors, which were reduced to 38 unique compounds following duplicate removal. Drug-likeness and ADMET screening using SwissADME and ProTox retained 22 compounds for further evaluation. Molecular docking was performed using AutoDock, followed by Dynamics simulations in GROMACS to assess stability. Machine Learning (ML) models were developed for both toxicity regression and binary toxicity classification analyses. Toxicity prediction models were developed using twenty physicochemical and pharmacokinetic descriptors. In the regression analysis, Random Forest demonstrated the strongest cross-validation performance (R² = 0.5998 ± 0.3439), while the stacking ensemble achieved the highest test-set performance (R² = 0.9765), although differences among ensemble methods were not statistically significant. In the complementary classification analysis, the Support Vector Machine (SVM) achieved the highest discriminative performance with an ROC-AUC value of 0.98. Docking studies revealed that Z4P exhibited the strongest binding affinity (- 7.93 kcal/mol) to the wild-type IRE1, compared with the control drug MKC8866 (- 6.7 kcal/mol). Additionally, Z4P exhibited a higher binding energy of - 9.5 kcal/mol, whereas MKC8866 had a binding energy of - 6.94 kcal/mol. MD simulations over 200 ns confirmed the stability of the IRE1-Z4P complex, with favourable RMSD, RMSF, Rg, and SASA profiles relative to the control. These findings highlight Z4P as a promising mutation-resilient IRE1 inhibitor and validate the effectiveness of the integrated computational pipeline for identifying potential anti-cancer therapeutics.
Nithisha L Bastin, P. K. Praveen Kumar, B. Ethiraj et al.· Scientific Reports· 0 citations
Background: Tankyrase 1 (TNKS1) is a poly(ADP-ribose) polymerase involved in Wnt/β-catenin signaling, telomere maintenance, and genomic stability, making it an attractive therapeutic target in oncology. This study aimed to develop and apply an integrated computational workflow to identify novel TNKS1 inhibitor candidates. Methods: A curated dataset of experimentally validated TNKS1 inhibitors and property-matched DUD-E decoys was used to develop a consensus supervised machine learning (ML) model prioritization framework for ligand-based virtual screening, integrating Morgan fingerprints with three complementary classifiers. The model screened more than 700,000 compounds, and prioritized hits were evaluated by structure-based virtual screening (SBVS), Prime MM-GBSA binding free-energy refinement, and 500 ns molecular dynamics simulations (MDs). The top candidates were subsequently tested in an in vitro TNKS1 enzymatic inhibition assay. Results: The consensus ML framework prioritized 670 compounds, yielding five candidates for experimental testing. Compound 3 displayed the most favorable computational profile and was experimentally confirmed as a TNKS1 inhibitor candidate, exhibiting approximately 80% TNKS1 inhibition at 0.1 μM, whereas the remaining candidates showed only limited activity. Conclusions: The proposed workflow efficiently reduced a large chemical space to a focused set of TNKS1 inhibitor candidates while substantially reducing the experimental screening burden. Compound 3 represents a promising starting point for future structure–activity relationship studies and lead optimization in the context of TNKS1 inhibition. Moreover, this work highlights the value of integrating consensus ML, SBVS, and experimental validation to accelerate early-stage hit discovery for TNKS1 and other therapeutic targets.
M. Bilotta, Adriana Gargano, R. Rocca et al.· Pharmaceuticals· 0 citations
Interferon-inducible RNA-dependent protein kinase (PKR) is an emerging therapeutic target involved in cancer, neurodegeneration, and inflammatory disorders; however, the discovery of potent and structurally diverse PKR inhibitors remains limited. In this study, we report an integrated computational–experimental strategy for the identification of novel PKR inhibitory chemotypes. Pharmacophore models were generated from flexible docking of structurally diverse reference inhibitors and validated using receiver operating characteristic (ROC) analysis. To better capture ligand flexibility and address data scarcity, multiple conformations of 223 PKR inhibitors were employed as a data augmentation strategy in machine learning-based QSAR modelling. Among several algorithms evaluated, a Naïve Bayes classifier combined with genetic function algorithm (GFA) feature selection provided the most predictive model. The optimized model, incorporating a single pharmacophore hypothesis and key physicochemical descriptors, was applied to virtual screening of the Boehringer Ingelheim opnMe compound library. Experimental validation using a PKR kinase assay identified the pre-synthesized compound BI-8128 as a potent PKR inhibitor, with an IC50 value of 174.8 nM, demonstrating higher potency than the reference inhibitor C16 under identical conditions. Notably, this compound represents a structurally distinct scaffold compared to reported PKR inhibitors, indicating effective scaffold hopping. To the best of our knowledge, this is the first report of PKR inhibitory activity for BI-8128. Overall, this work demonstrates that integrating docking-derived pharmacophores with conformational ensemble-based machine learning provides an effective approach for discovering novel inhibitors against underexplored kinase targets.
G. Shakhatreh, M. Taha, S. Daoud· RSC Advances· 0 citations