QSAR-ML- and Metadynamics-Guided Design of Symmetrical Bis-Indanones to Overcome Mutational Anchor Loss in Acetylcholinesterase

Mavchen-1: A Conformational Ensemble Platform for Protein–Ligand Pose Prediction That Substantially Outperforms Static Structure Prediction in a Category-Stratified Benchmark

A category-stratified, statistically powered benchmark comparing pose prediction from receptor conformational ensembles against AlphaFold2, used as a matched static-structure baseline, across 29 protein–ligand systems spanning cryptic-pocket, induced-fit, water-mediated, and autoimmune-indication target classes is presented.

Combining Stability-Centered Atomistic Design with Machine Learning for Targeted Enzyme Optimization

A machine-learning-assisted enzyme-engineering (MLEE) workflow that adds substrate-specific functional information to htFuncLib through an initial screening and sequencing round that may bypass the need for transition-state models and reduce the effort required for obtaining high-activity variants.

QSAR-Guided Virtual Screening and Molecular Dynamics Reveal Olaparib as a Repurposing Lead Against α-Synuclein Aggregation

Parkinson’s disease (PD) is characterised by the pathological aggregation of α-synuclein (α-syn) into Lewy body inclusions, yet no disease-modifying therapy exists. To address this, we developed an integrated computational pipeline combining quantitative structure–activity relationship (QSAR) modelling, structure-based virtual screening, molecular dynamics (MD) simulation, and molecular mechanics Poisson–Boltzmann surface area (MM-PBSA) binding free energy calculations to repurpose FDA-approved drugs as α-syn fibril inhibitors. Two complementary QSAR model families were trained on 501 α-syn binding affinity records from BindingDB: Morgan extended-connectivity fingerprint (ECFP4) classifiers and a frozen ChemBERTa-77M-MLM transformer encoder, each using Random Forest and Logistic Regression. The applicability domain (AD) was assessed using Morgan–Tanimoto similarity (Tc ≥ 0.40) and calibrated ChemBERTa cosine distance (θ ≤ 0.367). A three-stage funnel applying central nervous system (CNS) permeability filters, a consensus QSAR probability threshold (≥0.80), and AD gating reduced 2241 FDA-approved drugs to 205 candidates for AutoDock Vina 1.2.6 docking against two sites on the cryo-electron microscopy (cryo-EM) α-syn fibril structure, PDB 6SSX: the inter-protofilament cleft (Site 1) and the non-amyloid-beta component (NAC) groove (Site 2). The Morgan fingerprint models achieved an area under the receiver operating characteristic curve (AUROC) of up to 0.940 and a balanced accuracy of 0.810; the ChemBERTa models achieved an AUROC of 0.785 and a balanced accuracy of 0.728. Notably, ChemBERTa AD covered 76.8% of the FDA drugs versus only 5.5% for Morgan–Tanimoto, enabling broad-spectrum screening. The top docking candidates were Olaparib (−7.91 kcal/mol), Paliperidone (−7.75 kcal/mol), Niraparib (−7.18 kcal/mol), Dordaviprone (−7.06 kcal/mol), and Parecoxib (−6.89 kcal/mol). The MD simulations over 200 ns across three independent replicates confirmed stable NAC groove binding, and replicate-averaged MM-PBSA calculations yielded ΔG = −20.6 ± 1.9 kcal/mol for Olaparib at Site 2, −17.1 ± 0.9 kcal/mol for Risperidone, and −16.9 ± 0.8 kcal/mol for Paliperidone, reported as the mean ± standard error of the mean (SEM) across replicates. Olaparib additionally formed five hydrogen bonds in the representative pose, while MD trajectories maintained approximately 2–5 hydrogen bonds, together with a halogen bond within the NAC groove, the largest contact count of any screened compound. These findings identify Olaparib as a novel high-affinity repurposing lead, while Paliperidone and Risperidone are reported as chemically informative secondary NAC–groove binders rather than proposed antiparkinsonian therapeutics, given that their dopamine D2-antagonist pharmacology is clinically associated with drug-induced parkinsonism. All of the candidates warrant experimental validation via thioflavin-T fluorescence or nuclear magnetic resonance (NMR) spectroscopy.

Overcoming the accuracy–generalization tradeoff in docking and scoring for prospective virtual screening

Integrating Parallel-Bias Metadynamics-Metainference and Relaxation Dispersion NMR to Resolve Hidden Conformational States Driving Toxin-Channel Recognition.

Conformational dynamics in toxin inhibitors are an important contributor to ion channel affinity, yet toxin multidimensional energy landscapes remain largely unexplored. In the current work, we combine parallel-bias metadynamics-metainference (PBMetaD) simulations with relaxation dispersion NMR to define, at atomistic resolution, the thermodynamics and kinetics of Hui1, a de novo three disulfide toxin derived from the SAK-I family that targets K+-channels. Using the three χ3 disulfide dihedrals as collective variables, an extensive 48-replica well-tempered PBmetaD simulation (16.2 μs cumulative sampling) resulted in a fully converged three-dimensional (3D) free-energy surface comprising eight Hui1 conformers. These basins account for ∼96% of the bias-weighted ensemble and partition into four low- and four high-energy states separated by 7.5 kJ/mol associated with the (-) and (+)Cys12-Cys28 χ3 states, respectively. Transition-state theory identifies rota-isomerization of Cys3-Cys35 as the slowest, and therefore rate-determining, coordinate, while the analogous motions around Cys12-Cys28 and Cys17-Cys32 are ∼5-fold faster. 15N R1ρ relaxation dispersion NMR measurements confirmed these kinetics, identifying two structurally distinct residue clusters exhibiting intermediate and faster exchange processes. The Key Interaction Finder (KIF) approach reveals that Cys17-Cys32 conformation modifies connectivities within a dense interaction network between Cys17 and residues Gln14, Tyr23, Arg24, and Lys29, and correlates with the accessibility of residues Tyr23 and Arg24 of the helix-kink-helix region for interaction with the channel vestibule. Our work establishes PBMetaD as a powerful framework for mapping coupled disulfide and backbone dynamics in toxins and reveals specific conformers and interactions likely to control K+ channel recognition.

Chemical features consistently associated with kinase inhibitor activity: a cross-target SHAP analysis under scaffold-split evaluation.

Interpretable machine learning approaches to quantitative structure-activity relationship (QSAR) modelling are increasingly applied in drug discovery, yet most studies remain confined to single targets and report feature attributions without translating them into chemically meaningful insights. We introduce a cross-target SHAP entropy framework for quantifying shared versus target-specific structure-activity relationships across protein families, applied to 31 human kinase targets from BindingDB under scaffold-based train-test evaluation. Random Forest classifiers trained on Morgan ECFP4 fingerprints achieved a median AUROC of 0.994, AUPRC of 0.9998, and MCC of 0.674, confirming genuine SAR learning beyond class prevalence exploitation. Pairwise Spearman rank correlation of mean absolute SHAP profiles across targets yielded moderate cross-target consistency (mean r = 0.332; 465 pairs). Shannon entropy-based classification of the top 200 fingerprint bits identified 15 consensus features dominated by aromatic N-heterocycles, aliphatic rings, and hydrogen bond environments, and 50 divergent features showing 2.8-fold higher SHAP magnitude than consensus features. SAR validation confirmed genuine enrichment of two top consensus fragments in active compounds. These findings indicate that kinase QSAR models share a low-magnitude consensus descriptor signal across the kinase family, while target-specific features dominate predictive decision boundaries. All SHAP attributions describe model decision behavior and should not be interpreted as causal binding mechanisms. The entropy decomposition framework is generalisable to other protein families and provides a transferable workflow for converting SHAP outputs into chemically actionable insights. All code and data are publicly available.