Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 30 appraisals
Journal of chemical information and modeling
HELMify: A Hybrid Rule- and LLM-Based Generator of Peptide Monomer HELM Names
HELM is a hierarchical notation system for biopolymers that is an increasingly popular choice for representing peptides. In this system, each monomer name must uniquely identify a single monomer, and until now, chemists have named peptide monomers by hand using a limited set of rules. As the chemical space of monomers is practically unlimited, this naming process is too time-consuming to be done by chemists and is prone to inconsistency. To address this bottleneck, we developed a public benchmark data set for this task and two complementary automated naming approaches for HELM name generation: a zone-based namer, which relies on hand-curated dictionaries of substructures, and a generative AI namer, which relies on a pretrained Large Language Model (LLM) with chemistry world knowledge. We evaluated both approaches on a set of publicly disclosed alpha amino acids. For a time-split test set, the zone-based namer produced usable names for 84% of cases: 54% matched the human-generated name exactly, and an additional 30% did not match but were interpretable as the correct structure. The LLM-based namer produced usable names for 42% of cases: 33% of the names matched the human-generated name exactly, and an additional 9% were nonidentical but still interpretable as the correct structure. In another 49% of LLM-generated cases, the outputs were considered close matches, and while not directly usable, they can still inform the generation of correct names. The zone-based namer is deterministic, fast, and independent of any previous names, explicitly handling zone order and chirality. The LLM-based namer can generalize from human-named examples to propose names for novel chemical groups, helping fill library gaps. Together, the approaches offer complementary coverage: rules-based precision and speed paired with data-driven flexibility for novel moieties. Combining the strengths of both approaches paves the way for an automated peptide monomer naming process that can accelerate research by (i) helping the community establish standardized guidelines for the rapidly expanding noncanonical amino acid (ncAA) space, (ii) streamlining integration with synthesis planning, inventory management, and analysis pipelines, (iii) enabling more consistent annotation in ELNs, databases, and publications, and (iv) facilitating cross-lab replication and data exchange. Together, these advances can expedite method development, reduce ambiguity in collaboration, and ultimately support the discovery of novel peptides with improved drug-like properties.
29 July 2026
Read appraisal →Journal of chemical information and modeling
PegaPlus─Interactive Machine Learning by Human Observation for Efficient Clustering and Analysis of Structure-Activity Data
In the early stages of drug discovery, the identification of molecular series with lead-like properties is essential for structure-activity relationship studies. A common first step is to cluster data derived from high-throughput screening (HTS). This data then needs to be exhaustively refined by an expert to identify bioactive compound classes. Here, we present the novel, interactive, and visual approach named PegaPlus for the analysis of HTS data, facilitating the incorporation of expert knowledge with the strategy of learning by observation. PegaPlus uses a tailored variant of a stochastic proximity embedding algorithm to visualize the clustered data in a two-dimensional plane via a web interface. The user can iteratively modify the proposed compound clustering by dragging data points in the drawing plane. At each step, an online support vector machine learns and improves its clustering predictions by incorporating user modifications and the molecular features of the compounds. The visualization is also updated based on the user's input and molecular similarity. PegaPlus was developed with the exploratory process in mind and allows the user to freely organize molecules on the drawing plane, or filter them by various molecular properties. Our evaluation shows that PegaPlus reduces the number of refinement steps needed by half in comparison to the purely manual approach on multiple data sets. These results demonstrate how an interactive machine learning approach can reliably support medicinal chemists and automate time-consuming, manual refinement tasks. The PegaPlus web server is openly available at pegaplus.zbh.uni-hamburg.de.
14 July 2026
Read appraisal →Nucleic acids research
DeepCYP: an integrated deep learning web server for the holistic 'pathway-site product' prediction of CYP450 metabolism
CYP450 (cytochrome P450)-mediated drug metabolism is a critical determinant of pharmacokinetics and clinical safety, making comprehensive metabolic profiling essential for rational drug discovery. Here, we present DeepCYP (https://deepcyp.scbdd.com), a freely accessible deep-learning web server for end-to-end CYP450 metabolic profiling. Trained on an expanded dataset and a mechanism-based reaction rule library, DeepCYP uses a multi-task graph neural network (GNN) combined with multi-scale descriptors. Operating directly on 2D molecular graphs, this architecture bridges the entire "pathway-site-product" continuum across nine major CYP isoforms (CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A4) within a unified pipeline. Benchmarking demonstrates that DeepCYP outperforms established tools, including FAME3, SMARTCyp, and BioTransformer 3.0, improving Top-1 and Top-2 ranking metrics by over 10%. Furthermore, the server supports high-throughput batch processing, capable of evaluating ~280 molecules per minute. DeepCYP also enhances interpretability through an interactive visualization interface, featuring susceptibility radar charts and dynamic transformation tables. By translating abstract predictions into biological insights, DeepCYP provides a practical tool to accelerate lead optimization and mitigate toxicity risks.
13 July 2026
Read appraisal →Nucleic acids research
DeepKinomeWeb: a quantitative, panel-level platform for kinase inhibitor screening and selectivity profiling
Protein kinases are central targets in drug discovery, yet early-stage development of potent and selective inhibitors remains challenging due to high experimental costs and limited interpretability of large-scale screening data. Here, we present DeepKinomeWeb, an integrated web-based platform that transforms competition-based high-throughput screening data into actionable insights for kinase inhibitor prioritization. Built upon our previously validated deep learning regression model, DeepKinome, the platform enables quantitative prediction of kinase-inhibitor binding affinities and provides panel-level visualization of selectivity landscapes, selectivity metric calculations, and integrated structural and physicochemical analyses. Through its user-friendly interface, DeepKinomeWeb supports rational, data-driven decision-making for biologists and medicinal chemists, lowering the barrier to systematic selectivity assessment in kinase inhibitor discovery. DeepKinomeWeb is freely available to all users without any login requirement at https://str.kribb.re.kr/deepkinome.
13 July 2026
Read appraisal →Journal of medicinal chemistry
Machine Learning-Based Models to Predict Drug-Induced Liver Injury (DILI) to Assist Medicinal Chemistry
Drug-induced liver injury (DILI) is a leading cause of drug failure and post-market withdrawals. Traditional preclinical methods fail to detect up to 40-45% of clinical hepatotoxicity cases. Computational approaches, particularly those based on machine learning and deep learning (DL), are emerging as promising tools to support medicinal chemistry and early drug discovery, though their predictive capabilities remain under active investigation. In this perspective, we review the development of DILI annotation data sets, tracing their growth from small collections to large, comprehensive resources. We also outline the evolution of computational methods, from simple descriptor-based models to advanced DL and ensemble approaches that incorporate interpretable features. Finally, we highlight recent efforts to integrate standardized causality frameworks, pharmacogenomics, and mechanistic models, aiming to connect computational advances with clinical relevance. This perspective provides valuable insight for researchers and promotes the development of more robust and consensual DILI prediction strategies.
10 July 2026
Read appraisal →Journal of computer-aided molecular design
Two faces of drug repurposing data in machine learning-driven structure-based virtual screening supported by interaction fingerprints
Ensemble docking is a prominent source of protein-ligand complex structures, which could be conveniently analyzed with interaction fingerprints approach in a machine learning framework. Nevertheless, the chemical nature and provenance of the source data libraries used for model development may influence the predictive ability. We assessed the performance of SARS-CoV-2 main protease (Mpro) inhibitors classification by random forest models trained on different fingerprint implementations using three inhibitor libraries, comprising drug repurposing data or compounds rationally designed as Mpro inhibitors. Models trained on drug repurposing data showed a low prediction performance on their own, while infusion of these data into libraries with specific inhibitors greatly improved applicability domains of the models. Our results highlight the importance of balanced dataset preparation for machine learning applications in medicinal chemistry, especially when relied on drug repurposing experiments.
8 July 2026
Read appraisal →Molecular pharmaceutics
Computational Identification of Active Drug Metabolites for Human Protein Targets
Understanding drug metabolism helps mitigate toxicity risks and anticipate pharmacological effects beyond the parent compound. Here, we explore high-confidence in vitro human bioactivity annotations from the public domain associated with a curated data set of drug-metabolite pairs. Our findings show that 26% of all drug-metabolite-target combinations contain metabolites with retained or increased bioactivity relative to their parent drugs, showing a high proportion of active metabolites with potencies below 100 nM. Moreover, these active metabolites display similar predicted physicochemical and ADME properties to their parent drugs, indicating that their biological activity may be preserved in vivo. Two key features of the identified active metabolites are noted: (1) the metabolite is more likely to be formed via an O-demethylation or an N-demethylation at a tertiary nitrogen, and (2) these drug-metabolite pairs are frequently active against the membrane receptor target class. To support further research in this area, we make our data compilation, including pairs with identified active metabolites and recorded bioactivity data, openly available.
7 July 2026
Read appraisal →European journal of medicinal chemistry
Accurate and interpretable ADMET prediction: Integrating structural, geometric, and global molecular context representations
Predicting absorption, distribution, metabolism, excretion, and toxicity (ADMET) profiles early in drug discovery is essential to avoid late-stage failures. However, most current computational models depend on single-view molecular representations. This restricts them from capturing the complex interplay between 2D topology, 3D conformations, and specific functional groups that collectively drive structure-pharmacokinetic relationships. Here, we present LGSM, a comprehensive deep learning framework for integrative molecular property prediction. By fusing structural (2D topology), geometric (3D conformational descriptors), and sequence-derived semantic representations, LGSM provides accurate ADMET predictions. Specifically, a language model-based encoding of SMILES sequences is utilized to capture the global molecular context, seamlessly complementing local functional group characteristics. By fusing these different levels of information, our architecture accounts for both local chemical environments and broader spatial relationships. We evaluated this approach across four public ADMET datasets. To mimic real-world drug design scenarios where new chemotypes are constantly explored, we applied a strict scaffold split. Under these conditions, LGSM consistently outperformed standard computational baselines. Beyond predictive accuracy, we prioritized chemical interpretability. By combining the model's internal analysis with molecular docking, we show that LGSM successfully flags key substructures responsible for metabolic liabilities or transporter recognition. This transparency makes the framework a practical tool to guide rational lead optimization in medicinal chemistry.
6 July 2026
Read appraisal →Journal of chemical information and modeling
Generative Artificial Intelligence Optimization of Albumin Binders: Coumarin and Fatty Acid Derivatives
Previously, we reported a dual combination based on 4-hydroxycoumarin and dodecanedioic acid that could synergistically bind to human serum albumin (HSA). However, optimizing this combination remains challenging and could often be guided by empirical selection and extensive experimental screening, which may limit the chemical diversity and suboptimal affinity. In this study, we established a systematic artificial intelligence framework that integrates computational optimization with wet-lab synthesis and experimental validation, enabling improvement of the dual combination while preserving the core chemotypes. We first trained a machine learning classifier on curated HSA binding data and used it as an external scoring function to guide reinforcement learning-driven scaffold decoration with LibINVENT, enabling goal-directed generation of coumarin derivatives and fatty acid derivatives. Candidate molecules were prioritized through multiparameter filtering and diversity-aware selection, followed by synthesis and experimental validation using surface plasmon resonance. The optimized representatives show nanomolar HSA binding and enhanced affinity compared to the original ligands. Molecular docking and molecular dynamics simulations further provide a mechanistic rationale for the affinity improvements by revealing additional stabilizing interactions and more favorable binding energetics at the corresponding HSA sites. Besides, the optimized coumarin derivative (CD1) is a warfarin-derived coumarin analogue, yet it did not show detectable anticoagulant activity in an acute clotting time assay, whereas warfarin did. Overall, this work demonstrates a practical AI-guided route to expand chemical diversity and improve affinity for a synergistic HSA binding combination.
23 June 2026
Read appraisal →Journal of chemical information and modeling
A Reproducible Hierarchical Virtual Screening Framework Integrating Scaffold-Aware Machine Learning, Ensemble Docking, and Molecular Dynamics: Application to IDO1
Indoleamine 2,3-dioxygenase 1 (IDO1) is a heme-containing enzyme implicated in cancer immune escape and remains an attractive therapeutic target despite recent clinical setbacks. We report a fully reproducible hierarchical virtual screening framework integrating scaffold-aware machine learning, ensemble docking, consensus scoring, and molecular dynamics simulations for robust prioritization of IDO1 inhibitors. A curated ChEMBL data set of IDO1 inhibitors was subjected to strict standardization, duplicate removal, and activity binarization at pChEMBL ≥6. Models were trained using scaffold-based splitting and nested cross-validation to prevent chemical series leakage. An ensemble of Random Forest, XGBoost and SVM classifiers achieved balanced predictive performance (ROC-AUC ≈0.88-0.89) with applicability domain filtering to ensure reliability. Prospective screening of FDA-approved drugs yielded 39 compounds within the applicability domain predicted as active. These were further evaluated through ensemble docking against multiple IDO1 crystal structures using GNINA with CNN rescoring. Consensus strategies were systematically benchmarked, demonstrating that best-Z-score aggregation outperformed mean, rank-based, and weighted methods in enrichment factor (EF) metrics. Two top-ranked candidates were subjected to 300 ns molecular dynamics simulations, revealing stable binding modes and persistent interactions with key catalytic residues. This study demonstrates that hierarchical integration of scaffold-aware machine learning with structure-based ensemble strategies enhances robustness and reduces false positives in virtual screening campaigns. The proposed workflow is generalizable and supports reproducible candidate prioritization in computational drug discovery. The complete implementation, including data processing, model training, and analysis steps, is provided as a fully executable Jupyter notebook available at https://github.com/rocco-b/IDO1-inhibitors-ML-and-docking-data.
23 June 2026
Read appraisal →Science advances
Discovery of TYR inhibitors from de novo molecular generation to dual-track lead optimization: 'Competition' between AI and chemists.
This study introduces a unified framework combining artificial intelligence (AI)-directed de novo molecular generation with dual-track lead optimization-comprising expert-guided strategies and AI-driven pathways-to discover tyrosinase (TYR) inhibitors for hyperpigmentation disorders. Using a reinforcement learning (RL)-based generative model, the lead compound AI10 was identified. Subsequent optimization followed two parallel routes. The expert-guided approach yielded AI10-m15 as the most potent TYR inhibitor, with notable antipigmentation activity and excellent cellular safety profiles. In contrast, the AI-driven pathway explored broader chemical spaces, generating unconventional chemotypes, exemplified by the potent TYR inhibitor AI10-a2, highlighting AI's capacity to uncover nonintuitive activity cliffs despite greater output variability. Systematic comparison revealed that the AI model offers exploratory diversity, whereas expert-guided optimization provides predictable improvements in activity and developability. In summary, starting from an AI-generated lead and subsequently integrating both expert-guided and AI-driven structural optimization strategies, these findings further underscore that combining AI technologies with experts' medicinal chemistry insights can substantially accelerate the discovery of viable candidate compounds.
21 June 2026
Read appraisal →Journal of medicinal chemistry
Integrating AI in Medicinal Chemistry for Accelerated Drug Discovery: A Comprehensive SAR (CSAR) Optimization Strategy and Discovery of Potent ALDH3A1 Inhibitors
Developing potent, selective small-molecule inhibitors remains a major challenge in drug discovery. ALDH3A1, a detoxifying aldehyde dehydrogenase isoform implicated in cancer and neurodegeneration, is a promising yet underexplored therapeutic target. To accelerate inhibitor optimization, we developed an AI-guided, reaction-based hit-to-lead workflow combining sequential reaction enumeration, pharmacophore-informed docking, and predictive modeling to support scalable SAR expansion. Applied to ALDH3A1, two rounds of enumeration using Enamine building blocks generated about 250,000 virtual analogues. Combined deep learning and docking-based triage prioritized 150 compounds for synthesis, leading to a roughly 1,000-fold improvement in biochemical potency from 1.41 μM to 1 nM for NCATS-SM0707, together with 4 nM cellular activity for NCATS-SM0708. Crucially, this methodology can be expanded by applying various chemical reactions at different positions. This study highlights CSAR as a scalable, generalizable complementary strategy to accelerate hit optimization through reaction-based enumeration and AI-guided prioritization of larger synthetically accessible chemical space.
13 June 2026
Read appraisal →Journal of computer-aided molecular design
Target-aware molecule SMILES generation using a large language model with retrieval-augmented generation, multi-turn memory, and a predictive model
Given the vastness of chemical space and the cost and time requirements of high-throughput screening, resource-efficient computational strategies are needed to prioritize candidate drug molecules. Here, we evaluated whether an open-weight large language model (LLM) augmented with retrieval-augmented generation (RAG), multi-turn memory, and a pretrained drug-target interaction predictor can generate target-aware molecules without task-specific retraining. Protein-ligand-pKi examples retrieved from BindingDB, Davis, and KIBA were used as contextual guidance, while DeepPurpose provided the optimization signal during iterative SMILES refinement. Our approach produced a statistically significant increase in predicted pKi across successive multi-turn memory iterations, indicating that retrieval- and memory-guided refinement may improve target-conditioned molecular generation. The proposed framework also showed favorable molecular generation characteristics, with novelty reaching 100%, diversity up to 0.882, and uniqueness up to 1.0, suggesting that retrieval did not reduce the process to simple reproduction of known ligands but instead supported exploration of new regions of chemical space. In addition, the approach yielded supportive docking results consistent with the generation of chemically relevant candidate ligands. However, the increase in predicted affinity was accompanied by trade-offs in molecular quality, including reduced validity and drug-likeness in later iterations. Taken together, these findings suggest that this framework provides a flexible and comparatively resource-efficient strategy for target-aware de novo molecular design, while further multi-objective optimization and independent validation remain necessary.
31 May 2026
Read appraisal →Journal of chemical information and modeling
Integrating QSAR-Machine Learning, Biochemical Assays, and Molecular Dynamics for the Discovery of JAK2 Inhibitors in Cervical Cancer
Cervical cancer remains a major global health challenge, where dysregulated JAK2 signaling constitutes a key molecular driver. Nevertheless, selective small-molecule JAK2 inhibitors for HPV-positive cervical cancer are still limited. Here, we integrated biochemical assays, QSAR-machine learning, and molecular dynamics simulations to identify potent JAK2 inhibitors. A series of naphthalene-based derivatives, including hydroxynaphthalenamide and phosphorylated dihydronaphthylamide analogs, were evaluated for cytotoxicity in HeLa cells and JAK2 kinase inhibition. Several compounds exhibited selective cytotoxicity with minimal activity toward normal fibroblasts, among which 2q and 2s showed low-nanomolar JAK2 inhibition and strong apoptosis induction through suppression of the JAK2/STAT3/STAT5 pro-tumorigenic signaling pathway. To accelerate hit identification, a QSAR-machine learning (QSAR-ML) framework was employed to prioritize 13 newly designed derivatives. Among three ensemble boosting models, the Categorical Boosting (CB) model demonstrated the strongest predictive capability, achieving a high R2 of 0.955 for the training set and a low RMSE of 0.156 for the test set. This model successfully identified five active candidates with strong prediction-experiment agreement (MAPE = 2.6-14.4%), with D4 and D13 meeting drug-likeness criteria and displaying potent nanomolar JAK2 inhibition. Finally, 1-μs molecular dynamics simulations revealed that hydrophobic contacts and hydrogen bonding cooperatively stabilize these inhibitors within the JAK2 ATP-binding pocket. Collectively, these findings establish a QSAR-ML-guided strategy for accelerating JAK2 inhibitor discovery and highlight naphthalene-based scaffolds as promising leads for targeted cervical cancer therapy.
27 May 2026
Read appraisal →ChemMedChem
A 24-Day MicroCycle Journey to Interleukin-17 Inhibitors: From Library Design to Central Core Prioritization and Capping Group Identification
We herein report an application of the MicroCycle platform, which leverages advanced machine learning models, automation, and miniaturization of processes to accelerate the exploration of chemical space through libraries. This approach was applied to a novel thiazole-based series targeting interleukin-17, enabling rapid prioritization of central cores and identification of compelling new capping groups, thereby guiding further optimization efforts.
27 May 2026
Read appraisal →Journal of chemical theory and computation
MolSculptor: An Adaptive Diffusion-Evolution Framework Enabling Generative Drug Design for Multitarget Affinity and Selectivity
The rational design of molecules with tailored activity/selectivity across multiple protein targets is crucial for developing therapies for complex diseases like cancer, yet it remains a formidable challenge. While deep generative models show immense promise, their application to these tasks faces fundamental challenges, as they struggle to incorporate the structural information on multiple distinct protein pockets and require vast multitarget data sets or specialized expert knowledge that are rarely available. Here we introduce MolSculptor, an adaptive diffusion-evolution framework designed to generate inhibitors for any combination of on- and off-targets, circumventing the need for target-specific training data or prior expert knowledge. MolSculptor unifies both de novo design and lead optimization, provides a versatile workflow applicable to different stages of drug discovery, and allows for direct conditioning on key drug-like properties. It integrates a 3D-aware surrogate model to enable flexible guidance for any set of specified on- and off-targets. Furthermore, MolSculptor employs an active learning protocol to adaptively refine this guidance, ensuring high performance even in data-scarce scenarios. We demonstrate MolSculptor on a series of challenging multitarget and selective inhibitor design tasks, where it significantly outperforms state-of-the-art methods in generating high-quality candidates that satisfy all complex constraints. Notably, many of the generated molecules exhibit predicted affinity profiles superior to those of experimentally validated references. Using MolSculptor, we successfully designed and synthesized a novel and potent dual-target inhibitor for castration-resistant prostate cancer (CRPC), whose inhibitory activity was confirmed through wet-lab validation. MolSculptor provides a powerful and generalizable paradigm for designing ligands with complex, multitarget activity profiles, paving the way for data-efficient solutions to complex therapeutic problems.
27 May 2026
Read appraisal →Journal of computer-aided molecular design
ADMET-vault: an interactive framework for real-time ADMET prediction and molecular optimization
Drug discovery pipelines face persistent challenges, as candidates that show promising results in early laboratory studies are found ineffective in clinical trials-often because their absorption, distribution, metabolism, excretion, and toxicity (ADMET) profiles were not well understood early enough. In this study, an interactive machine-learning based platform, ADMET-Vault, is presented, which brings ADMET prediction directly into the molecular design process, allowing researchers to assess and optimize drug candidates during the molecule design phase itself. ADMET-Vault integrates multiple molecular representations including physicochemical descriptors, molecular fingerprints, and graph neural network embeddings. The platform evaluated 12 benchmark datasets from Therapeutics Data Commons using scaffold-based validation. ADMET-Vault demonstrated consistent predictive performance across multiple ADMET endpoints. Strong results were observed for clearance related properties, intestinal absorption, hepatotoxicity, and aqueous solubility. Moreover, regression showed mean absolute error value that ranged from 0.28 (Caco-2 permeability, expressed as log cm/s) to 7.68 (plasma protein binding rate, expressed as % bound), with Spearman's rank correlation coefficients between 0.44 and 0.66. Classification tasks showed area under the receiver operating characteristic curve values ranged between 0.87 and 0.99. ADMET-Vault's main contribution is not just in its predictive accuracy. It combines reliable machine learning models with a live molecular editor. This allows researchers to see how a structural change impacts drug-like properties instantly. This interactive design-predict-optimize loop fills a long-standing gap in early-stage drug discovery workflows.
27 May 2026
Read appraisal →Zeitschrift fur Naturforschung. C, Journal of biosciences
Exploring the potency of 6-bromo-3-methylquinoline analogues as prostaglandin F2α inhibitors
Prostaglandin F2α (PGF2α) is associated with preterm labor and preterm birth. PGF2α inhibitors have thus proven to be a promising target in the development of lead compounds to prevent preterm birth. In this work, Quantitative Structural Activity Relationship (QSAR) was implemented on a dataset of 77 compounds of 6-bromo-3-methylquinoline analogues using statistical approach and random selection in the QSARINS software. The Genetic Algorithm-Multiple Linear Regression (GA-MLR) approach was used to predict the best model (R 2 = 0.8943 and Q 2 LOO = 0.8836). The inclusion of descriptors FNSA-2 and WV.mass resulted in a well-fitted and highly predictable model. Artificial neural network (ANN) analysis was also carried out to validate the model effectiveness. Twenty eight new molecules with better predicted biological activity (pIC50) were designed. The binding energy from the docking study of seven compounds have shown higher binding activity than P10 into prostaglandin F synthase protein (PDB ID: 2F38). The stability of protein-ligand complex was further validated by 100 ns molecular dynamics simulation and MM-PBSA binding free energy. DFT and ADME-toxicity analysis also confirmed their drug-likeness properties. Collectively, our findings highlight novel quinoline derivatives as promising lead candidates, warranting further validation through collaborative in vitro and in vivo studies.
27 May 2026
Read appraisal →Journal of computer-aided molecular design
Dynamic stability-driven machine learning improves binding pose identification on human serum albumin.
Accurate identification of ligand binding modes on human serum albumin (HSA) is important for understanding drug-protein interactions that influence pharmacokinetics, including site-specific competition and protein-binding displacement. However, because HSA is a large, multisite binding protein that permits diverse binding configurations, conventional molecular docking often struggles to distinguish crystal-like poses from nonspecific surface-binding poses using static scoring functions alone. Here, we developed a machine-learning framework for HSA binding-pose identification that incorporates dynamic stability information derived from short molecular dynamics (MD) simulations. Time-dependent interaction patterns observed during MD trajectories, including contact persistence, distance fluctuations, and interaction energy-related descriptors, were quantified as interpretable features and integrated with docking scores. Using Leave-One-PDB-Out cross-validation on 31 HSA-ligand crystal structures, models incorporating MD-derived features significantly outperformed the docking score-only model under both pose-level (root-mean-square deviation [RMSD] ≤ 2.5 Å) and site-level (RMSD ≤ 5.0 Å) criteria (Holm-corrected p < 0.001). External validation using a temporally independent nateglinide-HSA complex further supported the utility of the framework for discriminating plausible binding modes within the HSA setting. Notably, informative dynamic features were obtained from 5 ns MD simulations, indicating that early-stage dynamic behavior is sufficient to improve pose discrimination. These findings show that short MD simulations can provide practically useful and interpretable features for machine-learning-based binding-pose identification on HSA. While broader validation will be required to establish applicability beyond HSA, the proposed framework offers a practical strategy for improving binding-mode analysis in pharmacokinetically relevant HSA studies.
15 May 2026
Read appraisal →Journal of molecular modeling
Exploring human GABA transporter 3 binders for epilepsy through quantum reactivity analysis, molecular dynamics, machine learning prediction, and network pharmacology
CONTEXT: Drug-resistant epilepsy (DRE) remains a major therapeutic challenge, affecting millions of patients globally who do not respond to conventional antiepileptic medications. Recent evidence highlights GABA transporter 3 (GAT3), a glial subtype responsible for regulating extracellular GABA levels, as a promising molecular target for seizure control. Despite its therapeutic potential, selective and potent GAT3 inhibitors remain largely unexplored. This study aims to identify novel GAT3 inhibitors through an integrative computational strategy combining quantum chemical, molecular dynamics, and machine learning approaches to accelerate antiepileptic drug discovery. The cryo-EM-resolved human GAT3-SNAP-5114 complex structure was used for virtual screening, leading to the identification of three highest-ranking compounds, viz. Comp_1, Comp_2, and Comp_3, from the diverse chemically optimized screening library. Analysis of their quantum chemistry profiles indicated HOMO-LUMO energy gap values of 4.14 eV, 4.21 eV, and 3.26 eV, respectively, suggesting them to be maximally reactive with favorable electronic properties for interacting with biological molecules. Molecular dynamics simulations were performed in a fully hydrated POPC lipid bilayer using the AMBER force field, providing a physiologically relevant environment for assessing ligand binding and transporter dynamic, revealed stable binding modes, with the highest structural integrity or least deviations in the structure with Comp_2, having RMSD values ranging from 2.5 to 3.0 Å, with strong hydrogen bonding, while the binding mode was more flexible with strong anchorage for Comp_3, with its mode involving diverse interactions in the complex. Principal component and free energy landscape analyses identified energetically favorable, ligand-induced conformational states. Machine learning-based pIC₅₀ predictions using the AdaBoost regressor yielded values of 7.95 (Comp_1), 7.98 (Comp_2), and 7.63 (Comp_3), all comparable or superior to the reference inhibitor (7.89). A novel approach for network pharmacology integrating detailed drug-gene interaction profiling was additionally used to improve systems-level understanding of the discovered compound. These findings validate the potential of Comp_1-3 as lead GAT3-targeting antiepileptic agents. METHODS: A docking-based virtual screening of a structurally diverse compound library was performed against the cryo-EM structure of human GAT3 in complex with the reference inhibitor SNAP-5114. The top-scoring ligands were subjected to density functional theory (DFT) analysis to assess electronic properties such as HOMO-LUMO energy gaps, reflecting potential binding reactivity. Lead candidates were refined, docked, and evaluated through 500 ns molecular dynamics (MD) simulations, analyzing structural stability via RMSD, RMSF, hydrogen bonding, and interaction network profiles. Principal component analysis and free energy landscape evaluations were used to explore ligand-induced conformational dynamics. Furthermore, machine learning (ML) models-particularly the AdaBoost algorithm-were employed to predict pIC₅₀ values, confirming bioactivity trends. Finally, network pharmacology analysis was conducted to elucidate gene targets and pathway interconnections, revealing the multi-pathway involvement of top candidates in seizure regulation. This integrative computational framework underscores the potential of GAT3-focused modelling and ML-guided screening for rational antiepileptic drug design.
14 May 2026
Read appraisal →Journal of chemical information and modeling
A Transformer for Reaction-Aware Compound Explorations with GFlowNet in QSAR-Guided Molecular Design
In drug discovery tasks, achieving a balance between high biological activity toward therapeutic targets and synthetic chemical feasibility is critically important. While the recently proposed deep learning-based molecular generation models have enabled explorations of vast chemical spaces, most existing approaches do not consider synthetic routes for generated compounds. To address this issue, TRACE-GFN is proposed for molecular optimization; this method incorporates chemical reaction pathways into a quantitative structure-activity relationship (QSAR)-guided molecular design procedure. The method integrates a transformer model to explicitly learn chemical reactions with a generative flow network (GFlowNet) that efficiently samples diverse candidates. In benchmark experiments involving dopamine receptor D2 (DRD2), AKT serine/threonine kinase 1 (AKT1), and C-X-C motif chemokine receptor 4 (CXCR4), TRACE-GFN demonstrated the ability to identify compounds with high QSAR values while maintaining strong diversity, outperforming the existing molecular generation models. These results demonstrate that the proposed model can efficiently explore promising compounds while accounting for real-world chemical reactions. The source code is publicly available under an MIT license at https://github.com/sekijima-lab/TRACE-GFN.
13 May 2026
Read appraisal →Journal of chemical information and modeling
Artificial Intelligence for Predicting Small-Molecule Bioactive Conformations
Precise prediction of bioactive conformations represents a central challenge in drug discovery. Traditional computational methods often rely on the assumption that the global minimum energy state corresponds to the bioactive conformation─a hypothesis with inherent limitations. Data-driven artificial intelligence (AI) models have revolutionized research methodologies in this domain. By learning the implicit probability distributions of conformations, AI demonstrates the potential to achieve high-precision predictions of bioactive structures. This review systematically surveys technical advancements ranging from geometric regression to generative and conditional generative models, analyzing their effectiveness in benchmarks and de novo design experiments. We deeply explore critical challenges, including data sparsity, sampling efficiency, generation capability, and interpretability. Finally, we provide an outlook on bridging the gap from "structural prediction" to "functional design" through the construction of dynamic data sets, the development of efficient algorithms, and the implementation of human-in-the-loop collaborative decision-making systems.
13 May 2026
Read appraisal →Journal of chemical information and modeling
Explainability Methods from Machine Learning Detect Important Drugs' Atoms in Drug-Target Interactions
Predicting drug-target interactions (DTI) with graph neural networks (GNNs) is hindered by their lack of interpretability. To address this, we benchmark four explainable artificial intelligence (XAI) attribution methods on GNN models trained for kinase and G-protein-coupled receptors (GPCR) targets. We assess the methods' consistency through atom-level intersection over union (IoU) and validate their biological relevance by mapping attributed atoms to three-dimensional (3D) protein-ligand structures. While consistency across methods was modest, consensus attributions were highly enriched for atoms directly contacting the binding pocket─up to 76% within 2 Å in the kinase-inhibitor complexes. Notably, these attributed atoms were frequently found contacting experimentally important regulatory residues such as those in the DFG motif. This indicates that XAI methods, despite their disagreements, can identify chemically meaningful ligand features, providing a foundation for developing more interpretable GNNs in drug discovery.
2 May 2026
Read appraisal →Journal of chemical information and modeling
A Reinforcement Learning-Guided Genetic Algorithm Integrating Medicinal Chemistry-Inspired Molecular Transformations
Achieving optimal target activity while maintaining synthetic accessibility and drug-likeness represents a major challenge in computational drug discovery. Existing de novo generative models often yield chemically invalid or synthetically intractable structures and struggle to optimize multiple objectives simultaneously. Here, we introduce ALCHIMIA, an interpretable hybrid framework combining reinforcement learning (RL) and a genetic algorithm (GA), built based on a vocabulary of 33 medicinal chemistry-inspired molecular transformations. The RL component trains a policy network to prioritize transformation sequences that improve synthetic accessibility (SA) and the quantitative estimate of drug-likeness (QED) scores, embedding these constraints directly into molecular generation. The GA component applies the learned policy as a mutational operator within population-based optimization guided by molecular docking, enabling the exploration of diverse chemical lineages while converging toward high-affinity ligands. ALCHIMIA was applied to two different pharmacologically relevant targets: human Cannabinoid Receptor 2 (CB2R) and human Sigma nonopioid intracellular Receptor 1 (S1R). We considered three different scenarios: (i) unconstrained hit identification; (ii) scaffold-constrained lead optimization; and (iii) design of dual modulators. The framework generated chemically valid molecules with QED and SA scores comparable to or better than those obtained with random baselines and selected de novo design methods. By codifying typical medicinal chemistry actions as learnable transformations and coupling multiobjective optimization with GA-based diversity maintenance, ALCHIMIA, freely available as a GitHub repository (https://github.com/alberdom88/ALCHIMIA), provides a practical, interpretable, and scalable framework for molecular de novo design.
2 May 2026
Read appraisal →Journal of medicinal chemistry
Machine Learning Approach to Anticancer Activity Prediction of Transition-Metal Complexes Based on a Large-Scale Experimental Database
In this work, we developed a straightforward data-driven approach to predict the cytotoxicity of metal complexes based entirely on their (metal + ligands) composition. To this end, we have manually curated MetalCytoToxDB─a comprehensive experimental database comprising 26,500 IC50 values for 7050 metal complexes against 754 cell lines from 1921 articles. Based on these, machine learning models were created to accurately assign the cytotoxicity class within the ruthenium and iridium subsets. Moreover, external validation of the best-performing model on the unseen data was carried out. The possibility of multimetal predictions was explored, enabling assessment of cytotoxicity among the complexes of metals, for which experimental data are relatively scarce. The interpretability and limitations of the developed models are discussed. Finally, a pipeline for the effective high-throughput computational screening of ruthenium complexes is proposed. The MetalCytoToxDB is available online for AI-assisted exploration at https://biometaldb.streamlit.app/.
25 Apr 2026
Read appraisal →Journal of medicinal chemistry
An Interpretable, Thermodynamics-Based Deep Learning Framework for Predicting and Optimizing Drug Membrane Permeability
Cellular membranes serve as selective barriers, and membrane permeability is crucial for drug pharmacokinetics. While in vitro and in vivo methods exist, predicting and designing membrane permeability remains challenging. We developed a thermodynamics-based deep learning framework to analyze the structure-permeability relationship, based on the concept that interactions between membrane lipids and small molecules influence permeability. We determined the membrane penetration thermodynamics of 8,239 compounds using coarse-grained molecular dynamics simulations and created interpretable graph neural network models to predict and design drug membrane permeability. As a proof-of-concept, we designed a novel nasal-administered melatonin analog, MT-A2, optimized for permeability. Compared with melatonin, MT-A2 showed superior nasal absorption, prolonged brain retention, and enhanced sleep efficacy. Our results provide a promising approach for predicting and designing membrane permeability, aiding in the development of drugs with better pharmacokinetics.
25 Apr 2026
Read appraisal →Journal of chemical information and modeling
AI-MedCraft: A Strategy-Driven AI Platform for Multi-Objective Molecular Design
Small-molecule drug discovery requires balancing potency, selectivity, solubility, safety, and synthetic feasibility, among other interdependent properties. Yet most computational workflows address these properties sequentially or collapse them into fixed weighted scores. Although artificial intelligence has accelerated molecular design, many platforms struggle to handle explicit multiobjective trade-offs. Here, we present AI-MedCraft, a strategy-driven molecular design framework that applies adaptive, Pareto-guided reinforcement learning to optimize multiple objectives concurrently within a unified workflow. When structural information is available, physics-aware scoring is incorporated to support binding-competent designs. In a structure-based benchmark on the solubility rescue of the BTK inhibitor GDC-0834, AI-MedCraft achieves broader Pareto-front coverage than the scalar-reward reinforcement learning framework REINVENT 4 under matched objectives and computational cost. In a second case study, AI-MedCraft redesigns Efavirenz to retain HIV-1 reverse transcriptase engagement while reducing predicted 5-HT2A off-target binding, demonstrating its use in multiconstraint molecular optimization.
14 Apr 2026
Read appraisal →Journal of chemical information and modeling
Compressing Chemistry Reveals Functional Groups
We introduce the first formal large-scale assessment of the utility of traditional chemical functional groups as used in chemical explanations. Our assessment employs a fundamental principle from computational learning theory: a good compression of data should reveal a good explanation. We introduce an unsupervised learning algorithm based on the Minimum Message Length (MML) principle that searches for substructures that compress around three million biologically relevant molecules. We demonstrate that the discovered substructures contain most human-curated functional groups as well as novel larger patterns with more specific functions. We also run our algorithm on 24 specific bioactivity prediction data sets to discover data set-specific functional groups. Fingerprints constructed from data set-specific functional groups are shown to significantly outperform other fingerprint representations, including the MACCS and Morgan fingerprint, when training ridge regression models on bioactivity regression tasks.
14 Apr 2026
Read appraisal →Journal of medicinal chemistry
Rational Design of Broad-Spectrum Non-Nucleoside Reverse Transcriptase Inhibitors via Pharmacophore-Oriented Generative Artificial Intelligence
Guided by the pharmacophore-oriented molecular generation platform PhoreGen, we employed rilpivirine (RPV) as a lead compound to generate 300 structurally diverse analogs that preserve key pharmacophoric features. Subsequent drug-likeness evaluation, molecular docking score, and synthetic feasibility led to the identification of compound No.102 (A19), which displayed potent inhibition activity against WT HIV-1 (EC50 = 3.15 nM) and low cytotoxicity (CC50 > 335 μM). Subsequent structure-activity relationship optimization identified A24, which demonstrated robust activity against clinically relevant drug-resistant HIV-1 mutants (EC50 = 2.07-28.8 nM). Notably, compared to RPV, A24 exhibited significantly reduced cytotoxicity (CC50 = 82.3 μM vs. 3.98 μM) and enhanced potency against the Y188L (EC50 = 28.8 nM vs. 79.4 nM) and F227L + V106A (EC50 = 19.0 nM vs. 81.6 nM) mutants. Pharmacokinetic evaluations revealed that A24 exhibited attenuated CYP enzyme inhibition (IC50 ≥ 1.88 μM), reduced hERG-related toxicity (IC50 = 2.633 μM), and improved metabolic stability (human t1/2 = 39.1 min). Collectively, these favorable properties position A24 as a promising NNRTI for further development.
10 Apr 2026
Read appraisal →Chemical communications (Cambridge, England)
Innovative integration of molecular docking and machine learning for drug discovery: from virtual screening to nanomolar inhibitors
This review highlights our group's systematic approach to integrating molecular docking, pharmacophore modeling, and machine learning methodologies for the rational discovery of bioactive leads. We describe innovative strategies including docking-based data augmentation, ligand-receptor contact fingerprints, genetic algorithm-guided feature selection, and SHAP-based model interpretation. These approaches have enabled the discovery of nanomolar inhibitors against multiple therapeutic targets including STAT3, TTK, LSD-1, and HER2. The presented workflow demonstrates how machine learning (ML) can be synergistically combined with traditional computer-aided drug design methods to achieve efficient scaffold hopping and identify novel chemotypes with potent biological activities.
10 Apr 2026
Read appraisal →