Journal of Cheminformatics最新文献

{"title":"Smiles-based bioactivity prediction through molecular encoder selection and data augmentation","authors":"Ju Hyung Lee,&nbsp;Seongik Choi,&nbsp;Utku Ozbulak,&nbsp;Joris Vankerschaver,&nbsp;Wesley De Neve","doi":"10.1186/s13321-026-01251-0","DOIUrl":"10.1186/s13321-026-01251-0","url":null,"abstract":"<p>Quantitative prediction of inhibitor potency can accelerate early-stage drug discovery. Recently, data-driven approaches have gained widespread interest in drug discovery, as evidenced by a growing number of benchmarking challenges and open competitions. In this context, we developed a machine learning-based methodology that can find the most effective way of predicting IC₅₀ values against ASK1 from SMILES, for <i>“Jump AI(.py) 2025: 3rd AI Drug Discovery Competition”</i>, hosted by the Korea Pharmaceutical and Bio-Pharma Manufacturers Association (KPBMA) on the Dacon platform. Applying our methodology achieved the highest overall predictive performance among all participating teams. Beyond this competition setting, we present a compact SMILES-based modeling workflow comprising (i) a pre-trained encoder, (ii) regression models, (iii) data augmentation, and (iv) hyperparameter tuning. We systematically compared molecular representations from sequence- and graph-based models, including ChemBERTa-2 and MolCLR. Across encoder–regressor combinations, ChemBERTa-77 M-MLM embeddings paired with support vector regression (SVR) yielded the strongest predictive performance. Embedding-level mix-up augmentation and SVR hyperparameter tuning further improved predictive performance. Our findings highlight that careful SMILES preprocessing and encoder selection have a critical influence on IC₅₀ values and provide a reproducible benchmark for single-target bioactivity prediction, thus contributing to a more efficient drug discovery process. </p><p><b>Scientific Contribution</b> In this study, we propose a machine learning methodology for predicting the IC₅₀ values of ASK1 inhibitors from SMILES representations, with a systematic comparison of molecular encoders and regression models. Our results show that the use of suitable encoder-regressor pairs together with embedding-level mix-up augmentation improves model generalizability without requiring SMILES-level augmentation. This strategy would be particularly useful for settings with imbalanced labels or limited data, and could be applied more broadly to IC₅₀ prediction for other kinase inhibitors.</p>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01251-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148393669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MINERVA: a public XAI-powered platform advancing multi-target discovery in Alzheimer’s disease","authors":"Nicola Gambacorta,&nbsp;Daniela Trisciuzzi,&nbsp;Fabrizio Mastrolorito,&nbsp;Francesco Leonetti,&nbsp;Cosimo D. Altomare,&nbsp;Modesto de Candia,&nbsp;Marco Catto,&nbsp;Nicola Amoroso,&nbsp;Fulvio Ciriaco,&nbsp;Orazio Nicolotti","doi":"10.1186/s13321-026-01253-y","DOIUrl":"10.1186/s13321-026-01253-y","url":null,"abstract":"<div><p>Alzheimer’s disease (AD) is a multifactorial neurodegenerative disorder involving a complex interplay of interconnected pharmacological targets, a feature that limits the success of traditional single-target medicinal chemistry approaches. Here, we present MINERVA (Multi-target Interactive Network for Explainable Research and Visualization in Alzheimer’s disease), a public web-based platform designed to support multi-target drug discovery assisted by eXplainable Artificial Intelligence (XAI) strategies. MINERVA integrates large-scale disease-focused high-quality data comprising as many as 70,960 small molecules annotated across 33 <span>AD</span> relevant targets, taken from ChEMBL and CADRO databases. To capture different levels of pharmacological relevance, four distinct pharmacological thresholds (i.e., 10 μM, 1 μM, 100 nM, and 10 nM) were set to enable the parallel exploration of weak to high-affinity ligand spaces. Independent Balanced Random Forest (BRF) classifiers were trained for each AD target threshold combination using an extended core-substituent fingerprint, which ensures robustness against class imbalance and chemical heterogeneity. MINERVA incorporates a probability binning based domain of applicability (DoA) to quantify prediction reliability and a SHAP-based explainability framework to fairly map fragment-level contributions directly onto chemical structures. MINERVA is freely accessible at https://prometheus.farmacia.uniba.it/minerva/.</p><p><b>Scientific contribution</b></p><p>Herein we introduce MINERVA, the first freely accessible platform that combines large-scale AD-related data curation, multi-target prediction, and multi-threshold bioactivity modeling within a fully explainable and user-friendly environment. By enabling transparent, ligand-based exploration of chemical space across multiple AD pathways, MINERVA provides a unique and practical resource for accelerating multi-target drug discovery in the neurodegenerative research area.</p></div>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01253-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148393766","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multimodal feature fusion for molecular property classification","authors":"Jing Liu,&nbsp;Li Xue,&nbsp;Yin Wang,&nbsp;Qiaorong Wu,&nbsp;Wenwei Tao,&nbsp;Yiwei Wang,&nbsp;Jianming Wu,&nbsp;Jiesi Luo","doi":"10.1186/s13321-026-01256-9","DOIUrl":"10.1186/s13321-026-01256-9","url":null,"abstract":"<div><p>Accurate molecular property prediction is a cornerstone of modern chemical science, driving progress in drug discovery, materials design, and environmental research. Yet, most existing models remain unimodal, while multimodal approaches often rely on simple aggregation, leaving much of the complementary chemical information underexploited. In this work, we present a multimodal feature fusion framework that unites the strengths of deep chemical language processing (CLP) models and molecular fingerprints, integrating sequential and structural representations for more comprehensive molecular characterization. Unlike previous heuristic combinations, our framework systematically investigates the principles of effective cross-modal fusion. We benchmark ten CLP architectures and eight fingerprint types through exhaustive combinatorial search to identify the most synergistic configurations. This exploration shows that aggregating multiple models does not necessarily improve performance; instead, successful fusion requires data-aware design guided by feature integration and complementarity. The proposed strategy effectively couples sequential features learned from SMILES with structural information captured by molecular fingerprints, resulting in a coherent and chemically interpretable molecular representation. Evaluated across 60 datasets from MoleculeNet and TOXRIC, our fusion models deliver consistent and substantial gains over state-of-the-art baselines. Beyond outperforming existing architectures, this work provides conceptual insights and practical guidelines for multimodal fusion in molecular property prediction, highlighting the importance of efficient fusion strategies in building robust and generalizable molecular models.</p><p><b>Scientific contribution</b></p><p>This study provides a large-scale empirical evaluation of multimodal feature fusion for molecular property classification by systematically integrating SMILES-based chemical language representations with fingerprint-based structural descriptors across 60 benchmark datasets. The framework introduces a data-aware combinatorial fusion strategy to identify task-specific complementary feature combinations, improving robustness and interpretability compared with unimodal models and baseline approaches. The results clarify how sequence-based and structure-based molecular representations complement each other, providing practical guidance for designing multimodal models in cheminformatics.</p></div>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01256-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148379817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"P2MAT: A machine learning (ML) driven software for Property Prediction of MATerial.","authors":"Methun Kamruzzaman, Alexander Landera, Nalini Menon, Mark D Allendorf, Kunal Poorey","doi":"10.1186/s13321-026-01247-w","DOIUrl":"10.1186/s13321-026-01247-w","url":null,"abstract":"<p><p>Accurate prediction of melting points for pure molecules remains a significant challenge in predictive chemistry, with implications across various scientific fields, including materials science, drug discovery, and separations chemistry. Traditional methods, such as group contribution (GC) techniques, have shown limited success due to the complex relationship between molecular structure and melting point. In this study, we present a data-driven machine learning (ML) approach to predict the melting points of organic compounds, leveraging both 2D and 3D molecular descriptors. Our results indicate that ML models can significantly improve melting-point predictions, providing a robust tool for the scientific community. Scientific contributionOur detailed analysis on melting point prediction, along with SHAP explainability, reveals the top influencing features for the prediction. The P2MAT application we developed as part of this study can predict both melting and boiling points from a SMILES string. P2MAT is available as an easy-to-install, user-friendly GUI for maximum outreach to the scientific community. Our benchmark analysis demonstrates the excellence of our method for predicting melting points.</p>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":" ","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148381430","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Computational design of low-volatility lubricants for space using interpretable machine learning","authors":"Daniel Miliate,&nbsp;Ashlie Martini","doi":"10.1186/s13321-026-01250-1","DOIUrl":"10.1186/s13321-026-01250-1","url":null,"abstract":"<p>The function and lifetime of moving mechanical assemblies (MMAs) in space depend on the properties of lubricants. MMAs that experience high speeds or high cycles require liquid-based lubricants due to their ability to reflow to the point of contact. However, only a few liquid-based lubricants have vapor pressures low enough for the vacuum conditions of space, each of which has limitations that add constraints to MMA designs. This work introduces a data-driven machine learning (ML) approach to predicting vapor pressure, enabling virtual screening and discovery of new space-suitable liquid lubricants. The ML models are trained with data from both high-throughput molecular dynamics simulations and experimental databases. The models are designed to prioritize interpretability, enabling the relationships between chemical structure and vapor pressure to be identified. Based on these insights, several candidate molecules are proposed that may have promise for future space lubricant applications in MMAs.</p><p>This work develops interpretable machine learning models for vapor pressure of low volatility molecules, identifying the structural features that drive volatility. The framework is accurate in the ultra-low-volatility regime where existing methods are unreliable, enabling virtual screening of candidate space lubricants. All code and data are publicly available, and the approach generalizes to materials discovery in other extreme environments.</p>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01250-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148373876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"OpenStats: how to combine statistics and research data management (RDM) to leverage efficient scientific data analysis by guided statistics","authors":"Konrad Krämer,&nbsp;Pierre Tremouilhac,&nbsp;Fabian Mauz,&nbsp;Christoph Grathwol,&nbsp;Nicole Jung,&nbsp;Stefan Bräse","doi":"10.1186/s13321-026-01241-2","DOIUrl":"10.1186/s13321-026-01241-2","url":null,"abstract":"<div><p>We developed OpenStats, a user-friendly web application that brings the power of the R language to researchers through a high-level interface and broad support for statistical methods such as t-tests and ANOVA. OpenStats was integrated into our electronic lab notebook Chemotion ELN via its third-party API, enabling direct data exchange with it. A pivotal feature of OpenStats is its ability to record each analysis step in a structured history. This record allows users to retrace their work and enables automatic replay of the entire analysis, promoting reproducibility and long-term data integrity. In the modern research landscape, Research Data Management (RDM) tools like electronic lab notebooks (ELNs) are crucial for generating reproducible, repeatable, and transparently documented data. However, integrating statistical analysis tools into RDM systems remained a challenge. We demonstrated how OpenStats can be seamlessly linked to Research Data Management (RDM) systems, using Chemotion ELN as a reference implementation, by our application to a standard dose–response assay scenario. This linkage allows the integration of statistical data analysis in the form of a closed, traceable workflow into the RDM world. It therefore complements systems that are aiming for reproducible and standardized workflows in the scientific environment.</p></div>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01241-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148366348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unified heterogeneity-aware benchmark of drug synergy prediction: a cross-study analysis of traditional machine learning and graph deep learning models","authors":"Yingjuan Cheng,&nbsp;Qing Ye,&nbsp;Linlong Jiang,&nbsp;Yu Kang,&nbsp;Chang-Yu Hsieh,&nbsp;Tingjun Hou","doi":"10.1186/s13321-026-01242-1","DOIUrl":"10.1186/s13321-026-01242-1","url":null,"abstract":"<div><p>Drug synergy prediction holds great promise in accelerating combination therapy development and improving treatment efficacy in cancer and other complex diseases. However, progress in this area is hindered by considerable heterogeneity across experimental datasets, including variability in the number of drug combinations, inconsistencies in synergy scoring methodologies, and differences in data quality. Here, we present the first comprehensive benchmarking framework specifically designed to accommodate inter-dataset heterogeneity. This framework integrates 13 independent datasets encompassing 454,794 retained drug combination–cell line entries, 4247 drugs, and 187 cell lines, all of which are cancer cell lines. Our comparative evaluation of seven computational models for drug synergy prediction reveals that model performance strongly depends on both the scale and quality of the datasets. The graph model JointSyn performed favorably on datasets with larger numbers (&gt; 10,000) of retained drug combinations, while the traditional random forest model performed competitively on smaller-scale datasets. We also observe that the ZIP scoring metric yields the highest accuracy in large-scale data, whereas HSA is more effective in sparse-data scenarios. However, different synergy metrics show significant variability in performance across datasets, suggesting that different synergy metrics capture distinct aspects of drug interactions, and the choice of metric can substantially affect model evaluation and cross‑dataset consistency. Furthermore, we find that well-designed small datasets can match or even surpass the performance of larger benchmarks, suggesting that different metrics are applicable to different datasets/testing scenarios. Our benchmark provides a robust foundation for fair model evaluation and paves the way for the development of more generalizable and preclinically relevant drug synergy prediction methods.</p><h3>Graphical abstract</h3>\u0000<div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01242-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148359119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Count your bits: fingerprint benchmarking to assess broad chemical space representation","authors":"Florian Huber,&nbsp;Julian Pollmann","doi":"10.1186/s13321-026-01245-y","DOIUrl":"10.1186/s13321-026-01245-y","url":null,"abstract":"<div><p>Quantifying molecular similarity is a cornerstone of cheminformatics, underpinning applications from virtual screening and nearest-neighbor search to chemical space visualization and the evaluation of machine-learning predictions. Yet, despite widespread use of Tanimoto comparisons of 2D fingerprints, the practical behavior of these similarity measures depends strongly on the fingerprint type, representation (binary vs. count), and whether fingerprints are folded into fixed-length vectors. In this work, our main goal is to provide a broad and systematic comparison of common fingerprint types, including dictionary-based, circular (Morgan/FCFP), path-based (RDKit), topological-distance-based (Atom Pair), hybrid distance-encoded (MAP4), torsion, LINGO, and Avalon, together with several representation variants. We benchmark these fingerprints across multiple large and heterogeneous datasets using complementary evaluation tasks that probe specificity, score distributions, compound-size dependence, top-k ranking agreement, and compare fingerprint similarities to a graph-based reference. Across benchmarks, count and often log-count variants generally improve specificity and structural alignment, while folding-induced bit collisions can strongly distort similarities for high-occupancy fingerprints, making unfolded variants particularly important for RDKit and often necessary for MAP4 on heterogeneous datasets. To support reproducible benchmarking and future extensions, we also introduce the open-source Python library chemap, which provides unified computation of folded, unfolded, and frequency-folded fingerprints together with optimized similarity calculations.</p><p><b>Scientific contribution</b></p><p>We introduce a multi-criteria benchmarking framework for molecular fingerprints that goes beyond classical virtual-screening retrieval tests by evaluating specificity (fingerprint duplicates), score behavior (including compound-size dependence), ranking agreement, and neighborhood structure on large, heterogeneous small-molecule datasets. Our results reveal that common default settings, especially folding for high-occupancy fingerprints, can introduce severe bit-collision artifacts, and that count (often log-count) and unfolded variants substantially improve specificity and agreement with structure-based references. We release the open-source Python library chemap to standardize these fingerprint variants and enable reproducible, extensible benchmarking for future fingerprint development.</p></div>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01245-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148349941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sampling out-of-distribution chemical spaces via Bayesian flow","authors":"Nianze Tao,&nbsp;Minori Abe","doi":"10.1186/s13321-026-01248-9","DOIUrl":"10.1186/s13321-026-01248-9","url":null,"abstract":"<div><p>Generating novel molecules with higher properties than the training space, namely the out-of-distribution generation, is important for <i>de novo</i> drug design. However, it is not easy for distribution learning-based models, for example diffusion models, to solve this challenge as these methods are designed to fit the distribution of training data as close as possible. In this paper, we show that Bayesian flow network, especially ChemBFN model, is capable of intrinsically generating high quality out-of-distribution samples that meet several scenarios. A reinforcement learning strategy is added to the ChemBFN and a controllable ordinary differential equation solver-like generating process is employed that accelerate the sampling processes. Most importantly, we introduce a semi-autoregressive strategy during training and inference that enhances the model performance and surpass the state-of-the-art models. A theoretical analysis of out-of-distribution generation in ChemBFN with semi-autoregressive approach is included as well.</p><p><b>Scientific contribution</b></p><p>Benchmarked on both small molecule and protein generative tasks, ChemBFN method showed outstanding out-of-distribution performance without complex modification, proving its feasibility of exploring chemical spaces outside the training data and its value as a tool to accelerate drug design and materials discovery.</p><h3>Graphic Abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01248-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148336133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hold on tight: the kinetic profiling of opioid receptor ligands using the CORAL-MD","authors":"Kinga Kurowska,&nbsp;Wiktor Rorat,&nbsp;Szymon K. Kordylewski,&nbsp;Sabina Podlewska","doi":"10.1186/s13321-026-01211-8","DOIUrl":"10.1186/s13321-026-01211-8","url":null,"abstract":"<div><p>Opioid receptors, particularly the μ-opioid receptor, are central targets in pain management, yet opioid therapeutics remain limited by adverse effects. Recent work indicates that ligand residence time may contribute to signaling bias (a potential avenue to dissociate therapeutic effects from deleterious side effects) by stabilizing distinct receptor conformations. Here, we investigated molecular determinants underlying prolonged residence time of μ-opioid receptor ligands. Seven compounds with experimentally measured residence times and diverse signaling profiles were studied using conventional MD, random accelerated MD, and metadynamics to characterize their unbinding behavior and contact patterns with the receptor.</p><p>To enable the systematic extraction and qualitative exploration of interaction features associated with residence time, we developed CORAL-MD (Correlational Analysis of Ligand–Protein Interactions)–an automated platform that analyzes MD-derived interaction profiles in relation to compound kinetics and other experimental outcomes (activity/properties). Application of CORAL-MD revealed specific interaction motifs associated with extended ligand retention, offering valuable guidance for designing safer, more selective opioid therapeutics. The tool is freely accessible at http://coralmd.if-pan.krakow.pl and it facilitates biological interpretation of MD simulations without requiring any programming expertise.</p><p><b>Scientific contribution</b></p><p>We identified key ligand-receptor interaction patterns and structural determinants that contribute to prolonged complex stability, including persistent interactions with Asp147^3.32 and Tyr148^3.33, as well as additional stabilizing contacts in the extracellular vestibule and at the TM5/TM6 interface. However, the interaction patterns can differ depending on the initial receptor conformation, which should be carefully selected on the basis of the functional effect triggered by the examined compounds. Our findings deepen the mechanistic understanding of binding kinetics in opioid receptors and offer valuable guidance for the rational design of next-generation opioid therapeutics with improved efficacy and reduced side effects. We also share CORAL-MD, an on-line tool (freely-available at http://coralmd.if-pan.krakow.pl), enabling analysis of dependencies between ligand-interaction patterns from molecular dynamics simulations and compound activities, kinetic properties or other experimentally verified features.</p><h3>Graphical Abstract</h3>\u0000<div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":617,"journal":{"name":"Journal of Cheminformatics","volume":"18 1","pages":""},"PeriodicalIF":7.9,"publicationDate":"2026-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s13321-026-01211-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148336134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}