Machine Learning‐Enabled Drug‐Induced Toxicity Prediction

Advanced Science, Год журнала: 2025, Номер unknown

Опубликована: Фев. 3, 2025

Abstract Unexpected toxicity has become a significant obstacle to drug candidate development, accounting for 30% of discovery failures. Traditional assessment through animal testing is costly and time‐consuming. Big data artificial intelligence (AI), especially machine learning (ML), are robustly contributing innovation progress in toxicology research. However, the optimal AI model different types usually varies, making it essential conduct comparative analyses methods across domains. The diverse sources also pose challenges researchers focusing on specific studies. In this review, 10 categories drug‐induced examined, summarizing characteristics applicable ML models, including both predictive interpretable algorithms, striking balance between breadth depth. Key databases tools used prediction highlighted, toxicology, chemical, multi‐omics, benchmark databases, organized by their focus function clarify roles prediction. Finally, strategies turn into opportunities analyzed discussed. This review may provide with valuable reference understanding utilizing available resources bridge mechanistic insights, further advance application drugs‐induced

Язык: Английский

---

Machine learning and molecular docking prediction of potential inhibitors against dengue virus

Frontiers in Chemistry, Год журнала: 2024, Номер 12

Опубликована: Дек. 24, 2024

Introduction Dengue Fever continues to pose a global threat due the widespread distribution of its vector mosquitoes, Aedes aegypti and albopictus . While WHO-approved vaccine, Dengvaxia, antiviral treatments like Balapiravir Celgosivir are available, challenges such as drug resistance, reduced efficacy, high treatment costs persist. This study aims identify novel potential inhibitors virus (DENV) using an integrative discovery approach encompassing machine learning molecular docking techniques. Method Utilizing dataset 21,250 bioactive compounds from PubChem (AID: 651640), alongside total 1,444 descriptors generated PaDEL, we trained various models Support Vector Machine, Random Forest, k-nearest neighbors, Logistic Regression, Gaussian Naïve Bayes. The top-performing model was used predict active compounds, followed by performed AutoDock Vina. detailed interactions, toxicity, stability, conformational changes selected were assessed through protein-ligand interaction studies, dynamics (MD) simulations, binding free energy calculations. Results We implemented robust three-dataset splitting strategy, employing Regression algorithm, which achieved accuracy 94%. successfully predicted 18 known DENV inhibitors, with 11 identified active, paving way for further exploration 2683 new ZINC EANPDB databases. Subsequent studies on NS2B/NS3 protease, enzyme essential in viral replication. ZINC95485940, ZINC38628344, 2′,4′-dihydroxychalcone ZINC14441502 demonstrated affinity −8.1, −8.5, −8.6, −8.0 kcal/mol, respectively, exhibiting stable interactions His51, Ser135, Leu128, Pro132, Ser131, Tyr161, Asp75 within site, critical residues involved inhibition. Molecular simulations coupled MMPBSA elucidated making it promising candidate development. Conclusion Overall, this approach, combining learning, docking, highlights strength utility computational tools discovery. It suggests pathway rapid identification development drugs against DENV. These silico findings provide strong foundation future experimental validations in-vitro aimed at fighting

Язык: Английский

---

Computational Toxicology and Risk Assessment

Elsevier eBooks, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

---

Quantitative structure–activity relationships of chemical bioactivity toward proteins associated with molecular initiating events of organ-specific toxicity

Journal of Cheminformatics, Год журнала: 2024, Номер 16(1)

Опубликована: Ноя. 5, 2024

The adverse outcome pathway (AOP) concept has gained attention as a way to explore the mechanism of chemical toxicity. In this study, quantitative structure–activity relationship (QSAR) models were developed predict compound activity toward protein targets relevant molecular initiating events (MIE) upstream organ-specific toxicities, namely liver steatosis, cholestasis, nephrotoxicity, neural tube closure defects, and cognitive functional defects. Utilizing bioactivity data from ChEMBL 33 database, various machine learning algorithms, features methods assess prediction reliability compared applied develop robust activity. results demonstrate high predictive performance across multiple targets, with balanced accuracy exceeding 0.80 for majority models. Furthermore, stability checks confirmed consistency training-test splits. obtained by using QSAR predictions identify known markers adversities highlighted utility risk assessment prioritizing compounds further experimental evaluation. Scientific contribution work describes development tools screening chemicals potential systemic toxicity, thus contributing resource savings providing indications better-targeted testing. This study provides advances in field computational modeling MIEs information AOP which is still relatively young unexplored. comprehensive procedure highly generalizable, offers framework predicting wide range toxicological endpoints.

Язык: Английский

---

Health risk assessment for dietary exposure to 3-monochloropropane-1,2-diol, 2-monochloropropane-1,2-diol, and glycidol for Italian consumers

Chemosphere, Год журнала: 2024, Номер unknown, С. 143339 - 143339

Опубликована: Сен. 1, 2024

Язык: Английский

---

Artificial Intelligence Methods for Evaluating Mitochondrial Dysfunction: Exploring Various Chemical Notations Suitable for Neural Language Processing Models

Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 116 - 131

Опубликована: Сен. 19, 2024

Язык: Английский

---

Exploring the future of biopharmaceutical drug discovery: can advanced AI platforms overcome current challenges?

Discover Artificial Intelligence, Год журнала: 2024, Номер 4(1)

Опубликована: Дек. 5, 2024

Язык: Английский

---