Exploring Hidden Dangers: Predicting Mycotoxin-like Toxicity and Mapping Toxicological Networks in Hepatocellular Carcinoma DOI

Jian Xiu,

Hengzheng Yang,

Xiaoli Shen

et al.

Journal of Chemical Information and Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: May 20, 2025

Mycotoxins are potent triggers of hepatocellular carcinoma (HCC) due to their intricate interplay with cellular macromolecules and signaling pathways. This study integrates machine learning biomolecular analyses elucidate the mechanisms underlying mycotoxin-induced hepatocarcinogenesis. Using a data set 1767 mycotoxins 1706 non-mycotoxin fungal metabolites, we evaluated 51 models. The KPGT model achieved optimal performance an ROC-AUC 0.979 balanced accuracy 0.930. Clustering analysis identified six distinct mycotoxin clusters unique structural features. Network toxicology revealed protein-protein interaction patterns across different clusters, identifying key regulatory proteins including EGFR, SRC, ESR1. GO enrichment uncovered cluster-specific effects on protein complexes macromolecular assemblies, particularly in membrane organization vesicular transport. KEGG pathway demonstrated systematic perturbation major cascades, each cluster distinctly modulating kinase networks receptor tyrosine Molecular docking validated these interactions, binding affinities ranging from -9.6 -4.7 kcal/mol. Notably, 5 showed strong SRC (-9.6 kcal/mol), EGFR (-9.5 ESR1 (-7.8 providing insights into toxin-macromolecule recognition. These findings enhance our understanding mycotoxin-protein interactions HCC development suggest potential therapeutic strategies targeting interfaces.

Language: Английский

Application and Development Trends of Network Toxicology in the Safety Assessment of Traditional Chinese Medicine DOI
Xiaoyan Li, Liangqing Lin, Li Pang

et al.

Journal of Ethnopharmacology, Journal Year: 2025, Volume and Issue: unknown, P. 119480 - 119480

Published: Feb. 1, 2025

Language: Английский

Citations

2
Machine Learning‐Enabled Drug‐Induced Toxicity Prediction DOI Creative Commons
Changsen Bai, Lianlian Wu, Ruijiang Li

et al.

Advanced Science, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 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

Language: Английский

Citations

1
Data-driven toxicity prediction in drug discovery: Current status and future directions DOI

Ningning Wang,

Xinliang David Li, Jing Xiao

et al.

Drug Discovery Today, Journal Year: 2024, Volume and Issue: unknown, P. 104195 - 104195

Published: Sept. 1, 2024

Language: Английский

Citations

4
Multimodal Representation Learning via Graph Isomorphism Network for Toxicity Multitask Learning DOI
Guishen Wang, Hui Feng,

Mengyan Du

et al.

Journal of Chemical Information and Modeling, Journal Year: 2024, Volume and Issue: 64(21), P. 8322 - 8338

Published: Oct. 21, 2024

Toxicity is paramount for comprehending compound properties, particularly in the early stages of drug design. Due to diversity and complexity toxic effects, it became a challenge compute toxicity tasks. To address this issue, we propose multimodal representation learning model, termed graph isomorphism network (MMGIN), multitask learning. Based on fingerprints molecular graphs compounds, our MMGIN model incorporates acquire comprehensive representation. This adopts two-channel structure independently learn fingerprint Subsequently, two feedforward neural networks utilize learned perform learning, encompassing classification multiple category simultaneously. test effectiveness constructed novel data set, (CTMTL) derived from TOXRIC set. We compare with other representative machine deep models CTMTL Tox21 sets. The experimental results demonstrate significant advancements achieved by model. Furthermore, ablation study underscores introduced fingerprints, graphs, showcasing model's superior predictive capability robustness.

Language: Английский

Citations

4
A multiscale molecular structural neural network for molecular property prediction DOI
Zhiwei Shi, Miao Ma, Hanyang Ning

et al.

Molecular Diversity, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 25, 2025

Language: Английский

Citations

0
Psychedelic Drugs in Mental Disorders: Current Clinical Scope and Deep Learning‐Based Advanced Perspectives DOI Creative Commons
Sung‐Hyun Kim, Sumin Yang,

Jeehye Jung

et al.

Advanced Science, Journal Year: 2025, Volume and Issue: unknown

Published: March 20, 2025

Mental disorders are a representative type of brain disorder, including anxiety, major depressive depression (MDD), and autism spectrum disorder (ASD), that caused by multiple etiologies, genetic heterogeneity, epigenetic dysregulation, aberrant morphological biochemical conditions. Psychedelic drugs such as psilocybin lysergic acid diethylamide (LSD) have been renewed fascinating treatment options gradually demonstrated potential therapeutic effects in mental disorders. However, the multifaceted conditions psychiatric resulting from individuality, complex interplay, intricate neural circuits impact systemic pharmacology psychedelics, which disturbs integration mechanisms may result dissimilar medicinal efficiency. The precise prescription psychedelic remains unclear, advanced approaches needed to optimize drug development. Here, recent studies demonstrating diverse pharmacological psychedelics reviewed, emerging perspectives on structural function, microbiota-gut-brain axis, transcriptome discussed. Moreover, applicability deep learning is highlighted for development basis big data. These provide insight into interindividual factors enhance discovery precision medicine.

Language: Английский

Citations

0
Exploring Hidden Dangers: Predicting Mycotoxin-like Toxicity and Mapping Toxicological Networks in Hepatocellular Carcinoma DOI

Jian Xiu,

Hengzheng Yang,

Xiaoli Shen

et al.

Journal of Chemical Information and Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: May 20, 2025

Mycotoxins are potent triggers of hepatocellular carcinoma (HCC) due to their intricate interplay with cellular macromolecules and signaling pathways. This study integrates machine learning biomolecular analyses elucidate the mechanisms underlying mycotoxin-induced hepatocarcinogenesis. Using a data set 1767 mycotoxins 1706 non-mycotoxin fungal metabolites, we evaluated 51 models. The KPGT model achieved optimal performance an ROC-AUC 0.979 balanced accuracy 0.930. Clustering analysis identified six distinct mycotoxin clusters unique structural features. Network toxicology revealed protein-protein interaction patterns across different clusters, identifying key regulatory proteins including EGFR, SRC, ESR1. GO enrichment uncovered cluster-specific effects on protein complexes macromolecular assemblies, particularly in membrane organization vesicular transport. KEGG pathway demonstrated systematic perturbation major cascades, each cluster distinctly modulating kinase networks receptor tyrosine Molecular docking validated these interactions, binding affinities ranging from -9.6 -4.7 kcal/mol. Notably, 5 showed strong SRC (-9.6 kcal/mol), EGFR (-9.5 ESR1 (-7.8 providing insights into toxin-macromolecule recognition. These findings enhance our understanding mycotoxin-protein interactions HCC development suggest potential therapeutic strategies targeting interfaces.

Language: Английский

Citations

0