Identification of STAT3 phosphorylation inhibitors using generative deep learning, virtual screening, molecular dynamics simulations, and biological evaluation for non-small cell lung cancer therapy DOI Creative Commons

Weiji Cai,

Beier Jiang,

Yichen Yin

et al.

Molecular Diversity, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 23, 2024

The development of phosphorylation-suppressing inhibitors targeting Signal Transducer and Activator Transcription 3 (STAT3) represents a promising therapeutic strategy for non-small cell lung cancer (NSCLC). In this study, generative model was developed using transfer learning virtual screening, leveraging comprehensive dataset STAT3 to explore the chemical space novel candidates. This approach yielded chemically diverse library compounds, which were prioritized through molecular docking dynamics (MD) simulations. Among identified candidates, HG110 molecule demonstrated potent suppression phosphorylation at Tyr705 inhibited its nuclear translocation in IL6-stimulated H441 cells. Rigorous MD simulations further confirmed stability interaction profiles top candidates within binding site. Notably, HG106 exhibited superior affinities stable conformations, with favorable interactions involving key residues pocket, outperforming known inhibitors. These findings underscore potential deep expedite discovery selective inhibitors, providing compelling pathway advancing NSCLC therapies.

Language: Английский

User-friendly and industry-integrated AI for medicinal chemists and pharmaceuticals DOI Creative Commons

Olga Kapustina,

Polina Burmakina,

Nina Gubina

et al.

Artificial Intelligence Chemistry, Journal Year: 2024, Volume and Issue: 2(2), P. 100072 - 100072

Published: July 14, 2024

Artificial intelligence has brought crucial changes to the whole field of natural sciences. Myriads machine learning algorithms have been developed facilitate work experimental scientists. Molecular property prediction and drug synthesis planning become routine tasks. Moreover, inverse design compounds with tunable properties as well on-the-fly autonomous process optimization chemical space exploration became possible in silico. Affordable robotic platforms exist able perform thousands experiments every day, analyzing results tuning protocols. Despite this, most these developments get trapped at stage code or overlooked, limiting their use by Meanwhile, visibility number user-friendly tools technologies available date is too low compensate for this fact, rendering development novel therapeutic inefficient. In Review, we set goal bridge gap between modern scientists improve efficacy. Here survey advanced easy-to-use help medical chemists research, including those integrated technological processes during COVID-19 pandemic motivated need fast yet precise solutions. review how are industry clinics streamline production. These already transform current paradigm scientific thinking revolutionize not only medicinal chemistry, but

Language: Английский

Citations

7
Molecular Generation for CNS Drug Discovery and Design DOI

Saifeng Chen,

Ding Luo, Weiwei Xue

et al.

ACS Chemical Neuroscience, Journal Year: 2025, Volume and Issue: unknown

Published: March 13, 2025

Computational drug design is a rapidly evolving field, especially the latest breakthroughs in generative artificial intelligence (GenAI) to create new compounds. However, potential of GenAI address challenges designing central nervous system (CNS) drugs that can effectively cross blood-brain barrier (BBB) and engage their targets remains largely unexplored. The integration techniques with experimental data sets advanced evaluation metrics provides unique opportunity enhance CNS discovery. In this viewpoint, we will introduce definition drug-like properties resources discovery, highlighting need train specialized models aimed at novel candidates by efficiently exploring space.

Language: Английский

Citations

0
A comprehensive review of neurotransmitter modulation via artificial intelligence: A new frontier in personalized neurobiochemistry DOI

Jaleh Bagheri Hamzyan Olia,

Arasu Raman, Chou‐Yi Hsu

et al.

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 189, P. 109984 - 109984

Published: March 14, 2025

Language: Английский

Citations

0
Can Machine Learning Overcome the 95% Failure Rate and Reality that Only 30% of Approved Cancer Drugs Meaningfully Extend Patient Survival? DOI
Duxin Sun,

Christian Macedonia,

Zhigang Chen

et al.

Journal of Medicinal Chemistry, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 10, 2024

Despite implementing hundreds of strategies, cancer drug development suffers from a 95% failure rate over 30 years, with only 30% approved drugs extending patient survival beyond 2.5 months. Adding more criteria without eliminating nonessential ones is impractical and may fall into the "survivorship bias" trap. Machine learning (ML) models enhance efficiency by saving time cost. Yet, they not improve success identifying root causes failure. We propose "STAR-guided ML system" (structure-tissue/cell selectivity-activity relationship) to addressing three overlooked interdependent factors: potency/specificity on/off-targets determining efficacy in tumors at clinical doses, on/off-target-driven tissue/cell selectivity influencing adverse effects normal organs optimal doses balancing efficacy/safety as determined selectivity. STAR-guided can directly predict dose/efficacy/safety five features design/select best drugs, enhancing development.

Language: Английский

Citations

2
Generative deep learning enables the discovery of phosphorylation-suppressed STAT3 inhibitors for non-small cell lung cancer therapy DOI Creative Commons

Weiji Cai,

Beier Jiang,

Yichen Yin

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 18, 2024

Abstract The discovery of phosphorylation-suppressed inhibitors for Signal Transducer and Activator Transcription 3 (STAT3) presents a novel therapeutic strategy non-small cell lung cancer (NSCLC). Despite the pivotal roles STAT3 in progression, effective remain limited, especially efficiently suppressing phosphorylation at Try705. This study harnesses generative deep learning to develop model de novo design that selectively target phosphorylated form subsequentially induce cellular apoptosis. Initially, we constructed utilizing with transfer virtual screening, trained on existing inhibitor datasets explore chemical space. We generated diverse library candidate compounds, which were subsequently screened through molecular docking pharmacophore modeling, identifying several promising inhibitors. Compared HG106, HG110 molecule can suppress STAT3, nucleus translocation H441, stimulated by IL6 pro-inflammatory factor. Rigorous dynamics (MD) simulations performed evaluate stability interaction profiles selected candidates within binding site. Among top candidates, compounds HG106 exhibited superior affinities compared known MD confirmed stable conformations favorable interactions key residues pocket, indicating potential vivo efficacy. demonstrates power accelerating identification inhibitors, providing direction NSCLC therapy.

Language: Английский

Citations

0
Identification of STAT3 phosphorylation inhibitors using generative deep learning, virtual screening, molecular dynamics simulations, and biological evaluation for non-small cell lung cancer therapy DOI Creative Commons

Weiji Cai,

Beier Jiang,

Yichen Yin

et al.

Molecular Diversity, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 23, 2024

The development of phosphorylation-suppressing inhibitors targeting Signal Transducer and Activator Transcription 3 (STAT3) represents a promising therapeutic strategy for non-small cell lung cancer (NSCLC). In this study, generative model was developed using transfer learning virtual screening, leveraging comprehensive dataset STAT3 to explore the chemical space novel candidates. This approach yielded chemically diverse library compounds, which were prioritized through molecular docking dynamics (MD) simulations. Among identified candidates, HG110 molecule demonstrated potent suppression phosphorylation at Tyr705 inhibited its nuclear translocation in IL6-stimulated H441 cells. Rigorous MD simulations further confirmed stability interaction profiles top candidates within binding site. Notably, HG106 exhibited superior affinities stable conformations, with favorable interactions involving key residues pocket, outperforming known inhibitors. These findings underscore potential deep expedite discovery selective inhibitors, providing compelling pathway advancing NSCLC therapies.

Language: Английский

Citations

0