Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 16, 2026

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids
08:21

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Published on: April 13, 2022

Diffusion-Based Generative Model With Scaffold-Hopping Strategy Yields Highly Potent Bioactive Molecules.

Yuwei Yang1, Xiaoqing Gong1, Shukai Gu1

  • 1Faculty of Applied Sciences, Macao Polytechnic University, Macao SAR, China.

Advanced Science (Weinheim, Baden-Wurttemberg, Germany)
|May 15, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Bioreactor Controls-III01:22

Bioreactor Controls-III

Strain improvement is a foundational strategy in industrial microbiology aimed at maximizing microbial productivity, particularly because natural isolates typically yield commercially valuable products in very low concentrations. Although optimizing the culture medium and environmental conditions can improve yields, these adjustments are inherently limited by the organism’s genetic potential. As a result, the focus shifts toward genetic modifications to enhance biosynthetic capacity. The...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Facilitating structure-based drug discovery with an artificial intelligence-driven virtual screening platform.

Nature protocols·2026
Same author

A Multistage Virtual Screening Strategy Integrating Molecular Similarity, Deep Learning Scoring, and Molecular Docking toward the Discovery of Novel LRRK2 Inhibitors.

Journal of chemical information and modeling·2026
Same author

Comprehensive Assessment and Benchmark of Deep Generative Models for Proteolysis TArgeting Chimera (PROTAC) Design.

Journal of chemical information and modeling·2026
Same author

LaMGen: LLM-based 3D molecular generation for multi-target drug design.

Nature communications·2026
Same author

High-Throughput Screening of FDA-Approved Drugs for Antibacterial and Antibiofilm Activities Against Multidrug-resistant <i>Pseudomonas aeruginosa</i>.

ACS omega·2026
Same author

Metadynamics Simulation Reveals Allosteric Communication Effects of the Flipping Process of the Atypical DLG Motif in RIPK1.

Journal of chemical information and modeling·2026
Same journal

Learning Moisture-Induced Damage From Vision: Diffusion Models for Real-Time Monitoring of Additive Manufacturing Processes.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same journal

Intrinsic Dual-Phase Regulated GeSe<sub>2</sub> Nanoparticles Triggered by Ball-Milling Treatment for Photonic Multi-Valued Logic Circuits.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same journal

A Plant Photoregulator-Inspired S-Type Heterojunction System for Diabetic Keratopathy via Tri-Modal Light-Driven Immunometabolic Reprogramming, Tissue Repair, and Antibacterial Activity.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same journal

eEF1G Orchestrates Translation to Ensure Meiotic Progression in Transcriptionally Quiescent Spermatocytes.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same journal

Ultrasound-Recharged Sub-Nanometer Palladium Catalysts for on-Demand and Self-Terminating Bioorthogonal Prodrug Activation in Cancer Therapy.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same journal

Graphene Aerogels With Spherical Pore Structure for Broad Frequency Regulation and Enhanced Low-Frequency Response.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
See all related articles

A new AI model, SMarT-Diff, enhances drug discovery by optimizing multiple molecular properties simultaneously. It successfully generated potent drug candidates, including one with superior performance against LRRK2 compared to existing controls.

Area of Science:

  • Artificial intelligence in drug discovery
  • Computational chemistry and molecular modeling
  • Medicinal chemistry and lead optimization

Background:

  • Lead optimization is crucial but challenging in drug discovery, requiring balancing multiple conflicting properties.
  • Existing generative models struggle to balance multi-objective optimization with scaffold exploration.

Purpose of the Study:

  • To develop a generative model that balances multi-objective optimization with scaffold-level exploration for lead optimization.
  • To introduce SMarT-Diff (Scaffold-based Multi-property Tuning Diffusion) for improved molecular generation and optimization.

Main Methods:

  • Developed SMarT-Diff, a scaffold-based multi-property tuning diffusion model.
  • Applied the model to single-target (LRRK2, HPK1, GLP-1R) and dual-target (GSK3β/JNK3) molecular optimization tasks.
Keywords:
deep learningdiffusion modeldrug discoverymolecular generationmulti‐objective optimization

More Related Videos

Microfluidic Synthesis of Microgel Building Blocks for Microporous Annealed Particle Scaffold
09:34

Microfluidic Synthesis of Microgel Building Blocks for Microporous Annealed Particle Scaffold

Published on: June 16, 2022

Related Experiment Videos

Last Updated: May 16, 2026

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids
08:21

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Published on: April 13, 2022

Microfluidic Synthesis of Microgel Building Blocks for Microporous Annealed Particle Scaffold
09:34

Microfluidic Synthesis of Microgel Building Blocks for Microporous Annealed Particle Scaffold

Published on: June 16, 2022

  • Validated model-generated compounds through wet-lab experiments against LRRK2.

    • Main Results:

    • SMarT-Diff demonstrated superior performance in molecular generation and optimization metrics.
    • The model generated drug-like molecules with enhanced structural diversity, preserved pharmacophores, and high synthetic accessibility.
    • A generated compound showed exceptional potency against LRRK2 (IC50 of 1.544 nM), outperforming the positive control.

    Conclusions:

    • SMarT-Diff effectively balances multi-objective optimization with scaffold hopping, enabling structural novelty and property enhancement.
    • The model shows strong potential for designing and optimizing novel, highly effective drug candidates.
    • Wet-lab validation confirms the practical utility and success of AI-driven drug design using SMarT-Diff.