Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Solving Constrained Optimization Problems Using Hybrid Qubit-Qumode Quantum Devices.

Journal of chemical theory and computation·2026

Same author

Crystal structure and molecular dynamics simulations of rademikibart Fab-IL-4Rα complex reveal biochemical basis for next-generation potent IL-4Rα inhibition in type 2 allergic and inflammatory diseases.

bioRxiv : the preprint server for biology·2026

Same author

Insights on Covalent Attachment of a Re Bipyridyl Catalyst onto <i>p</i>-Si(111) Using an Alkyl Linker for Photoelectrochemical CO<sub>2</sub> Reduction.

Inorganic chemistry·2026

Same author

Electrochemical Nitrate Reduction to Ammonia Driven by Catalytic Monovacancies in Single-Walled Carbon Nanotubes.

The journal of physical chemistry letters·2026

Same author

Mechanism of Tyrosine-Driven Deprotonation in Photosystem II Revealed by Multiscale Simulations.

Journal of the American Chemical Society·2026

Same author

Computational Evolution of Anti-PD-1 Antibodies Induces Structural Refolding for High-Affinity Interactions.

Biochemistry·2026

Search research articles

相关实验视频

Updated: Jul 15, 2025

Rapid, Enzymatic Methods for Amplification of Minimal, Linear Templates for Protein Prototyping using Cell-Free Systems

Rapid, Enzymatic Methods for Amplification of Minimal, Linear Templates for Protein Prototyping using Cell-Free Systems

Published on: June 14, 2021

化学空间AL:一种有效的积极学习方法,应用于特定蛋白质的分子生成.

Gregory W Kyro¹, Anton Morgunov¹, Rafael I Brent¹

¹Yale University.

|September 25, 2023

概括

这项研究引入了一种有效的主动学习方法,用于药物发现中的生成人工智能. 这种方法成功地产生了新型分子,甚至可以复制针对特定蛋白质标的现有抑制剂.

科学领域:

计算化学是一种计算化学.
人工智能的人工智能是人工智能.
药物发现药物发现

背景情况:

生成型人工智能 (AI) 模型越来越多地应用于药物发现.
探索广的化学空间需要有效的方法来识别具有所需性质的分子.

研究的目的:

开发一种计算效率高的主动学习方法,使生成模型与分子生成中的特定目标保持一致.
为了证明这种方法在针对蛋白质标的向分子生成中的应用.

主要方法:

实施了需要评估生成数据子集的积极学习策略.
使用有针对性的方法微调基于GPT的分子发生器.
将该方法应用于c-Abl激酶和Cas9的HNH域.

主要成果:

该生成模型成功地学会了产生类似于已知的c-Abl激酶抑制剂的分子,包括两种FDA批准的药物的精确复制.
该方法在产生针对Cas9的HNH域的分子方面被证明是有效的,这种目标缺乏现有的抑制剂.
该方法通过仅对生成数据的一个子集进行评估来证明计算效率.

结论:

开发的积极学习方法对于使用生成性AI进行向分子生成是有效的.

更多相关视频

Inducible T7 RNA Polymerase-mediated Multigene Expression System, pMGX

Inducible T7 RNA Polymerase-mediated Multigene Expression System, pMGX

Published on: June 27, 2017

Improving Student Outcomes with an Adaptable Molecular Cloning Course-Based Undergraduate Research Experience

Improving Student Outcomes with an Adaptable Molecular Cloning Course-Based Undergraduate Research Experience

Published on: November 15, 2024

相关实验视频

Last Updated: Jul 15, 2025

Rapid, Enzymatic Methods for Amplification of Minimal, Linear Templates for Protein Prototyping using Cell-Free Systems

Rapid, Enzymatic Methods for Amplification of Minimal, Linear Templates for Protein Prototyping using Cell-Free Systems

Published on: June 14, 2021

Inducible T7 RNA Polymerase-mediated Multigene Expression System, pMGX

Inducible T7 RNA Polymerase-mediated Multigene Expression System, pMGX

Published on: June 27, 2017

Improving Student Outcomes with an Adaptable Molecular Cloning Course-Based Undergraduate Research Experience

Improving Student Outcomes with an Adaptable Molecular Cloning Course-Based Undergraduate Research Experience

Published on: November 15, 2024

该方法是可通用的,适用于各种蛋白质标,包括那些没有现有抑制剂的蛋白质标.

开源的ChemSpaceAL Python套件有助于实现和复制这种方法.