相关实验视频

Updated: Jan 14, 2026

Quantitative Structure-Activity Relationship, Activity Prediction, and Molecular Dynamics of Non-nucleotide Reverse Transcriptase Inhibitors

Quantitative Structure-Activity Relationship, Activity Prediction, and Molecular Dynamics of Non-nucleotide Reverse Transcriptase Inhibitors

Published on: May 9, 2025

将GCN结构学习与LLM化学知识相结合,用于增强虚拟选

Radia Berreziga¹, Mohammed Brahimi^2,3, Khairedine Kraim³

¹Laboratory of Research in Artificial Intelligence (LRIA), Faculty of Computer Science, University of Science and Technology Houari Boumediene (USTHB), Algiers 16111, Algeria.

Journal of chemical information and modeling

|October 21, 2025

概括

本研究介绍了一种混合深度学习模型,将图形卷积网络 (GCN) 和大语言模型 (LLM) 结合起来,用于在药物发现中进行增强的虚拟选. 新型架构提高了识别潜在药物候选者的准确性.

科学领域:

计算化学是一种计算化学.
药物发现药物发现
在化学信息学中的机器学习.

背景情况:

传统的虚拟选的机器学习方法通常会因为预定义的分子表示而丢失信息.
深度学习,特别是图形卷积网络 (GCNs),通过直接处理分子图提供了更具表现力的方法.
大型语言模型 (LLM) 擅长从大型数据集中捕获复杂的化学模式.

研究的目的:

开发一种新的混合深度学习架构,以改善药物发现中的虚拟查.
为了更有效的分子表示,将GCN与LLM衍生的嵌入整合起来.
为了证明模型在不同目标类别的广泛适用性.

主要方法:

提出了一种混合架构,将GCN与LLM衍生嵌入组合在一起.
实施了一种层 wise 连接策略,以逐步丰富分子表示.
评估了酶和非酶数据集的模型,与GCN,Molformer,SVM和XGBoost进行比较.

主要成果:

混合型号的精度达到了88.7%,超过了独立的GCN (86.8%),Molformer (85.1%),XGBoost (85.0%) 和SVM (84.7%).
层层地注入LLM嵌入物逐渐丰富了分子表示与全球化学背景.
可以预先计算LLM嵌入,确保训练和推理期间的计算效率.

结论:

拟议的混合GCN-LLM架构显著提高了虚拟选性能.
这种方法提供了一种更准确,更有效的方法来识别有前途的候选药物.
该模型在各种数据集中的有效性突显了其在药物发现中广泛应用的潜力.

更多相关视频

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Published on: April 13, 2022

Pharmacophore Modeling for Targets with Extensive Ligand Libraries: A Case Study on SARS-CoV-2 Mpro

Pharmacophore Modeling for Targets with Extensive Ligand Libraries: A Case Study on SARS-CoV-2 Mpro

Published on: September 26, 2025

相关概念视频

Ligand Binding and Linkage

Ligand Binding and Linkage

Ligand Binding and Linkage

Ligand Binding and Linkage

Allosteric proteins have more than one ligand binding site; the binding of a ligand to any of these sites influences the binding of ligands to the other sites. When a protein is allosteric, its binding sites are called coupled or linked. In the case of enzymes, the site that binds to the substrate is known as the active site and the other site is known as the regulatory site. When a ligand binds to the regulatory site, this leads to conformational changes in the protein that can influence...

Ligand Binding Sites

Ligand Binding Sites

Proteins are dynamic macromolecules that carry out a wide variety of essential processes; however, the activities of most proteins depend on their interactions with other molecules or ions, known as ligands.
Protein-ligand interactions are quite specific; even though numerous potential ligands surround a cellular protein at any given time, only a particular ligand can bind to that protein. Moreover, a ligand binds only to a dedicated area on the surface of the protein, known as the...

Ligand Binding Sites

Ligand Binding Sites

Molecular Models

Molecular Models

Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.

Predicting Molecular Geometry

Predicting Molecular Geometry

VSEPR Theory for Determination of Electron Pair Geometries