相关实验视频

Updated: Jun 7, 2025

Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules

Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules

Published on: July 25, 2013

ConfRank:通过对联培训改善GFN-FF符合者排名

Christian Hölzer¹, Rick Oerder^2,3, Stefan Grimme¹

¹Mulliken Center for Theoretical Chemistry, University of Bonn, Beringstr. 4, 53115 Bonn, Germany.

Journal of chemical information and modeling

|November 20, 2024

概括

机器学习方法ConfRank通过使用对联训练显著改善了药物发现的分子符合者排名. 这种方法提高了准确性和速度,在识别低能耗适配器方面表现优于现有的方法.

科学领域:

计算化学是一种计算化学.
药物发现药物发现
机器学习机器学习

背景情况:

符合性排名对于药物发现至关重要,但目前的方法在速度和准确性方面存在局限性.
强力计算的限制阻碍了大规模的选应用.

研究的目的:

介绍 ConfRank,一种用于增强符合者排名的机器学习方法.
提高识别低能分子对应物的准确性和效率.

主要方法:

开发了机器学习模型ConfRank,利用对联训练.
利用DimeNet++架构在大型数据集 (GEOM,QM9) 上进行训练.
使用各种指标 (RMSD,斯皮尔曼相关性) 对GFN-FF和GFN2-xTB的评估性能.

主要成果:

ConfRank 减少了对联 RMSD 从 5.65 到 0.71 kcal mol-1.1 的时间.
在识别最低能耗对应物方面,达到高达81%的准确性,明显优于现有方法.
与GFN2-xTB相比,已经证明了高达100倍的运行时间加速.
提高了斯皮尔曼相关性至0.90,将标志翻转概率降低到7%.

结论:

ConfRank提供了一个具有成本效益和可扩展的解决方案,用于准确的符合者排名.
配对训练方法显著提高了排名表现,尽管对绝对能量的预测可能有很小的影响.
在药物发现和 (生物) 分子结构分析方面,ConfRank显示出有前途的应用.

更多相关视频

A Protocol for Computer-Based Protein Structure and Function Prediction

A Protocol for Computer-Based Protein Structure and Function Prediction

Published on: November 3, 2011

Optimization of Synthetic Proteins: Identification of Interpositional Dependencies Indicating Structurally and/or Functionally Linked Residues

Optimization of Synthetic Proteins: Identification of Interpositional Dependencies Indicating Structurally and/or Functionally Linked Residues

Published on: July 14, 2015

相关概念视频

Conserved Binding Sites

Conserved Binding Sites

Many proteins’ biological role depends on their interactions with their ligands, small molecules that bind to specific locations on the protein known as ligand-binding sites. Ligand-binding sites are often conserved among homologous proteins as these sites are critical for protein function.
Binding sites are often located in large pockets, and if their location on a protein’s surface is unknown, it can be predicted using various approaches. The energetic method computationally...