基于分子嵌入的算法选择在蛋白质-配体对接中
Jiabao Brad Wang1, Siyuan Cao1, Hongxuan Wu1
1Division of Natural and Applied Sciences, Duke Kunshan University, 8 Duke Av., Suzhou, 215316, Jiangsu, China.
一个新的算法选择模型MolAS通过预测每个算法的有效性来提高分子对接性能. 它比单一最佳解决方案 (SBS) 提供了显著的优势,并有助于缩小与虚拟最佳解决方案 (VBS) 的差距.
科学领域:
- 计算化学是一种计算化学.
- 药物发现 药物发现
- 在生物信息学中的机器学习.
背景情况:
- 分子对接算法的选择是具有挑战性的,因为其依赖于上下文的性能.
- 没有单一的算法可靠地在所有结构,化学和协议变异中执行.
- 现有的方法往往缺乏适应性和通用性.
研究的目的:
- 介绍MolAS,一个轻量级的算法选择模型用于分子对接.
- 使用预训的蛋白质和连接体嵌入来预测每个算法的性能.
- 为了提高单一最佳解决方案 (SBS) 的性能,并接近虚拟最佳解决方案 (VBS) 的性能.
主要方法:
- 使用预训练的蛋白质和带嵌入物.
- 使用注意力聚合和浅余解码器来预测性能.
- 在五个分子对接基准中评估了MolAS,其中有数百到数千个标记的复合体.
主要成果:
- 莫拉斯比单一最佳解决器 (SBS) 取得了高达15个百分点的绝对改进.
- 该模型缩小了虚拟最佳解决者 (VBS) 和SBS之间的差距的17-66%.
- 性能在低Oracle和可分离的顶部解决器区域最有效,但在协议不匹配下退化.
结论:
- 莫拉斯提供了一个强大的,基于嵌入的方法来选择对接算法.
- 它的有效性与预言景观特征和协议稳定性有关.
- 该模型作为域内选择器和诊断工具来评估选择的正确性.
更多相关视频
08:49Incorporating Target Protein Structure Flexibility and Dynamics in Computational Drug Discovery Using Ensemble-Based Docking Analysis
Published on: June 20, 2025
05:08Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins
Published on: July 8, 2025
相关概念视频
Ligand Binding Sites
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
Conserved Binding Sites
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...
Ligand Binding and Linkage
Ligand Binding and Linkage
The Equilibrium Binding Constant and Binding Strength
