MCST-AFN:基于低保真分子动力学模型的多通道时空特征自适应融合网络框架
Xing Chen1, Weichen Liu2, Tiantian Ruan3
1School of Information Science and Technology, Nantong University, Nantong 226019, Jiangsu, China.
ACS omega
|July 29, 2025
概括
这项研究引入了一种用于分子性质预测的新型深度学习框架,通过集成动态构造来增强分子表示,并降低药物开发的计算成本.
科学领域:
- 计算化学的计算化学
- 药物发现 药物发现 药物发现
- 机器学习 机器学习
背景情况:
- 预测分子特性对于药物开发至关重要.
- 分子表示是属性预测的关键.
- 动态分子构造,而不仅仅是静态结构,影响性质.
研究的目的:
- 开发一种有效的分子表示方法,包括动态分子构造.
- 为了降低与传统的4D-QSPR方法相关的高计算成本.
- 为了提高药物发现中的分子性质预测准确度.
主要方法:
- 提出了一个多通道时空特征自适应融合网络 (MCST-AFN) 框架.
- 使用低保真分子动力学 (MD) 模型进行高效的坐标更新和多通道嵌入.
- 采用以注意力为基础的网络来适应时空特征的融合和自我监督的学习来增强表现.
主要成果:
- 在12个分子性质预测数据集中,MCST-AFN框架实现了2.10%的平均性能改善.
- 在ESOL数据集中观察到19.70%的显著绩效提升.
- 该方法有效地整合了动态分子信息,同时降低了计算成本.
结论:
- MCST-AFN框架为分子性质预测提供了一种计算效率高,准确的方法.
- 通过深度学习整合动态分子构造,可以增强分子表征.
- 这种方法在加速药物开发方面具有重大前景.
相关概念视频
The Fluid Mosaic Model
153.1K
The fluid mosaic model was first proposed as a visual representation of research observations. The model comprises the composition and dynamics of membranes and serves as a foundation for future membrane-related studies. The model depicts the structure of the plasma membrane with a variety of components, which include phospholipids, proteins, and carbohydrates. These integral molecules are loosely bound, defining the cell’s border and providing fluidity for optimal function.
153.1K
Adaptability of Cytoskeletal Filaments
3.9K
The cytoskeleton is a complex dynamic structure performing varied functions based on cellular requirements. The adaptability of the individual filaments in the cytoskeleton determines their ability to perform various functions within the cell. It can undergo rapid reorganization during processes like cell division or remain stable for several hours as in the interphase. The adaptability of these filaments depends on stringent regulatory mechanisms. The microfilament and microtubules of the...
3.9K
Molecular Models
40.5K
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.
40.5K
¹H NMR of Conformationally Flexible Molecules: Temporal Resolution
917
At room temperature, the chair conformer of cyclohexane undergoes rapid ring flipping between two equivalent chair conformers at a rate of approximately 105 times per second. These two chair conformers are in equilibrium. The rapid ring flipping results in the interconversion of the axial proton to an equatorial proton and an equatorial to the axial proton. Such interconversions are too rapid and cannot be detected on the NMR timescale. Hence, the NMR spectrometer cannot distinguish between the...
917
Fluid Mosaic Model
12.9K
Scientists identified the plasma membrane in the 1890s and its principal chemical components (lipids and proteins) by 1915. The model for plasma membrane structure, proposed in 1935 by Hugh Davson and James Danielli, was the first model to be widely accepted in the scientific community. The model was based on the plasma membrane's "railroad track" appearance in early electron micrographs. Davson and Danielli theorized that the plasma membrane's structure resembled a sandwich...
12.9K
Model Approaches for Pharmacokinetic Data: Distributed Parameter Models
127
Pharmacokinetic models are mathematical constructs that represent and predict the time course of drug concentrations in the body, providing meaningful pharmacokinetic parameters. These models are categorized into compartment, physiological, and distributed parameter models.
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
127


