对正规和非正规氨基酸的分子动力学 (MD) 衍生特征
Tiffani Hui1, Maxim Secor1, Minh Ngoc Ho1
1Department of Chemistry, Tufts University, Medford, Massachusetts 02155, United States.
Journal of chemical information and modeling
|February 3, 2025
概括
机器学习模型现在可以通过使用来自分子动力学模拟的新特性来预测具有新氨基酸的结构. 这种方法可以提高准确性,而不需要大量的新培训数据.
科学领域:
- 计算化学和结构生物学.
- 机器学习在生物信息学中的应用.
- 和蛋白质结构的预测.
背景情况:
- 用于和蛋白质结构预测的机器学习 (ML) 模型通常在正规氨基酸上进行训练.
- 这些模型无法准确预测训练数据中缺少非正规氨基酸的结构.
- 收集新数据和重新训练ML模型是耗时的,通常是不切实际的.
研究的目的:
- 开发可通用的氨基酸特征,以提高ML模型的可扩展性.
- 为了能够准确地预测含有新型氨基酸的结构.
- 减少对ML模型进行广泛数据收集和再培训的需求.
主要方法:
- 使用二的分子动力学 (MD) 模拟生成了新的氨基酸特征.
- 从脊柱二面角 (φ, ψ) 分布和静电潜力中提取的特征.
- 训练有素的ML模型使用这些新特性预测循环五结构组合.
主要成果:
- 与传统特征 (R2 = 0.4300.508) 相比,具有新 (φ, ψ) 特征的ML模型在预测含有未见氨基酸的结构方面获得了更高的准确性 (R2 = 0.700).
- 这些新功能显示出对原始训练套件中不存在的氨基酸的优越通用性.
- 使用传统的摩根指纹和MACCS键的性能随着扩展的氨基酸库而显著降低.
结论:
- 使用MD模拟开发了强大的和可泛化的氨基酸特征.
- 证明了这些特性的有效性,提高了ML模型的性能,以预测具有新型氨基酸的结构.
- 这种方法提供了一种更有效的替代方案,即用新的实验数据重新训练机器学习模型.
更多相关视频
相关概念视频
01:49Conserved Binding Sites
4.2K
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...
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...
4.2K
03:42Amino acids
87.7K
Amino acids are the monomers that comprise proteins. Each amino acid has the same fundamental structure, which consists of a central carbon atom, or the alpha (α) carbon, bonded to an amino group (NH2), a carboxyl group (COOH), and to a hydrogen atom. Every amino acid also has another atom or group of atoms bonded to the central atom known as the R group. There are 20 common amino acids present in proteins, each with a different R group. Variation in the amino acid sequence is responsible...
87.7K
02:00Molecular Models
37.8K
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.
37.8K
02:36Molecular Geometry and Dipole Moments
12.5K
The VSEPR theory can be used to determine the electron pair geometries and molecular structures as follows:
12.5K
02:34Molecular Structure and Acidity
16.8K
An acid can be deprotonated to form a conjugate base or an anion. If the produced anion is more stable, then the acid is stronger. On the contrary, if the anion is unstable, then the acid is weaker. Hence, to determine the acidity of the compound, the stability of its conjugate base is studied using various factors.
The size effect explains the change in atomic size on acidity. When comparing the acids formed from elements that belong to the same column in the periodic table, their atomic sizes...
The size effect explains the change in atomic size on acidity. When comparing the acids formed from elements that belong to the same column in the periodic table, their atomic sizes...
16.8K
02:06Newman Projections
16.3K
Different notations are used to represent the three-dimensional structure of molecules on two-dimensional surfaces. One of the most commonly used representations is the dash-wedge formula. The dashed wedges, solid wedges, and the plane lines indicate the groups situated behind the plane, coming out of the plane, and in the plane, respectively.
The organic molecules rotate across the single bonds leading to numerous temporary three-dimensional structures of varying energy known as...
The organic molecules rotate across the single bonds leading to numerous temporary three-dimensional structures of varying energy known as...
16.3K