可解释的图形神经网络与数据增强用于预测C-H酸的pKa
Hongle An1,2, Xuyang Liu1,2, Wensheng Cai1,2
1Research Center for Analytical Sciences, Tianjin Key Laboratory of Biosensing and Molecular Recognition, State Key Laboratory of Medicinal Chemical Biology, College of Chemistry, Nankai University, Tianjin 300071, China.
Journal of chemical information and modeling
|September 14, 2023
概括
预测C-H酸pKa是一个挑战. 一个新的图形神经网络模型与知识注入的数据增强提高了预测准确性,特别是对于有限的实验数据.
科学领域:
- 计算化学计算化学
- 有机化学 有机化学
- 机器学习 机器学习
背景情况:
- 预测C-H酸的酸度 (pKa) 对有机合成,药物发现和材料科学至关重要.
- 目前的预测方法因实验数据稀缺和化学洞察力不足而面临限制.
研究的目的:
- 开发一种用于预测CH酸pKa值的新型模型.
- 通过先进的图形神经网络和数据增强技术来提高预测准确性.
主要方法:
- 使用图形神经网络 (GNN) 与传递单元 (MPU) 来提取分子图形信息.
- 通过替换原子来增加数据多样性来实施知识注入的数据增强策略.
- 采用读取层和完全连接的网络进行pKa预测,专注于电离位的碳原子.
主要成果:
- 知识注入的数据增强显著提高了模型的预测准确性,特别是在小数据集.
- 模型可解释性研究确定了影响C-H酸pKa值的关键替代物.
- 该模型在两个独立的数据集 (DMSO和水) 上显示出强大的性能.
结论:
- 拟议的GNN模型与知识注入的数据增强相结合,为预测C-H酸pKa提供了一种强大的方法.
- 这种方法提高了预测准确度,并提供了化学洞察力,解决了现有技术的局限性.
更多相关视频
相关概念视频
Basicity of Aliphatic Amines
5.9K
Amines can behave as Brønsted–Lowry bases by accepting a proton from the acid to form corresponding conjugate acids. Due to a lone pair of nonbonding electrons, aliphatic amines can also act as Lewis bases by forming a covalent bond with an electrophile.
To measure the basicity of amines, two conventions are generally used. The first defines Kb as the basicity constant for the deprotonation reaction of water by the amine, as presented in Figure 1. Conventionally, lower Kb indicates...
To measure the basicity of amines, two conventions are generally used. The first defines Kb as the basicity constant for the deprotonation reaction of water by the amine, as presented in Figure 1. Conventionally, lower Kb indicates...
5.9K
End Point Prediction: Gran Plot
371
A Gran plot is used to predict the equivalence volume or endpoint of a potentiometric or acid-base titration without reaching the endpoint. Typically, titration data is collected as a function of the titrant's volume up to a point less than the equivalence volume and then transformed into a linear format. The straight line is extended to the x-axis, indicating the necessary titrant volume to achieve the equivalence point.
For potentiometric titration, the Gran plot is created by plotting...
For potentiometric titration, the Gran plot is created by plotting...
371
Ladder Diagrams: Acid–Base Equilibria
518
Understanding the chemistry between the reagents is necessary for performing any experiment. To this end, scientists have designed a tool called a ladder diagram, which is a graphical representation that helps illustrate the chemistry of a system.
A ladder diagram for acid-base equilibria consists of a vertical axis that represents pH and horizontal bars (steps on the ladder) that help position all the pKa values in the system. At equilibrium, the pH value of the system corresponds to one of...
A ladder diagram for acid-base equilibria consists of a vertical axis that represents pH and horizontal bars (steps on the ladder) that help position all the pKa values in the system. At equilibrium, the pH value of the system corresponds to one of...
518
IR and UV–Vis Spectroscopy of Carboxylic Acids
4.2K
In IR spectroscopy of carboxylic acids, the C=O bond shows a characteristic band between 1710 and 1760 cm⁻¹, and the O–H bond exhibits a broad band between 2500 and 3300 cm⁻¹.
However, the stretching absorptions for the C=O bond vary depending on the structure of carboxylic acids. The C=O bond of the free carboxylic acids shows a higher stretching frequency, 1760 cm−1, while H-bonded carboxylic acids (dimers) exhibit stretching absorptions at a lower frequency,...
However, the stretching absorptions for the C=O bond vary depending on the structure of carboxylic acids. The C=O bond of the free carboxylic acids shows a higher stretching frequency, 1760 cm−1, while H-bonded carboxylic acids (dimers) exhibit stretching absorptions at a lower frequency,...
4.2K
NMR and Mass Spectroscopy of Carboxylic Acids
4.0K
In ¹H NMR spectroscopy, acidic protons (–COOH) of carboxylic acids are highly deshielded and absorb far downfield, at around 9–12 ppm. The chemical shift value depends on the concentration and solvent used.
While α protons of carboxylic acids absorb at 2–2.5 ppm, β protons absorb further upfield.
Carboxylic acids are easily identified by dissolving them in deuterium oxide, which results in a rapid exchange of the acidic protons with deuterium. This leads to the...
While α protons of carboxylic acids absorb at 2–2.5 ppm, β protons absorb further upfield.
Carboxylic acids are easily identified by dissolving them in deuterium oxide, which results in a rapid exchange of the acidic protons with deuterium. This leads to the...
4.0K
Molecular Models
38.6K
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.
38.6K


