PySIDT:子图等态决策树用于分子性质预测
Matthew S Johnson1, Hao-Wei Pang2, Anna C Doner2
1Combustion Research Facility, Sandia National Laboratories, Livermore, California 94551-0969, United States.
The journal of physical chemistry. A
|October 22, 2025
概括
亚图等态决策树 (SIDT) 为分子性质预测提供了比深度神经网络 (DNN) 更易于解释和数据效率更高的替代方案. PySIDT软件表现出卓越的性能,特别是在有限的数据中,超过了DNN和梯度增强树.
科学领域:
- 计算化学的计算化学
- 机器学习在化学中的应用
- 材料科学 材料科学 材料科学
背景情况:
- 精确的分子性质预测在化学中至关重要.
- 深度神经网络 (DNN) 很受欢迎,但需要大量的数据集,难以解释,并且难以结合化学知识.
- 现有的方法往往缺乏解释性,并与较小的数据集作斗争.
研究的目的:
- 介绍子图同态决策树 (SIDTs) 作为分子性质预测的新方法.
- 开发和介绍PySIDT软件,用于使用SIDT进行训练和推理.
- 在数据效率,可解释性和性能方面展示SIDT与DNN和梯度增强树相比的优势.
主要方法:
- 开发了Subgraph同型决策树 (SIDTs),一种使用分子子结构的基于图形的决策树方法.
- 实现PySIDT软件用于训练和运行SIDT上的推理.
- 将SIDT应用于各种分子预测任务,包括速率系数,热化学和稳定性.
主要成果:
- 在各种分子性质预测任务中,SIDT表现出强的性能.
- 与流行的DNN (Chemprop) 和梯度增强树木 (XGBoost) 方法相比,PySIDT表现优越,特别是在有限的训练数据下.
- 在形成预测的度方面,PySIDT在所有测试的训练/验证集大小中都超过了Chemprop和XGBoost.
结论:
- SIDT为DNN提供了一个可扩展,可解释和数据效率高的替代方案,用于分子性质预测.
- PySIDT促进了化学知识和不确定性估计的整合.
- 对于推进计算化学和材料科学应用,SIDT方法显示出显著的前景.
相关概念视频
02:27Predicting Molecular Geometry
44.9K
VSEPR Theory for Determination of Electron Pair Geometries
44.9K
03:21Phylogenetic Trees
49.2K
Phylogenetic trees come in many forms. It matters in which sequence the organisms are arranged from the bottom to the top of the tree, but the branches can rotate at their nodes without altering the information. The lines connecting individual nodes can be straight, angled, or even curved.
49.2K
02:00Molecular Models
43.5K
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.
43.5K
02:52Predicting Products: Substitution vs. Elimination
13.8K
When a nucleophile and an alkyl halide react, nucleophilic substitution and β-elimination reactions compete to generate products.
The following factors can influence the mechanisms competing against each other:
The following factors can influence the mechanisms competing against each other:
13.8K
01:20Applications of Molecular Taxonomy
506
Molecular taxonomy has revolutionized the understanding and classification of bacteria, providing precise insights into their diversity, evolutionary relationships, and ecological roles. By utilizing molecular techniques such as DNA sequencing and fingerprinting, researchers have made significant strides in various fields related to bacterial studies.Resolving Taxonomic AmbiguitiesMolecular taxonomy has been instrumental in distinguishing closely related bacterial species initially thought to...
506
02:36Molecular Geometry and Dipole Moments
18.1K
The VSEPR theory can be used to determine the electron pair geometries and molecular structures as follows:
18.1K