无碰撞的摩根指纹:一种原则性的方法来提高机器学习性能和化学中的可解释性
Jibai Li1, Xintong Qu1, Wenlong Zhang2
1Department of Environmental Science, School of Ecological and Environmental Sciences, East China Normal University, Shanghai, 200241, People's Republic of China.
Journal of cheminformatics
|March 2, 2026
概括
无碰撞的摩根指纹 (CF-MF) 解决了标准分子表示中的位碰撞缺陷,提高了AI模型的准确性和化学解释性. 这确保了更可靠的AI用于药物发现和材料科学.
科学领域:
- * 计算化学 计算机化学
- * 机器学习 * 机器学习
- * 化学信息学
背景情况:
- * 摩根指纹由于效率和可解释性,在化学机器学习中被广泛使用.
- * 一个关键的缺陷,比特碰撞,错误地映射了不同的子结构,破坏了结构-属性关系并阻碍了模型的解释性.
研究的目的:
- * 引入无碰撞的摩根指纹 (CF-MF),以消除位碰撞并确保信息保真.
- * 为分子表示提供一个原则,适应性,数据驱动的框架.
主要方法:
- * 开发了CF-MF的适应性,数据驱动的尺寸机制.
- *对25个不同的数据集 (>50,000个分子) 和各种机器学习模型进行了CF-MF评估.
- *采用信息理论分析,将碰撞引起的损失与性能降低相关联.
主要成果:
- * CF-MF实现了显著的绩效增长:高达16.81%的RMSE减少和11.1%的精度增加.
- *通过消除归因错误恢复化学解释性,并将可靠的预测域扩大了60-100%.
- * 在信息忠实性和预测性能之间显示出强烈的相关性 (R2=0.854).
结论:
- *CF-MF为AI在药物发现,材料科学和环境评估方面提供了可靠和值得信赖的基础.
- * 确立了信息忠实性作为下一代分子表示的关键原则.
- *CF-MF提供了与深度学习模型竞争的性能,同时保持了指纹的简单性.
相关概念视频
MALDI-TOF Mass Spectrometry
7.2K
Mass spectrometry is a powerful characterization technique that can identify and separate a wide variety of compounds ranging from chemical to biological entities, based on their mass-to-charge ratio (m/z). The instruments that allow this detection, known as mass spectrometers, have three components: an ion source, a mass analyzer, and a detector. These spectrometers differ based on the nature of their ion source and analyzers.Matrix-assisted laser desorption ionization (MALDI) is a commonly...
7.2K
Molecular Models
44.3K
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.
44.3K
Predicting Molecular Geometry
46.5K
VSEPR Theory for Determination of Electron Pair Geometries
46.5K
Atomic Force Microscopy
4.6K
Atomic force microscopy (AFM) is a type of scanning probe microscopy that can analyze topographic details of various specimens like ceramics, glass, polymers, and biological samples. AFM offers over 1000 times more resolution than the optical imaging system. Images generated from AFM are three-dimensional surface profiles, offering an advantage over the flat, two-dimensional images from other imaging techniques.
The AFM Probe
The probe is regarded as the heart of any AFM setup and comprises the...
The AFM Probe
The probe is regarded as the heart of any AFM setup and comprises the...
4.6K
IR Frequency Region: Fingerprint Region
2.1K
IR spectra are divided into two main regions: the diagnostic region and the fingerprint region. The diagnostic region of the spectrum lies above 1500 cm−1. The absorptions resulting from single-bond vibrations of the N–H, C–H, and O–H stretch at higher wavenumbers and appear on the left side of the spectrum. The stretching absorptions of the C≡C and C≡N occur between 2100–2300 cm−1. In contrast, those arising from stretching absorptions of the...
2.1K
Molecular Orbital Theory I
48.3K
Overview of Molecular Orbital Theory
48.3K


