莫尔德伯塔:物理化学和结构信息分子表示学习的基础模型
bioRxiv : the preprint server for biology
|February 27, 2026
概括
MolDe-BERTa是一种新的分子语言模型,通过学习分子结构和特性来增强药物和材料的发现. 它在预测任务中表现优于现有模型,加速化学研究.
科学领域:
- * 计算化学 计算机化学
- * 化学信息学 化学信息学
- * 机器学习 * 机器学习
背景情况:
- *基础模型对于加速材料和药物发现至关重要.
- *现有的分子语言模型使用通用目标,忽视了物理化学性质.
- * 需要基于结构的模型来弥合语言和物理分子表示.
研究的目的:
- * 介绍MolDe-BERTa,一个结构知情的自我监督的分子编码器.
- * 开发新的预训练目标,将分子特性嵌入到潜在空间中.
- * 推进基于编码器的无监督基础模型,用于化学信息的表示学习.
主要方法:
- * 使用了字节级别的字节对编码 (BPE) 代码化策略.
- * 在PubChem的1.23亿个SMILES分子的大体上预先训练了MolDe-BERTa.
- * 引入了三个新的预训练目标,以偏向分子性质和结构相似性.
主要成果:
- *MolDe-BERTa在9个下游的MoleculeNet基准上表现优于现有的掩盖语言模型.
- * 实现了回归误差降低16%.
- *在分类任务中,在ROC-AUC上获得了高达3.0分的改进.
结论:
- *MolDe-BERTa代表了无监督,化学信息的表示学习的重大进步.
- * 该模型通过整合结构和财产信息,使数据效率学习成为可能.
- * 公开可用的代码和数据集促进了分子发现的进一步研究.
相关概念视频
Molecular Models
44.2K
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.
44.2K
Molecular Shapes
62.9K
Molecules have characteristic shapes that are crucial for their function. The arrangement of various electron groups around the central atom dictates their molecular geometry. Electron pairs in the valence shell of a central atom will adopt an arrangement that minimizes repulsions between the electron pairs by maximizing the distance between them. The valence electrons form either bonding pairs, located primarily between bonded atoms, or lone pairs.
Two regions of electron density in a diatomic...
Two regions of electron density in a diatomic...
62.9K
VSEPR Theory
15.2K
Valence shell electron-pair repulsion theory (VSEPR theory) enables us to predict the molecular structure around a central atom from an examination of the number of bonds and lone electron pairs in its Lewis structure. The VSEPR model assumes that electron pairs in the valence shell of a central atom will adopt an arrangement that minimizes repulsions between these electron pairs by maximizing the distance between them. The electrons in the valence shell of a central atom form either bonding...
15.2K
Predicting Molecular Geometry
46.4K
VSEPR Theory for Determination of Electron Pair Geometries
46.4K
Fischer Projections
16.9K
Learning to draw Fischer projections of molecules and understanding their relevance plays a crucial role in the visual depiction of organic molecules. A Fischer projection is a two-dimensional projection on a planar surface to simplify the three-dimensional wedge–dash representation of molecules. This is especially helpful in the case of molecules with multiple chiral centers that can be difficult to draw. Here, all the bonds of interest are represented as horizontal or vertical lines. While...
16.9K
Mechanistic Models: Overview of Compartment Models
441
Mechanistic models, a category encompassing both physiological and compartmental modeling, differ from empirical models' approaches to incorporating known factors about the systems being modeled. Empirical models describe data with minimal assumptions, while mechanistic models aim to provide a robust description of available data by specifying assumptions and integrating known factors about the system. Compartmental analysis is a key example of a mechanistic model in pharmacokinetics and...
441


