IUPAC-GPT:基于IUPAC的大规模分子预训练模型,用于属性预测和分子生成
Jiashun Mao1,2, Tang Sui3, Kwang-Hwi Cho4
1School of Medical Information and Engineering, Southwest Medical University, Luzhou, 610199, China.
Molecular diversity
|July 3, 2025
概括
一个新的化学语言模型,IUPAC-GPT,在分子生成和属性预测任务中表现出色. 它的性能优于在SMILES字符串上训练的模型,提供更好的解释性和化学结构的语义理解.
科学领域:
- 计算化学的计算化学
- 化学信息学 化学信息学
- 自然语言处理自然语言处理.
背景情况:
- 国际纯化学和应用化学联盟 (IUPAC) 命名法是化学命名的一个人类友好的标准.
- 简化分子输入线输入系统 (SMILES) 是一种流行的计算机友好的分子表示语言.
- 将IUPAC名称和SMILES字符串进行比较对于推进分子建模任务至关重要.
研究的目的:
- 开发和评估一种新的化学语言模型,IUPAC-GPT,基于IUPAC命名法.
- 评估IUPAC-GPT在分子生成和下游任务 (如回归和分类) 的表现.
- 将IUPAC-GPT与现有模型进行比较,例如smilesGPT.
主要方法:
- 开发IUPAC-GPT,一种类似于变压器的语言模型,预先训练在IUPAC公司.
- 整合了参数结和轻量级的网络连接器,用于微调.
- 对分子生成,二进制分类和属性回归预测任务的模型的评估.
主要成果:
- 预先训练的IUPAC-GPT有效地将一般化学知识转移到下游任务中.
- 与微笑GPT相比,IUPAC-GPT在某些属性预测任务中表现优越.
- 该模型显示了在分子生成,分类和回归方面的强大能力.
结论:
- 在IUPAC corpora上预先训练的基于变压器的语言模型是化学信息学的有效替代品.
- 与基于SMILES的模型相比,IUPAC-GPT提供了更好的解释性和语义抽象.
- 该研究强调了IUPAC命名体系在先进的计算化学应用中的潜力.
更多相关视频
06:50Author Spotlight: A Computational Approach to Decipher Amino Acid Preferences in Multispecific Protein-Protein Interactions
Published on: January 26, 2024
2.0K
10:21Author Spotlight: Streamlining Protein Target Prediction and Validation via Molecular Docking and CETSA
Published on: February 23, 2024
2.9K
相关概念视频
Predicting Molecular Geometry
36.2K
VSEPR Theory for Determination of Electron Pair Geometries
36.2K
Molecular Models
40.7K
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.
40.7K
Molecular Weight of Step-Growth Polymers
2.3K
Step growth polymerization involves bi or multifunctional monomers. Bifunctional monomers react to form linear step growth polymers, whereas multifunctional monomers react to form non-linear or branched polymers.
As the step-growth polymerization involves step-wise condensation of monomers, the molecular weight also builds up eventually. Consequently, high molecular weight polymers are obtained at the late stages of the polymerization, where 99% of monomers have been consumed.
The extent of the...
As the step-growth polymerization involves step-wise condensation of monomers, the molecular weight also builds up eventually. Consequently, high molecular weight polymers are obtained at the late stages of the polymerization, where 99% of monomers have been consumed.
The extent of the...
2.3K
IUPAC Nomenclature of Carboxylic Acids
10.2K
IUPAC names of carboxylic acids are systematically derived following a few rules discussed below.
For acyclic saturated monocarboxylic acids, the longest hydrocarbon chain containing the –COOH carbon is identified as the parent chain. Then, the last -e of the parent hydrocarbon name is replaced with a suffix -oic acid.
For acyclic saturated monocarboxylic acids, the longest hydrocarbon chain containing the –COOH carbon is identified as the parent chain. Then, the last -e of the parent hydrocarbon name is replaced with a suffix -oic acid.
10.2K
Conformations of Ethane and Propane
14.7K
In an organic molecule, free rotation about the carbon-carbon single bond results in energetically different conformers of the molecule. Due to this rotation, called the internal rotation, ethane has two major conformations — staggered and eclipsed.
Staggered conformation is a low energy and more stable conformation with the C-H bonds on the front carbon placed at 60°dihedral angles relative to the C-H bonds on the back carbon, leading to a reduced torsional strain. In staggered...
Staggered conformation is a low energy and more stable conformation with the C-H bonds on the front carbon placed at 60°dihedral angles relative to the C-H bonds on the back carbon, leading to a reduced torsional strain. In staggered...
14.7K
Predicting Reaction Outcomes
8.6K
Kinetics describes the rate and path by which a reaction occurs. In contrast, thermodynamics deals with state functions and describes the properties, behavior, and components of a system. It is not concerned with the path taken by the process and cannot address the rate at which a reaction occurs. Although it does provide information about what can happen during a reaction process, it does not describe the detailed steps of what appears on an atomic or a molecular level. On the other hand,...
8.6K
