相关实验视频

Updated: Sep 17, 2025

Dual-color Correlative Light and Electron Microscopy for the Visualization of Interactions between Mitochondria and Lysosomes

Dual-color Correlative Light and Electron Microscopy for the Visualization of Interactions between Mitochondria and Lysosomes

Published on: September 27, 2024

用有限的计算资源将SMILES化学变压器的域调整为SELFIES

Obaid Khaleifah Alhmoudi¹, Mahmoud Aboushanab¹, Muhammed Thameem¹

¹Department of Chemical & Petroleum Engineering, Khalifa University of Science and Technology, P.O. Box 127788, Abu Dhabi, United Arab Emirates.

Scientific reports

|July 2, 2025

概括

域适应预训练使变压器模型,最初训练在简化分子输入线输入系统 (SMILES),有效地利用自我引用嵌入字符串 (SELFIES) 进行分子性质预测,提供一个成本效益高的替代方案.

科学领域:

计算化学的计算化学
机器学习在化学中的应用
化学信息学化学信息学

背景情况:

准确的分子性质预测依赖于强大的分子表示.
简化分子输入线输入系统 (SMILES) 允许多个表示,但不保证有效性.
自引用嵌入字符串 (SELFIES) 提供基于语法的方法,确保结构有效性.

研究的目的:

在SELFIES上研究SMILES预训练变压器 (ChemBERTa--基-v1) 的域适应预训练.
用预训练模型评估SELFIES作为用于分子性质预测的输入表示的有效性.
与SMILES相比,评估SELFIES适应的成本效益和性能.

主要方法:

在PubChem的700,000个SELFIES分子上进行ChemBERTa-zinc-base-v1的域适应性预训练.
使用t-SNE,等号相似性和QM9属性回归与冷变压器权重的嵌入级别评估.
在ESOL,FreeSolv和Lipophilicity数据集上进行端到端微调,用于下游物业预测.

主要成果:

适应SELFIES的模型表现优于原来的SMILES基线.
性能略高于ChemBERTa-77M-MLM,尽管在预训练尺度上有100倍的差异.
在ESOL,FreeSolv和Lipophilicity数据集上分别获得了较低的RMSE值 (0.944,2.511,0.746).

结论:

基于SELFIES的域调整是一种成本效益高的分子性质预测方法.
这种方法绕过了对分子描述器,3D特征或广泛的计算基础设施的需求.
将预先训练的模型调整为SELFIES,为化学信息学任务提供了可行且高性能的替代方案.

更多相关视频

Cardiac Muscle Cell-based Actuator and Self-stabilizing Biorobot - Part 2

Cardiac Muscle Cell-based Actuator and Self-stabilizing Biorobot - Part 2

Published on: May 9, 2017

A Guide to Structured Illumination TIRF Microscopy at High Speed with Multiple Colors

A Guide to Structured Illumination TIRF Microscopy at High Speed with Multiple Colors

Published on: May 30, 2016

相关概念视频

Nonconscious Mimicry

Nonconscious Mimicry

Nonconscious mimicry occurs when individuals alter their mannerisms to match the behaviors and expressions of those nearby, without intention.

Facial Feedback Hypothesis

Facial Feedback Hypothesis

Charles Darwin proposed that facial expressions are an evolutionary adaptation for communication. He argued that these expressions are not influenced by culture but are universal across species. For example, a snarling expression with exposed teeth signals a threat in many animals, including humans. Darwin also suggested that displaying an emotion can intensify the feeling. Smiling, for example, could enhance one's sense of happiness. This idea laid the foundation for understanding the role...

Transformation

Transformation

Microbial communities are dynamic environments where cell lysis releases free DNA into the surroundings. Other cells can take up this extracellular DNA through a process known as transformation.When a cell incorporates this foreign DNA into its genome, resulting in genetic modification, the process is known as transformation. Cells capable of this process are termed competent. Competence can be natural, as observed in certain bacteria and archaea, or artificially induced in the...