相关实验视频

Updated: May 1, 2026

A High-throughput Assay for the Prediction of Chemical Toxicity by Automated Phenotypic Profiling of Caenorhabditis elegans

A High-throughput Assay for the Prediction of Chemical Toxicity by Automated Phenotypic Profiling of Caenorhabditis elegans

Published on: March 14, 2019

基于AI的纳米毒性数据提取和纳米毒性的预测

Eunyong Ha¹, Seung Min Ha¹, Zayakhuu Gerelkhuu^2,3

¹Department of Chemistry, Hanyang University, Seoul 04763, Republic of Korea.

Computational and structural biotechnology journal

|April 21, 2025

概括

人工智能 (AI) 使用大语言模型 (LLM) 自动从研究文章中提取纳米毒性数据. 这种方法提高了数据收集效率,并改善了纳米毒性预测模型.

科学领域:

计算毒理学计算毒理学
纳米材料的安全性
科学领域的人工智能

背景情况:

纳米材料 (NM) 的使用正在增加,需要有效的毒性评估.
传统的毒性测试是昂贵的,并引发了伦理方面的担忧.
计算模型为预测纳米毒性提供了一个替代方案.

研究的目的:

为纳米毒性研究开发基于人工智能的自动化数据提取管道.
探索大型语言模型 (LLM) 的使用,并为提取纳米毒性数据提供提示工程.
从科学文献中有效收集高质量的纳米毒性数据.

主要方法:

使用Python和LangChain开发了一个人工智能驱动的数据提取管道.
利用了216个纳米毒性研究文章进行快速改进和LLM绩效评估.
从605篇研究文章中提取了测试数据,使用了优化的LLM和提示.

主要成果:

在训练数据上获得了F1数据提取 (F1_DE) 的F1分数,在84.6%和87.6%之间.
构建自动机器学习 (AutoML) 模型,其纳米毒性预测 (F1_NP) 的F1得分超过86.1%.
评估模型的可靠性和适用性,基于基本事实,尺寸和平衡.

结论:

基于AI的数据提取显著提高了收集纳米毒性信息的效率.
LLM和快速工程显示了推动纳米毒性研究的巨大潜力.
开发的管道和模型有助于更可靠和可访问的纳米毒性评估.

关键词:

自动机器学习自动化机器学习数据提取数据提取长链 (LangChain) 是一个长链.大型语言模型纳米毒性纳米毒性快递工程是指快递的工程.

更多相关视频

Demonstration of the Sequence Alignment to Predict Across Species Susceptibility Tool for Rapid Assessment of Protein Conservation

Demonstration of the Sequence Alignment to Predict Across Species Susceptibility Tool for Rapid Assessment of Protein Conservation

Published on: February 10, 2023

Author Spotlight: Advances in Evaluating Human Lung Epithelial Cells' Response to Metal-Organic Frameworks

Author Spotlight: Advances in Evaluating Human Lung Epithelial Cells' Response to Metal-Organic Frameworks

Published on: May 26, 2023

相关概念视频

Toxicity Testing in Animals

Toxicity Testing in Animals

Toxicity tests in animals are grounded on two main assumptions: first, the effects observed in laboratory animals can be extrapolated to humans, especially when adjusted for body surface area; second, high-dose exposure in animals is essential to identify potential human hazards from lower doses. This is based on the quantal dose-response concept, which faces the challenge of extrapolating results from relatively few test animals to much larger human populations. For example, a 0.01% incidence...