相关实验视频

Updated: Sep 12, 2025

The Lambda Select cII Mutation Detection System

The Lambda Select cII Mutation Detection System

Published on: April 26, 2018

"通过全面的细胞绘画分析计算预测突变性"

Natacha Cerisier¹, Emily Truong¹, Taku Watanabe²

¹Université Paris Cité, INSERM U1133, CNRS UMR 8251, 75013, Paris, France.

|August 4, 2025

概括

细胞绘画成像与机器学习相结合,准确预测化学变异性,优于传统方法. 这种高含量分析揭示了与DNA损伤相关的细胞变化,为毒理学提供了一种新的方法.

科学领域:

毒理学和药物开发研究
计算生物学计算生物学
化学安全化学安全

背景情况:

变异性评估对于毒理学,药物开发和环境安全至关重要.
传统的突变性测试是耗时且昂贵的.
像细胞绘画这样的高含量成像测试为预测毒理学提供了潜在的潜力.

研究的目的:

开发和评估机器学习模型,以使用细胞绘画数据预测化学变异性.
将基于细胞绘画的模型与传统基于结构的定量结构-活动关系 (QSAR) 工具的性能进行比较.
为了识别表征突变性的形态特征.

主要方法:

利用了两个大型数据集 (布罗德研究所和美国环保署) 采用细胞绘画配置文件和化学结构.
训练并比较了三个机器学习算法:随机森林,支持向量机器和极端梯度增强.
确定了表型改变度 (PAC) 以优化预测准确性.

主要成果:

极端梯度增强模型在预测两个数据集的突变性方面表现出卓越的表现.
细胞绘画模型在大多数化合物中表现优于传统的QSAR工具,如VEGA和CompTox仪表板.
形态分析揭示了变异性和DNA/RNA,ER,线粒体和核中的扰动之间的关联.

结论:

基于Cell Painting的机器学习模型显示出预测化学变异性的巨大潜力.
这些模型为基于结构的方法提供了一个互补的方法,为细胞机制提供了洞察力.
未来的研究应该专注于标准化细胞绘画协议,并探索深度学习以提高准确性.

更多相关视频

Implementation of In Vitro Drug Resistance Assays: Maximizing the Potential for Uncovering Clinically Relevant Resistance Mechanisms

Implementation of In Vitro Drug Resistance Assays: Maximizing the Potential for Uncovering Clinically Relevant Resistance Mechanisms

Published on: December 9, 2015

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

相关概念视频

Mutagenicity and Carcinogenicity

Mutagenicity and Carcinogenicity

Mutagenicity and carcinogenicity refer to the ability of drugs to cause genetic defects and induce cancer, respectively. The International Agency for Research on Cancer (IARC) classifies agents into four groups based on their carcinogenic potential. Group 1 agents are known human carcinogens; group 2A agents are probably carcinogenic to humans; group 3 agents lack data to support their role in carcinogenesis; and group 4 includes agents for which data support that they are not likely to be...