相关实验视频

Updated: Jun 3, 2025

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps

Published on: October 28, 2018

使用生成模型预测细胞对干扰的形态反应

Alessandro Palma^1,2, Fabian J Theis^3,4,5, Mohammad Lotfollahi^6,7,8

¹Department of Computational Health, Institute of Computational Biology, Helmholtz Zentrum München, Munich, Germany.

Nature communications

|January 8, 2025

概括

我们开发了一种新的AI模型,即图像扰乱自编码器 (IMPA),用于分析药物发现查中的复杂显微镜图像. IMPA准确地预测细胞变化,提高高内容成像分析的效率.

科学领域:

细胞成像细胞成像
药物发现药物发现
计算生物学是一种计算生物学.

背景情况:

高含量显微镜产生丰富的表型数据用于药物发现.
由于采样不完整和技术差异,分析大型成像数据集具有挑战性.

研究的目的:

开发一种生成模型,用于预测高含量成像屏幕中的形态变化.
解决分析复杂查数据的挑战,并考虑技术差异.

主要方法:

开发了IMage Perturbation Autoencoder (IMPA),这是一个生成式风格传输模型.
应用IMPA来预测由遗传和化学扰动引起的形态变化.
评估IMPA对乳腺癌和骨髓瘤细胞系的表现,评估其处理批量效应的能力.

主要成果:

对于已知和新出现的扰动,IMPA准确地捕捉了形态和人口水平的变化.
该模型通过考虑批量效应和技术变化,在各种不同的实验条件中展示了稳定性.
IMPA成功地模拟了技术变化的各种来源中的扰动.

结论:

IMPA是一个强大的工具,用于分析大规模的高内容成像屏幕.
该模型通过in-silico扰动预测促进了高效的实验设计.
IMPA准备在药物发现管道中加强对显微镜数据的分析.

更多相关视频

Author Spotlight: Enhancing PSC-to-Functional Cell Differentiation Using ML Models Based on Live-Cell Bright-Field Imaging

Author Spotlight: Enhancing PSC-to-Functional Cell Differentiation Using ML Models Based on Live-Cell Bright-Field Imaging

Published on: October 4, 2024

Patterning the Geometry of Human Embryonic Stem Cell Colonies on Compliant Substrates to Control Tissue-Level Mechanics

Patterning the Geometry of Human Embryonic Stem Cell Colonies on Compliant Substrates to Control Tissue-Level Mechanics

Published on: September 28, 2019

相关概念视频

Morphogenesis

Morphogenesis

Plant morphogenesis—the development of a plant’s form and structure—involves several overlapping developmental processes, including growth and cell differentiation. Precursor cells differentiate into specific cell types, which are organized into the tissues and organ systems that make up the functional plant.