Statistical analysis of drug treated cell morphologies from HCS image data

We have developed a framework for analyzing image data from High Content Screening (HCS) experiments. The Kolomogorov-Smirnov Statistic is used to identify statistically significant image parameters for use in K-means clustering. Clusters that are underrepresented in drug-treated cell populations can be "enriched" via normalizing by the control clusters. This general methodology can be applied at different drug treatment conditions to identify "interesting" clusters. We demonstrate how the resulting clusters of morphologies aid in the understanding of the underlying biology of drug-treated cell populations PRIB 2008 proceedings found at: http://dx.doi.org/10.1007/978-3-540-88436-1

Contributors: Monash University. Faculty of Information Technology. Gippsland School of Information Technology ;
Chetty, Madhu ;
Ahmad, Shandar ;
Ngom, Alioune ;
Teng, Shyh Wei ;
Third IAPR International Conference on Pattern Recognition in Bioinformatics (PRIB) (3rd : 2008 : Melbourne, Australia) ;
Coverage:
Rights: Copyright by Third IAPR International Conference on Pattern Recognition in Bioinformatics. All rights reserved.

本研究团队构建了一套用于分析高内涵筛选（HCS）实验图像数据的框架。在该框架中，采用柯尔莫哥洛夫-斯米尔诺夫统计量以识别具有统计学意义的图像参数，并应用于K-means聚类分析。对于在药物处理细胞群体中代表性不足的聚类，通过控制聚类的标准化方法实现“富集”。该通用方法可应用于不同的药物处理条件，以识别“有趣”的聚类。本研究展示了形态学聚类结果如何有助于理解药物处理细胞群体的潜在生物学机制。相关研究成果可在PRIB 2008会议论文集中查阅，链接为：http://dx.doi.org/10.1007/978-3-540-88436-1。