Phenotypic Fingerprinting of Small Molecule Cell Cycle Kinase Inhibitors for Drug Discovery

Jonathan Low1, Arunava Chakravartty2, Wayne Blosser1, Michele Dowless1, Christopher Chalfant3, Patty Bragger3, Louis Stancato*, 1
1 Departments of Cancer Growth and Translational Genetics, Eli Lilly and Company, Indianapolis, IN 46265, USA
2 Statistics and Information Science, Eli Lilly and Company, Indianapolis, IN 46265, USA
3 Discovery Informatics, Eli Lilly and Company, Indianapolis, IN 46265, USA

Article Metrics

CrossRef Citations:
Total Statistics:

Full-Text HTML Views: 390
Abstract HTML Views: 392
PDF Downloads: 106
Total Views/Downloads: 888
Unique Statistics:

Full-Text HTML Views: 237
Abstract HTML Views: 220
PDF Downloads: 85
Total Views/Downloads: 542

© Low et al.; Licensee Bentham Open.

open-access license: This is an open access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestrictive use, distribution, and reproduction in any medium, provided the original work is properly cited.

* Address correspondence to this author at the Cancer Growth and Translational Genetics Lilly Corporate Center Bldg. 98C, DC0434 Indianapolis, IN 46285, USA; Tel: 317-655-6910; Fax: 317-276-1414; E-mail:


Phenotypic drug discovery, primarily abandoned in the 1980’s in favor of targeted approaches to drug development, is once again demonstrating its value when used in conjunction with new technologies. Phenotypic discovery has been brought back to the fore mainly due to recent advances in the field of high content imaging (HCI). HCI elucidates cellular responses using a combination of immunofluorescent assays and computer analysis which increase both the sensitivity and throughput of phenotypic assays. Although HCI data characterize cellular responses in individual cells, these data are usually analyzed as an aggregate of the treated population and are unable to discern differentially responsive subpopulations. A collection of 44 kinase inhibitors affecting cell cycle and apoptosis were characterized with a number of univariate, bivariate, and multivariate subpopulation analyses demonstrating that each level of complexity adds additional information about the treated populations and often distinguishes between compounds with seemingly similar mechanisms of action. Finally, these subpopulation data were used to characterize compounds as they relate in chemical space.