Visualization and interpretation of high content screening data

High content screening is a method for identifying small molecule modulators of mammalian cell biology. The nature of the experiment generates an enormous amount of data in the form of photographic images of cells after treatment with compounds of interest. The interpretation of data from these experiments is challenging both in terms of automatically perceiving the images, extracting, and understanding differences between screened compounds and visualizing the results. This paper discusses the application of statistical and visual methods that have been used to interpret data from a simplified DNA stain (DAPI) screen to quickly identify compounds of interest. An understanding of the mechanism of action of the screened compounds can be obtained by comparing them to control compounds of known mechanism of action. Statistical and visual methods will be shown that facilitate easy comparison of screened compounds against these control compounds. As an example, a subset of the internal repository at ArQule was screened, together with control compounds that were known to induce characteristic mitotic arrest. Subsequent data processing described in this paper permitted the easy identification of compounds that were similar to (and very different from) the control compounds.