DESCRIPTION (provided by applicant): Adverse drug reactions (ADRs) are one of the leading causes of hospitalization and death in the United States. ADRs are often associated with unfavorable drug bioavailability or biodistribution profiles. Thus, ADRs could be prevented by optimizing drug transport properties from the systemic, organ level down to the microscopic, cellular level. To improve the quality of drugs entering clinical trials, a new generation of microscopic imaging instruments known as "high content screening" or "HCS" systems has been developed. HCS instruments can provide preclinical, human cell-based data to complement animal studies in predictive toxicology testing. As a high-throughput platform, HCS systems can be used to screen large collections of small molecules in physiologically-relevant assays. Now the challenge is to advance HCS technology to facilitate development of less toxic drug candidates with improved clinical success rates. To face this challenge, we propose to develop a cheminformatics and image data management and analysis plan to study the subcellular localization of small molecules in living cells. Inspired by machine vision approaches currently being used as a tool to analyze the subcellular distribution of proteins on a genome-wide scale ("location proteomics"), we propose that machine vision could also be adopted as a tool to analyze the distribution of small molecule drug candidates possessing a detectable optical signature. In analogy to how protein location is encoded by signal peptides, we hypothesize that subcellular small molecule localization is encoded by "Chemical Address Tags" to be discovered within the chemical structure of small molecules. To test this hypothesis, we plan to: 1) Develop automated, image analysis and cheminformatics tools to reverse-engineer Chemical Address Tags in an objective, quantitative and high-throughput manner; 2) Develop and compare two quantitative, machine vision approaches to assay the transport properties of organelle-targeting molecules; 3) Demonstrate how a cheminformatics-driven, image data management and analysis plan can impact a drug lead optimization effort.

Scientific advances to reduce the risk of adverse drug reactions (ADRs) will transform health care by: i) reducing the incidence of drug-related morbidity and mortality; ii) reducing withdrawal of otherwise effective drugs from the market; and, iii) facilitating regulatory approval of new, safer drugs. Strategies to reduce the incidence f side-effects by improving drug transport properties will be of great benefit to patients across all therapeutic areas.