AvalonRx®
Compound Profiling
Compound profiling enables the classification of compounds in the context of known pathways, mechanisms of action, and functional and structural categories.

It is used to characterize compounds that are identified in our HITS® screening process and from novel compounds derived from external sources. We also use compound profiling to guide the chemical synthesis of analogs during our drug optimization process.

Compound profiling begins with the comprehensive analysis of compound effects in relevant cell types using our high-throughput gene expression microarray capabilities. In this way we monitor thousands of genes simultaneously for both "on target" and "off target" effects of compounds. From this characterization process we obtain insight on potential mechanism of action, selectivity, specificity and novelty as well as early information on toxicity, tumor targets, and predictions of in vivo efficacy.

The observed profiles are analyzed using our custom built compound profiling toolkit. This compound profiling toolkit is used for mining of Avalon gene expression databases as well as external gene signature sources during all stages of drug discovery and development. The Avalon compound profiling toolkit consists of proprietary statistical methods and datasets that are applied to identify associations between novel compounds and known classes and functions, and activities. A key component of this system is the Avalon reference compound database. This database consists of transcriptional profiles derived from cellular treatments with hundreds of known reference compounds. These reference compounds include well-known physiological modulators, drug candidates in the earlier stages of development, failed clinical candidates as well as marketed drugs. These profiles have been classified according to the compound mechanism of action, and pathway and biological functions. Model gene signatures have been developed to represent each class, function, and activity. We have also integrated into our analysis signatures based on known pathway and functional analysis, and on data derived from relevant public data sources. This system was designed to be a flexible and allow the integration of new tools and datasources as they become available.