Integrating metabolism and toxicity in multi-organ systems.

A wide range of experimental and predictive approaches are currently employed by the pharmaceutical industry in an attempt to avoid developing molecules with likely toxicities. Experimental systems may be adapted to handle the increasing number of molecules that require screening. The cost implication of such extensive screening is, however, immense and in silico models must therefore become the first lines of investigation to assess the metabolic and toxicological potential of new compound libraries. In silico systems have been based on the relationship between the chemical structure of novel molecules and toxicological outcomes in terms of types, but not always, locations of lesions. There should be sufficient published and in-house data available on the distribution of diverse molecules to identify the rules that govern organ, cellular and subcellular distribution of molecules (and their metabolites) as the basis of tissue and cell targeting and, more importantly, the possibility of toxicological processes. In addition, the extensive cell biology knowledge needs to be linked to current in silico systems to enhance our understanding of metabolic activation and other processes that affect likely toxicological consequences in different tissues and specific cell types. There is a need to provide a single knowledge platform that can access all current in silico products that use different datasets. The closer integration of these diverse tools will help predict other approaches to help identify the most likely toxicological potential of novel molecules, and ensure the 'least' toxic molecules are developed.