Computational toxicology: heading toward more relevance in drug discovery and development.

Computational tools for predicting toxicity have been envisioned to have the potential to broadly impact the attrition rate of compounds in early research and development, and prove successful in predicting adverse drug reactions (ADRs) in patients enrolled in clinical trials, and particularly prior to the marketing of drugs. The impact of such tools to date, however, has been modest and relatively narrow in scope. It is important to note that advances within medical science and newer approaches in clinical development will require predictive toxicology applications to be viable, and therefore efforts must be directed into making these tools relevant to the goal of preventing undesired toxicity in patients. In this Editorial Opinion, the current status of computational toxicology within industry is reviewed and areas in which advances can be made are highlighted. While predicting the potential of a compound to induce specific ADRs continues to be a formidable task, the field of computational biology is now heading in a direction more relevant to human disease and adverse outcomes.