Adverse drug reactions (ADRs) are a significant human health problem. Any organ system can be affected, including the liver, skin and kidney. Drug-induced liver injury is the most frequent reason for the withdrawal of an approved drug from the market, and it also accounts for up to 50% of cases of acute liver failure. The clinical picture is often diverse, even for the same drug. Mild, asymptomatic effects occur at a relatively high frequency with a number of drugs. Idiosyncratic toxicity is rare but potentially life-threatening. Many serious ADRs that occur in man are unpredictable from routine pathology and clinical chemistry in laboratory animals and are therefore poorly understood. The drug metabolist can determine the propensity of a novel chemical entity to either accumulate in the hepatocyte or undergo bioactivation in numerous model systems, from expressed enzymes, genetically engineered cells to whole animals. Bioactivation can be measured using trapping experiments with model nucleophiles or by measurement of non-specific covalent binding. The chemistry of the process is defined and the medicinal chemist can address the issue by seeking a metabolically stable pharmacophore to replace the potential toxicophore. However, we require a more fundamental understanding of the role of drug chemistry and biochemistry in ADRs. This requires knowledge of the ultimate toxin, signalling in cell defense and the sequence of molecular events, which ultimately lead to cell and tissue damage. It is imperative that such studies have a clinical level, but then translated into laboratory-based molecular studies. This will provide a deeper understanding of potential toxicophores for drug design and define candidate genes for pharmacogenomic approaches to individualized medicines.