Despite several thousands of drugs are in use currently, research on new drug molecules is continuing. Because, there are diseases still without medication, successor/better drugs make the predecessor ones obsolete, and advancement in both life sciences and analytical technologies provide identification of previously unknown mechanisms of diseases, and discovery of novel drug targets. The two main criteria which a drug candidate should meet are high affinity for the target, and no or acceptable/tolerable toxicity in humans. Among these two, toxicity is the limiting one; developing a drug candidate with unacceptable toxicity has to be discontinued, even if it has an extremely high pharmacological activity. Drug would be withdrawn, if serious toxicity arises after marketing. Since drug development is a long (approximately 10 years), expensive, and infertile (one lead in 10.000 molecules) process, it is extremely important to detect the potential toxicity of drug candidate as early as possible. Today, it is believed that a great majority of toxic effects are caused by reactive intermediates generated by biotransformation of the parent drug. However, there are experimental difficulties in identifying such metabolite(s) in vivo. Their formation is affected by multi-factorial events; they can further be metabolized to structurally different products, and/or they may bind to a huge variety of biological sites or macromolecules. Hence, some reactive intermediates and their corresponding stable derivatives are generated in trace amounts, which make their determination more difficult. The ability of cytochrome P450s (CYP450) and other biotransformation enzymes to function in vitro offers a great flexibility to researchers, biotransformation of any compound can be simulated in a test tube, and metabolites/reactive intermediates are generated in an environment which has relatively much less background and less interfering multi-factorial events compared to in vivo. To simulate biotransformation, microsomal fraction is used most frequently from human and non-human sources. Purified or recombinant enzymes are used in determining the individual isoenzymes responsible for certain metabolites. Because of the chemical reactivity of intermediates, relevant, usually nucleophilic trapping agent(s) such as glutathione (GSH), N-acetylcysteine (NAC) and cyanide (CN-) are used to stabilize the metabolite. Trapped metabolites are subjected to spectrometric and/or nuclear magnetic resonance spectroscopic analyses for structural identification. Vertiginous advances especially in mass spectrometric technologies offer researchers new challenges in this area. This review is aimed at briefly summarizing the state of the art and compiling the highlighted studies in characterization of the reactive metabolites from drug molecules.