Computationally predicting the metabolic fates of drugs is a very complex task. This complexity is not only due to the huge and diverse biochemical network in the living cell, but also due to the fact that the majority of in vivotransformationsoccur by the action of hepatocytes and gastrointestinal microflora, thus not by a single cell type or even organism. However, the prediction of metabolic fates is definitely a problem worth solving, since it would allow facilitating the development of drugs and rely less on animal testing. As a first step in this direction, PharmBiosimis being developed as a biosimulation tool which is based on massive data reduction and on attributing metabolic fates of drug molecules to functional groups and substituents. Initial work is done with yeast as a model organism and is restricted to drugs that are mainly transformed by central metabolism, especially sugar metabolism. The reason for the latter is that the qualitative functioning of the involved biochemistry is very similar in diverse cell types involved in drug metabolism. Results from the model are compared and validated with data from mammalian systems.