ABSTRACT
If all theoretically possible biotransformations are shown for each consecutive product a combinatorial explosion of products occurs. MetabolExpert and META terminate a sequence when a Phase II metabolite was produced or when the product was predicted to be very water soluble, respectively. METEOR has facilities to apply a combination of several prioritization methods to identify the most likely metabolites. First, the likelihood of the biotransformations are qualitatively described, for example, as probable, plausible, equivocal, or doubted. A filter is applied to display only those at or above a user-selected level of likelihood (plausible is the default level). Second, the likelihood of a biotransformation is defined differently according to the octanol-water partition coefficient of the substrate, which is divided into a value of high, medium or low. Log Kow is estimated for every submitted structure or predicted (Phase I) metabolite using either a call to an external program (ClogP for Windows, BioByte Corp.) or a less accurate internal calculation using the Moriguchi method (Moriguchi, 1992). Generally enzymic biotransformations become less likely with increasing hydrophilicity (Testa et al. 2000a; 2000b; Testa and Cruciani, 2001), so a biotransformation is ranked plausible for a lipophilic substrate (log Kow >2) and ranked doubted for a hydrophilic one (log Kow < 0). Metabolism generally proceeds to make more water soluble metabolites with a resultant reduction in rates of metabolism for many subsequent biotransformations.
