Alignment-Based Prediction of Sites of Metabolism

Prediction
of metabolically labile atom positions in a molecule
(sites of metabolism) is a key component of the simulation of xenobiotic
metabolism as a whole, providing crucial information for the development
of safe and effective drugs. In 2008, an exploratory study was published
in which sites of metabolism were derived based on molecular shape-
and chemical feature-based alignment to a molecule whose site of metabolism
(SoM) had been determined by experiments. We present a detailed analysis
of the breadth of applicability of alignment-based SoM prediction,
including transfer of the approach from a structure- to ligand-based
method and extension of the applicability of the models from cytochrome
P450 2C9 to all cytochrome P450 isozymes involved in drug metabolism.
We evaluate the effect of molecular similarity of the query and reference
molecules on the ability of this approach to accurately predict SoMs.
In addition, we combine the alignment-based method with a leading
chemical reactivity model to take reactivity into account. The combined
model yielded superior performance in comparison to the alignment-based
approach and the reactivity models with an average area under the
receiver operating characteristic curve of 0.85 in cross-validation
experiments. In particular, early enrichment was improved, as evidenced
by higher BEDROC scores (mean BEDROC = 0.59 for α = 20.0, mean
BEDROC = 0.73 for α = 80.5).