High confidence predictions of drug-drug interactions: predicting affinities for cytochrome P450 2C9 with multiple computational methods.

Four different models are used to predict whether a compound will bind to 2C9 with a K(i) value of less than 10 microM. A training set of 276 compounds and a diverse validation set of 50 compounds were used to build and assess each model. The modeling methods are chosen to exploit the differences in how training sets are used to develop the predictive models. Two of the four methods develop partitioning trees based on global descriptions of structure using nine descriptors. A third method uses the same descriptors to develop local descriptions that relate activity to structures with similar descriptor characteristics. The fourth method uses a graph-theoretic approach to predict activity based on molecular structure. When all of these methods agree, the predictive accuracy is 94%. An external validation set of 11 compounds gives a predictive accuracy of 91% when all methods agree.

Line-walking method for predicting the inhibition of P450 drug metabolism.

A new method, called line-walking recursive partitioning (LWRP), for partitioning diverse structures on the basis of chemical properties that uses only nine descriptors of the shape, polarizability, and charge of the molecule is described. We use a training set of over 600 compounds and a validation set of 100 compounds for the cytochrome P450 enzymes 2C9, 2D6, and 3A4. The LWRP algorithm itself incorporates elements from support vector machines (SVMs) and recursive partitioning, while circumventing the need for the linear or quadratic programming methods required in SVMs. We compare LWRP with a many-descriptor SVM model, using the same dataset as that described in the literature.(1) The line-walking method, using nine descriptors, predicted the validation set with about 84-90% accuracy, a success rate comparable to that of the SVM method. Furthermore, line-walking was able to find errors in the assignment of inhibitor values within the validation set for the 2C9 inhibitors. When these errors are corrected, the model predicts with an even higher level of accuracy. Although this method has been applied to P450 enzymes, it should be of general use in partitioning molecules on the basis of function.