Prediction of blood-brain barrier permeability of organic compounds.

Using fragmental descriptors and artificial neural networks, a predictive model of the relationship between the structure of organic compounds and their blood-brain barrier permeability was constructed and the structural factors affecting the readiness of this penetration were analyzed. This model (N = 529, Q (2) = 0.82, RMSE cv = 0.32) surpasses the previously published models in terms of the prediction accuracy and the applicability domain and can be used for the optimization of the pharmacokinetic parameters during drug development.