Theoretically-derived molecular descriptors important in human intestinal absorption

A quantitative structure-human intestinal absorption relationship was devel oped using artificial neural network (ANN) modeling. A set of 86 drug compo unds and their experimentally-derived intestinal absorption values used in this study was gathered from the literature and a total of 57 global molecu lar descriptors, including constitutional, topological, chemical, geometric al and quantum chemical descriptors, calculated For each compound. A superv ised network with radial basis transfer function was used to correlate calc ulated molecular descriptors with experimentally-derived measures of human intestinal absorption. A genetic algorithm was then used to select importan t molecular descriptors. Intestinal absorption values (IA%) were used as th e ANN's output and calculated molecular descriptors as the inputs. The best genetic neural network (GNN) model with 15 input descriptors was chosen, a nd the significance of the selected descriptors for intestinal absorption e xamined. Results obtained with the model that was developed indicate that l ipophilicity, conformational stability and inter-molecular interactions (po larity, and hydrogen bonding) have the largest impact on intestinal absorpt ion. (C) 2001 Elsevier Science B.V. All rights reserved.