PREDICTION OF AQUEOUS SOLUBILITY OF ORGANIC-COMPOUNDS FROM MOLECULAR-STRUCTURE

Multiple linear regression (MLR) and computational neural networks (CN N) are utilized to develop mathematical models to relate the structure s of a diverse set of 332 organic compounds to their aqueous solubilit ies. Topological, geometric, and electronic descriptors are used to nu merically represent structural features of the data set compounds. Gen etic algorithm and simulated annealing routines, in conjunction with M LR and CNN, are used to select subsets of descriptors that accurately relate to aqueous solubility. Nonlinear models with nine calculated st ructural descriptors are developed that have a training set root-mean- square error of 0.394 log units for compounds which span a -log(molari ty) range from -2 to +12 log units.