PARM - A GENETIC EVOLVED ALGORITHM TO PREDICT BIOACTIVITY

Based on Waiters' GERM (Genetic Evolved Receptor Model) algorithm, an improved algorithm FARM (Pseudo Atomic Receptor Model) was put forward . PARM uses a combination of a genetic algorithm and a cross-validatio n technique to produce an atomic-level pseudoreceptor model, based on a set of known structure-activity relationships. During the genetic pr ocess, an artificial interfering method, which is based on a complemen tary principle of ligand-receptor interaction, was used to accelerate the search speed. The evolved models show a high correlation between i ntermolecular energy and bioactivity and can predict the bioactivity o f an unknown molecule by interpolating in the regression equation of t he structure-activity relationship. This algorithm was applied to two systems and produced reasonable results.