Genetic Programming as a tool for identification of analyte-specificity from complex response patterns using a non-specific whole-cell biosensor

► Genetic Programming designs algorithms by a heuristic machine-learning approach. ► GP-generated classifiers identify analyte specific patterns emerging from an array-type biosensor. ► A significant classification of several herbicides is successfully demonstrated. ► Analyte-specificity can be trained either on the level of individual substances or chemical classes.