Quantitative structure activity relationship models based on heuristic method and gene expression programming for the prediction of the pK_a values of sulfa drugs

Quantitative structure-property relationships (QSPR) were developed to predict the pK_a values of sulfa drugs via heuristic method (HM) and gene expression programming (GEP).  The descriptors of 31 sulfa drugs were calculated by the software CODESSA, which can calculate constitutional, topological, geometrical, electrostatic, and quantum chemical descriptors.  HM was also used for the preselection of 4 appropriate    molecular descriptors.  Linear and nonlinear QSPR models were developed based on the HM and GEP separately and two prediction models lead to a good correlation coefficient (R) of 0.90 and 0.95.  The two QSPR models are useful in predicting pK_a during the discovery of new drugs and providing theory information for studying the new drugs.