Quantitative structure–retention relationship (QSRR) studies have proved to be a valuable approach in the prediction of the high resolution gas chromatographic relative retention time (HRGC-RRT) of organic chemicals. Polychlorinated biphenyls (PCBs) are mainly ubiquitous pollutants due to their properties according to inflammability, chemical stability and solubility in most organic solvents. Moreover, these organochlorinated compounds are bioaccumulated because of their affinity to lipids and slow degradation rate. PCBs are also globally distributed in the environment and found in human, wildlife, air, water, soil and sediments. To estimate and predict the HRGC-RRT values of all 209 PCBs on 18 different stationary phases, 282 molecular descriptors from DRAGON program and the topological structures of PCBs molecules were calculated. By means of the final variable selection method which is genetic and elimination selection stepwise regression algorithms, four optimal descriptors were selected to develop a QSRR model for the prediction of GC-RRT of PCBs on each stationary phase. The accuracy of all the developed models was cross-validation using leaveone- out, Y-randomization and external validation through an odd-even number and division of the entire dataset into training and test sets.
