摘要
Using the molecular electronegativity distance vector descriptors derived directly from the molecular topological structures, the relative retention time （RRT） of polybrominated diphenyl ethers （PBDEs） were predicted. A four-variable regression model （M30） with the correlation coefficient of 0.9816 and the root mean square errors of 0.061 was developed using a training set including 30 PBDEs. The correlation coefficient of 0.9841 and the root mean square errors of 0.054 between the values of RRT predicted by M30 and the RRT observed for 16 external PBDEs show a good predictive potential of M30. The descriptors included in the M30 represent four interactions between four pairs of atom types, i.e., atom -C= and -C=, -C= and 〉C=, 〉C= and 〉C=, -C= and -Br.
Using the molecular electronegativity distance vector descriptors derived directly from the molecular topological structures, the relative retention time （RRT） of polybrominated diphenyl ethers （PBDEs） were predicted. A four-variable regression model （M30） with the correlation coefficient of 0.9816 and the root mean square errors of 0.061 was developed using a training set including 30 PBDEs. The correlation coefficient of 0.9841 and the root mean square errors of 0.054 between the values of RRT predicted by M30 and the RRT observed for 16 external PBDEs show a good predictive potential of M30. The descriptors included in the M30 represent four interactions between four pairs of atom types, i.e., atom -C= and -C=, -C= and 〉C=, 〉C= and 〉C=, -C= and -Br.
基金
the National Basic Research Program of China(2003CB415002)
the National Natural Science Foundation of China(No.20377022)
the Guangxi Natural Science Fund(No.0236063)for their financial supports.
