QSPR study of supercooled liquid vapour pressures of PBDEs by using molecular distance-edge vector index

• Main Authors: Jiao Long, Wang Xiaofei, Bing Shan, Xue Zhiwei, Li Hua
• Format: Article
• Language: English
• Published: Serbian Chemical Society 2015-01-01
• Series: Journal of the Serbian Chemical Society
• Subjects: QSPR; PBDEs; supercooled liquid vapour pressures; molecular distance-edge vector index; artificial neural network
• Online Access: http://www.doiserbia.nb.rs/img/doi/0352-5139/2015/0352-51391400087J.pdf
• ISSN: 0352-5139; 1820-7421

The quantitative structure property relationship (QSPR) for supercooled liquid vapour pressures (PL) of PBDEs was investigated. Molecular distance-edge vector (MDEV) index was used as the structural descriptor. The quantitative relationship between the MDEV index and lgPL was modeled by using multivariate linear regression (MLR) and artificial neural network (ANN) respectively. Leave-one-out cross validation and k-fold cross validation were carried out to assess the prediction ability of the developed models. For the MLR method, the prediction root mean square relative error (RMSRE) of leave-one-out cross validation and k-fold cross validation is 9.95 and 9.05 respectively. For the ANN method, the prediction RMSRE of leave-one-out cross validation and k-fold cross validation is 8.75 and 8.31 respectively. It is demonstrated the established models are practicable for predicting the lgPL of PBDEs. The MDEV index is quantitatively related to the lgPL of PBDEs. MLR and L-ANN are practicable for modeling this relationship. Compared with MLR, ANN shows slightly higher prediction accuracy. Subsequently, an MLR model, which regression equation is lgPL = 0.2868 M11 - 0.8449 M12 - 0.0605, and an ANN model, which is a two inputs linear network, were developed. The two models can be used to predict the lgPL of each PBDE.