QSPR Study on the Soil-Water Partition Coefficient of Polychlorinated Biphenyls by Using Artificial Neural Network

Abstract:

A practicable quantitative structure property relationship (QSPR) model for predicting the soil-water partition coefficient, Koc, of 16 polychlorinated biphenyls (PCBs) was developed. The structure of the investigated PCBs is encoded by five quantum structural descriptors and on topological index. The calibration model of Koc was developed by using artificial neural network (ANN). The input variables of ANN were generated from 6 structural descriptors by using principal component analysis (PCA). Leave one out cross validation was carried out to assess the predictive ability of the developed model. The prediction RMS%RE for the 16 PCBs is 6.35. The R2 between the predicted and experimental logKoc is 0.8522. It is demonstrated that ANN combined with PCA is a practicable method for developing QSPR model for Koc of these PCBs.

• Periodical: Advanced Materials Research (Volumes 455-456)
• Pages: 930-934
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.455-456.930
• Online since: January 2012
• Authors: Long Jiao
• Keywords: Artificial Neural Network (ANN), PCB, Principal Component Analysis (PCA), QSPR, Soil-Water Partition Coefficient

References

[1] European Commission, http: /ec. europa. eu/environment/pops/index _en. htm, updated on Jan. 15th (2011).

[2] X.Y. Han, Z.Y. Wang, Z.C. Zhai, and L.S. Wang, Estimation of n-octanol/water partition coefficients (Kow) of all PCB congeners by ab initio and a Cl substitution position method, QSAR Comb. Sci., vol. 25, p.333 – 341, (2006).

DOI: 10.1002/qsar.200530141

[3] B.G. Hansen, A.B. Paya-Perez, M. Rahman, and B.R. Larsen, QSARs for Kow and Koc of PCB congerners: a critical examination of data, assumptions and stastistical approaches, Chemosphere. vol. 39, pp.2209-2228, (1999).

DOI: 10.1016/s0045-6535(99)00145-9

[4] R. Zhou, L. Zhu, K. Yang, and Y.Y. Chen, Distribution of organochlorine, pesticides in surface water and sediments from Qiantang River, East China, J. Hazard. Mater., vol. 137, p.68–75, (2006).

DOI: 10.1016/j.jhazmat.2006.02.005

[5] J.R. Baker, J.R. Mihelcic, and A. Sabljic, Reliable QSAR for estimation Koc for persistent organic pollutans: correlation with molecular connetivity indices, Chemosphere, vol. 45, pp.213-221, (2001).

DOI: 10.1016/s0045-6535(00)00339-8

[6] V. Uddameri, and M. Kuchanur, Fuzzy QSARs for predicting log Koc of persistent organic pollutants, Chemosphere, vol. 54, p.771–776, (2004).

DOI: 10.1016/j.chemosphere.2003.08.023

[7] J.Y. Dai, S.K. Han, and L.S. Wang, Prediction of n-octanol/water partition coefficients and soil sorption coefficients for polychlorinated organic compounds (PCOCs) using quantum chemical parameters, Acta Sci. Circum., vol. 20, no. 6, pp.693-698, (2000).

[8] P. Wang, Q.H. Zhang, Y.W. Wang, T. Wang, X.M. Lia, L. Ding, and G.B. Jiang, Evaluation of soxhlet extraction, accelerated solvent extraction and microwave-assisted extraction for the determination of polychlorinated biphenyls and polybrominated diphenyl ethers in soil and fish samples, Anal. Chimi. Acta, vol. 663, pp.43-48, (2010).

DOI: 10.1016/j.aca.2010.01.035

[9] H.A. Martens, and P. Dardenne, Validation and verification of regression in small data sets, Chemometr. Intell. Lab. Syst., vol. 44, 99–121, (1998).

[10] M. Randic, The connectivity index 25 years after, J. Mole. Graph. Model., vol. 20, pp.19-35, (2001).

[11] L. Jiao, Gao S.Y., Z.H. Jia, and H. Li, Quantification of Benzoic Acid and Salicylic Acid by Capillary Zone Electrophoresis Combined with Artificial Neural Network, Comput. Appl. Chem., vol. 24, pp.1595-1599, (2007).

[12] Y. X Zhang, H. Li, A.X. Hou, and J. Havel, Artificial neural networks based on principal component analysis input selection for quantification in overlapped capillary electrophoresis peaks,. Chemometr. Intell. Lab. Sys., vol. 82, pp.165-175, (2006).

DOI: 10.1016/j.chemolab.2005.08.012

[13] L. Jiao, and H. Li, QSPR studies on the aqueous solubility of PCDD/Fs by using artificial neural network combined with stepwise regression, Chemometr. Intell. Lab. Sys., vol. 103, pp.90-95, (2010).

DOI: 10.1016/j.chemolab.2010.05.019

[14] M. Jalali-Heravi, and Z. Garkani-Nejad, Prediction of electrophoretic mobilities of alkyl- and alkenylpyridines in capillary electrophoresis using artificial neural networks, J. Chromatogr. A, vol. 971, pp.207-215, (2002).

DOI: 10.1016/s0021-9673(02)01043-9

[15] Y.X. Zhang, H. Li, A. X. Hou, and J. Havel, Artificial neural networks based on genetic input selection for quantification in overlapped capillary electrophoresis peaks, Talanta, vol. 65, pp.118-128, (2005).

DOI: 10.1016/j.talanta.2004.05.050

[16] L. Jiao, QSPR studies on soot-water partition coefficients of persistent organic pollutants by using artificial neural network, Chemosphere, vol. 80, pp.671-675, (2010).

DOI: 10.1016/j.chemosphere.2010.04.013

[17] Y.J. Ye, D.X. Ding, X.Y. Li, X.H. Zhou, and Y. Tan, Application of BP-ANN in forecasting the equivalent radon exhalation rate of uranium ore-rock in the course of mine ventilation, Proceedings of Computational Intelligence and Software Engineering, Wuhan, China, 2009, pp.1-4.

DOI: 10.1109/cise.2009.5366769