DOI QR코드

DOI QR Code

Prediction and analysis of acute fish toxicity of pesticides to the rainbow trout using 2D-QSAR

2D-QSAR방법을 이용한 농약류의 무지개 송어 급성 어독성 분석 및 예측

  • Received : 2011.11.29
  • Accepted : 2011.11.30
  • Published : 2011.12.25

Abstract

The acute toxicity in the rainbow trout (Oncorhynchus mykiss) was analyzed and predicted using quantitative structure-activity relationships (QSAR). The aquatic toxicity, 96h $LC_{50}$ (median lethal concentration) of 275 organic pesticides, was obtained from EU-funded project DEMETRA. Prediction models were derived from 558 2D molecular descriptors, calculated in PreADMET. The linear (multiple linear regression) and nonlinear (support vector machine and artificial neural network) learning methods were optimized by taking into account the statistical parameters between the experimental and predicted p$LC_{50}$. After preprocessing, population based forward selection were used to select the best subsets of descriptors in the learning methods including 5-fold cross-validation procedure. The support vector machine model was used as the best model ($R^2_{CV}$=0.677, RMSECV=0.887, MSECV=0.674) and also correctly classified 87% for the training set according to EU regulation criteria. The MLR model could describe the structural characteristics of toxic chemicals and interaction with lipid membrane of fish. All the developed models were validated by 5 fold cross-validation and Y-scrambling test.

본 연구는 농약류에 대하여 구조-활성의 정량적 관계(QSAR)를 이용하여 무지개 송어(학명: Oncorhynchus mykiss)의 급성 독성을 예측-분석하는 과정을 수행하였다. 모델 구현을 위해 사용된 275종의 농약류에 대한 수중 독성(96h $LC_{50}$) 값은 DEMETRA프로젝트의 데이터를 사용하였다. 예측 모델에 사용된 2차원 분자 표현자는 PreADMET프로그램으로부터 계산을 하였고, 선형 (다중 선형 회귀 방법)모델과 비선형(서포트 벡터 머신, 인공 신경망) 학습 방법들은 실험값과 예측값의 적합도를 고려하여 최적화 되었다. 데이터 전처리 과정을 거친 뒤에, 5묶음 교차 검증과정을 포함한 모집단 기반 전진 선택법을 통해서 각 학습 방법의 최적의 표현자 집합을 결정하였다. 가장 좋은 결과는 SVM 방법 ($R^2_{CV}$=0.677, RMSECV=0.887, MSECV=0.674) 이었고, EU의 규제 기준에 따른 분류에서는 87%의 정확도를 나타내었다. MLR방법을 통해서는 무지개 송어의 급성 독성에 대하여 독성을 나타내는 농약류의 구조적 특징과 지질 층과의 상호작용을 설명할 수 있었다. 개발된 모든 모델들은 5묶음 교차 검증과 Y-scrambling test을 통해 검증되었다.

Keywords

References

  1. J. Ahlers, F. Stock and B. Werschkun, Environ. Sci. Pollut. Res., 15(7), 565-572 (2008). https://doi.org/10.1007/s11356-008-0043-y
  2. G. T. Miller, In 'Sustaining the Eart', 6th Ed., pp 211- 216, Thompson Learning, Pacific Grove, CA, 2004.
  3. OECD, 'OECD Guidelines for Testing Chemicals, METHOD 203, Fish, Acute Toxicity Test', Paris, 1984.
  4. 'Technical Guidance Document in support of Commission directive 93/67/EEC for new notified substances and Commission regulation (EC) No 1488/94 on risk assessment for existing substances', Brussels, 1996.
  5. J. S. Jaworska, M. Comber, C. Auer and C. J. Van Leeuwen, Environ. Health Perspect, 111(10), 1358-1360 (2003). https://doi.org/10.1289/ehp.5757
  6. 'Development of Environmental Modules for Evaluation of Toxicity of pesticide Residues in Agriculture (DEMETRA)', project supported by the European Commission, Contract No. QLK5-CT-2002-00691.
  7. A. Roncaglioni, E. Benfenati, E. Boriani and M. Clook, J. Environ. Sci. Health, Pt. B:Pestic., Food Contam., Agric. Wastes, 39(4), 641-652 (2004). https://doi.org/10.1081/PFC-200026890
  8. M. Casalegno, G. Sello and E. Benfenati, Chem. Res. Toxicol., 19(11), 1533-1539 (2006). https://doi.org/10.1021/tx0601814
  9. S. K. Lee, S. H. Park, I. H. Lee and K. T. No, PreADMET Ver.v2.0, BMDRC: Seoul, Korea, 2007.
  10. G. Schneider, W. Neidhart, T. Giller and G. Schmid, Angew. Chem. Int. Ed. Engl., 38(19), 2894-2896 (1999). https://doi.org/10.1002/(SICI)1521-3773(19991004)38:19<2894::AID-ANIE2894>3.0.CO;2-F
  11. N. R. Draper and H. Smith, In 'Applied Regression Analysis', 2nd Ed., pp 294-379, John Wiley & Sons Inc., New York, 1981.
  12. C. Cortes and V. Vapnik, Mach. Learn., 20(3), 273-297 (1995).
  13. B. Scholkopf, A. J. Smola, R. C. Williamson and P. L. Bartlett, Neural Comput., 12(5), 1207-1245 (2000). https://doi.org/10.1162/089976600300015565
  14. D. E. Rumelhart, G. E. Hinton and R. J. Williams, Nature, 323(6088), 533-536 (1986). https://doi.org/10.1038/323533a0
  15. Rapidminer Ver.5.0, Rapid Miner is unquestionable the world-leading open-source system for data mining., Rapid-I: Dortmund, Germany, 2010.
  16. B. L. Podlogar, I. Muegge and L. J. Brice, Curr. Opi. Drug Dis. Dev., 4(1), 102-109 (2001).
  17. A. K. Ghose, V. N. Viswanadhan and J. J. Wendoloski, J. Phys. Chem. A, 102(21), 3762-3772 (1998). https://doi.org/10.1021/jp980230o
  18. J. Huang and X. Fan, Mol. Pharm., 8(2), 600-608 (2011). https://doi.org/10.1021/mp100423u
  19. T. Jager and S. A. L. M. Kooijman, Ecotoxicology, 18(2), 187-196 (2009). https://doi.org/10.1007/s10646-008-0271-4
  20. P. Ertl, B. Rohde and P. Selzer, J. Med. Chem., 43(20), 3714-3717 (2000). https://doi.org/10.1021/jm000942e
  21. K. T. No, J. A. Grant, M. S. Jhon and H. A. Scheraga, J. Phys. Chem., 94(11), 4740-4746 (1990). https://doi.org/10.1021/j100374a067
  22. K. T. No, J. A. Grant and H. A. Scheraga, J. Phys. Chem., 94(11), 4732-4739 (1990). https://doi.org/10.1021/j100374a066

Cited by

  1. Prediction of Human Health and Ecotoxicity of Chemical Substances Using the OECD QSAR Application Toolbox vol.39, pp.2, 2013, https://doi.org/10.5668/JEHS.2013.39.2.130
  2. QSPR analysis for predicting heat of sublimation of organic compounds vol.28, pp.3, 2015, https://doi.org/10.5806/AST.2015.28.3.187
  3. Development of QSAR Model Based on the Key Molecular Descriptors Selection and Computational Toxicology for Prediction of Toxicity of PCBs vol.54, pp.5, 2016, https://doi.org/10.9713/kcer.2016.54.5.621