Forecasting wildfire disease on tobacco: toward developing a high-accuracy prediction model for disease index using local climate factors and support vector regression

  • X. H. Cai
  • T. Chen
  • R. Y. Wang
  • Y. J. Fan
  • Y. Li
  • S. N. Hu
  • Z. M. Yuan
  • H. G. Li
  • X. Y. Li
  • S. Y. Zhao
  • Q. M. ZhouEmail author
  • W. ZhouEmail author
Original Paper


Tobacco wildfire disease is common globally, and climate change may increase the risk of outbreaks. Therefore, there is an urgent need to establish an effective climate model to forecast the occurrence of wildfire disease. To design such a model, we collected data for 40 wildfire disease indices via tobacco field surveys and data for 15 climate factors of Guiyang County in China from 2012 to 2016. First, we built multiple linear regression (MLR), stepwise linear regression (SLR) and support vector regression (SVR) models using three climate features (precipitation, mean daily temperature and sunshine duration), and we could not find an effective model. Second, we built three corresponding models using expanded 15 climate features and an in-house WDEM method (the worst descriptor elimination multi-roundly), and the independent test results showed that the best SVR model had not only a higher predictive accuracy (\( {Q}_{ext}^2 \) = 0.94) but also a better stability. Finally, we further evaluated the biological significance of their retained climate features and the single-factor effects of the best model according to the interpretability analysis, and our results indicated that (1) the three climate factors (minimum value of wind velocity, daily range of temperature and daily pressure) strongly affected the occurrence of wildfire disease; (2) the ranges of relative humidity and sunshine hours were negatively correlated with the occurrence of wildfire disease, while daily mean vapour pressure was positively correlated with the occurrence of the disease. Our work enables a useful theoretical prediction for wildfire disease, especially in terms of climate-related predictions.



The authors thank other members of the laboratory at Department of Bioinformatics, College of Plant Protection, Hunan Agricultural University, Changsha, China for their help during the manuscript preparation.

Funding information

This research was supported by China Postdoctoral Science Foundation (No.2015T80870 and No.2014M562109), China Scholarship Council (No.201708430002) and Scientific Research Fund of Hunan Provincial Education Department (NO.17C0770).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.


  1. Anderson PJ, Chapman GH (1923) Tobacco wildfire in 1922. Mass Agricult Exper Stat Bull 213:1–27Google Scholar
  2. Andrew AM (2001) An introduction to support vector machines and other kernel-based learning methods [J]. Kybernetes 32(1):1–28Google Scholar
  3. Anne J (2010) Wildfire, angular leaf spot. Bacterial diseases, Chapter 15.Google Scholar
  4. Atkinson NJ, Urwin PE (2012) The interaction of plant biotic and abiotic stresses: from genes to the field [J]. J Exp Bot 63(10):3523CrossRefGoogle Scholar
  5. Barón M, Flexas J, DeLucia E H, 2012. Photosynthesis responses to biotic stress. En: Terrestrial photosynthesis in a changing environment. A molecular, physiological and ecological approach. Ed. J. Flexas, F. Loreto and H. Medrano. Cambridge: Cambridge Press, 331–350.Google Scholar
  6. Bostock RM, Pye MF, Roubtsova TV (2014) Predisposition in Plant Disease: Exploiting the Nexus in Abiotic and Biotic Stress Perception and Response [J]. Ann Rev Phytopathol 52(52):517CrossRefGoogle Scholar
  7. Chen Y, Yuan ZM, Zhou W et al (2009) A Novel QSAR Model Based on Geostatistics and Support Vector Regression [J]. ActaPhysico-ChimicaSinica 25(8):1587–1592(6)Google Scholar
  8. Diachun S, Valleau WD, Johnson EM (1942) Relation of moisture to invasion of tobacco leaves by Bacterium tabacum and Bacterium angulatum [J]. Phytopathology, (32), 379–387.Google Scholar
  9. Eriksson L, Johansson E, Müller M et al (1997) Cluster-based Design in Environmental QSAR [J]. Mol Inform 16(5):383–390Google Scholar
  10. Griebel T, Zeier J (2008) Light Regulation and Daytime Dependency of Inducible Plant Defenses in Arabidopsis: Phytochrome Signaling Controls Systemic Acquired Resistance Rather Than Local Defense [J]. Plant Physiol 147(2):790–801CrossRefGoogle Scholar
  11. Ichinose Y, Taguchi F, Mukaihara T (2013) Pathogenicity and virulence factors of Pseudomonas syringae [J]. J Gen Plant Pathol 79(5):285–296CrossRefGoogle Scholar
  12. Kangasjärvi S, Neukermans J, Li S et al (2012) Photosynthesis, photorespiration, and light signalling in defence responses [J]. J Exp Bot 63(4):1619CrossRefGoogle Scholar
  13. Lei T, Sun H, Kang Y et al (2017a) ADMET Evaluation in Drug Discovery. 18. Reliable Prediction of Chemical-Induced Urinary Tract Toxicity by Boosting Machine Learning Approaches [J]. Mol Pharm 14(11):3935–3953CrossRefGoogle Scholar
  14. Lei T, Chen F, Liu H, et al., 2017b. ADMET Evaluation in Drug Discovery. 17. Development of Quantitative and Qualitative Prediction Models for Chemical-Induced Respiratory Toxicity [J]. Mol Pharm, 14(7): 2407–2421.Google Scholar
  15. Ludovic C, Juan MS, Eric Q et al (2016) vSDC: a method to improve early recognition in virtual screening when limited experimental resources are available [J]. J Cheminformatics 8:1CrossRefGoogle Scholar
  16. Tang QY and Feng MG (2007) DPS data processing system-experimental design, statistical analysis and data mining, science press, pp.625–644.Google Scholar
  17. Shu M, Jiang Y, Yang L, et al., 2009. Application of ‘HESH’ Descriptors for the Structure-Activity Relationships of Antimicrobial Pep tides [J]. Protein Pept Lett, 16(2), 143–149.Google Scholar
  18. Su M, Wang L, Dai Z et al (2012) Primary Structural Characterizations of Polypeptide and Antimicrobial Peptides QSAM Modeling [J]. Chem J Chin Universities 33(11):2526–2531Google Scholar
  19. Suzuki N, Rivero R M, Shulaev V, et al., 2014. Abiotic and biotic stress combinations [J]. New Phytologist, 203(1), 32–43.Google Scholar
  20. Tan XS, Yuan ZM, Zhou TJ et al (2008) Multi-KNN-SVR combinatorial forecast and its application to QSAR of fluorine-containing compounds [J]. Chem J Chin Univ 29(1):95–99Google Scholar
  21. Vapnik V (1995) The nature of statistical learning theory [C]// Springer-Verlag New York, Inc.988–999.Google Scholar
  22. Wang LF, Tan XS, Bai LY et al (2012) Establishing an Interpretability System for Support Vector Regression and Its Application in QSAR of Organophosphorus Insecticide [J]. Asian J Chem 24(4):1575–1578Google Scholar
  23. Wang S, Sun H, Liu H et al (2016) ADMET Evaluation in Drug Discovery. 16. Predicting hERG Blockers by Combining Multiple Pharmacophores and Machine Learning Approaches [J]. Mol Pharm 13(8):2855–2866CrossRefGoogle Scholar
  24. Wei Z, Dai Z, Yuan C et al (2012) High-Dimensional Descriptor Selection and Computational QSAR Modeling for Antitumor Activity of ARC-111 Analogues Based on Support Vector Regression (SVR)[J]. Int J Mol Sci 13(1):1161–1172CrossRefGoogle Scholar
  25. Wang ZG, Ding W (2012) Studies on Key-Factor Analysis of Tobacco Wildfire Disease’s Occurrence and Its Control Techniques. Southwest UniversityGoogle Scholar
  26. Zhang S, Wei L, Bastow K et al (2007) Antitumor agents 252. Application of validated QSAR models to database mining: discovery of novel tylophorine derivatives as potential anticancer agents [J]. J Comput-Aided Mol Des 21(1–3):97–112CrossRefGoogle Scholar
  27. Zhou W, Dai Z J, Chen Y, et al., 2013. Computational QSAR models with high-dimensional descriptor selection improve antitumor activity design of ARC-111 analogues[J]. Med Chem Res, 22(1), 278–286.Google Scholar
  28. Zhou W, Wu SB, Dai ZJ et al (2015) Nonlinear QSAR moels with high-dimensional descriptor selection and SVR improve toxicity prediction and evaluation of phenols on photobacterium phosphoreum[J]. Chemometr Intell Lab Syst 145:30–38CrossRefGoogle Scholar
  29. Zhang XG (2000) Introduction to statistical learning theory and support vector machines[J]. Acta Automatica Sinica 26(01):32–42Google Scholar

Copyright information

© Springer-Verlag GmbH Austria, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Hunan Provincial Engineering and Technology Research Center for Agricultural Big Data Analysis and Decision-MakingHunan Agricultural UniversityChangshaChina
  2. 2.Hunan Provincial Key Laboratory for Biology and Control of Plant Diseases and Insect PestsHunan Agricultural UniversityChangshaChina
  3. 3.Hunan Provincial Engineering and Technology Research Center for Biopesticide and Formulation ProcessingHunan Agricultural UniversityChangshaChina
  4. 4.College of AgricultureHunan Agricultural UniversityChangshaChina
  5. 5.Chenzhou Company of Hunan Tobacco CompanyChenzhouChina
  6. 6.Hunan Tobacco CompanyChangshaChina
  7. 7.Department of Soil and Crop SciencesTexas A&M UniversityCollege stationUSA

Personalised recommendations