Scientia Agricultura Sinica ›› 2022, Vol. 55 ›› Issue (17): 3426-3436.doi: 10.3864/j.issn.0578-1752.2022.17.013

• FOOD SCIENCE AND ENGINEERING • Previous Articles     Next Articles

Research on the Application of a Balanced Sampling-Random Forest Early Warning Model for Aflatoxin Risk in Peanut

GUO Can1,4(),YUE XiaoFeng1,3,4(),BAI YiZhen1,3,4,5,ZHANG LiangXiao1,2,3,4,ZHANG Qi1,2,3,4,5(),LI PeiWu1,2,3,4,5()   

  1. 1Oil Crops Research Institute, Chinese Academy of Agricultural Sciences, Wuhan 430062
    2Key Laboratory of Biology and Genetic Improvement of Oil Crops, Ministry of Agriculture and Rural Affairs, Wuhan 430062
    3National Agricultural Testing Benchmark Laboratory (Biotoxin), Wuhan 430062
    4Laboratory of Risk Assessment for Oilseeds Products (Wuhan), Ministry of Agriculture and Rural Affairs, Wuhan 430062
    5Quality Inspection and Test Center for Oilseeds Products, Ministry of Agriculture and Rural Affairs, Wuhan 430062
  • Received:2022-01-14 Accepted:2022-05-17 Online:2022-09-01 Published:2022-09-07
  • Contact: Qi ZHANG,PeiWu LI E-mail:guocan2020@163.com;yuexf2017@caas.cn;zhangqi01@caas.cn;peiwuli@oilcrops.cn

RichHTML

29

PDF

88

Abstract:

【Objective】 Peanuts are highly vulnerable to aflatoxin contamination. Based on the Balanced Sampling-Random Forest early warning model for aflatoxin contamination in peanut established previously, this study aimed to analyze the main technical parameters and practical application effects of the model through systematic application research, which could provide a critical technical support for risk prediction of aflatoxin in post-harvest peanuts in China.【Method】 The model was used to predict the aflatoxin contamination risk of 153 main peanut producing cities in China from 2019 to 2020 by selecting the data of one month before the peanut harvest, including one geographical variable (latitude) and three climatic variables (precipitation, average air pressure, and daily average temperature of 8:00-20:00) as the key input parameters of the model. The immunoaffinity chromatography-high performance liquid chromatography-fluorescence detection method was used to determine the aflatoxin content of 2 164 peanuts to obtain the aflatoxin contamination data areas. The accuracy, precision, sensitivity, and false-positive rate of the model were analyzed to clarify the application effect according to the predicted risk and the actual risk of the model. 【Result】 A total of 125 areas were predicted as low-risk areas of aflatoxin, of which 116 areas were consistent with the actual measurement results, but 9 high-risk producing areas were misjudged as low-risk areas (False negative). Meanwhile, 28 areas were predicted as high-risk areas of aflatoxin, of which 15 areas were consistent with the actual measurement results, but 13 low-risk producing areas were misjudged as high-risk producing areas (False positive). Therefore, the accuracy of the model was 85.61%, the false-negative rate was 8.49%, and the false-positive rate was 5.88%. 【Conclusion】 The application of the Balanced Sampling-Random Forest early warning model could predict the risk of aflatoxin contamination in peanuts, which provided the technical support for scientifically guiding the harvesting, storage and utilization in post-harvest peanuts in China, thereby reducing the loss of aflatoxin contamination and guaranteeing the quality and safety of agricultural products.

Key words: peanut, aflatoxin, balanced sampling-random forest, early warning model, classification matrix

Table 1

Classification matrix of classification effect"

分类矩阵
Classification matrix		 预测为阳性
Predicted as positive		 预测为阴性
Predicted as negative
观察值为阳性
Originally classified as positive		 T-P F-N
观察值为阴性
Originally classified as negative		 F-P T-N

Fig. 1

Balanced Sampling-Random Forest warning model roadmap for aflatoxin contamination"

Fig. 2

Average content and rates of aflatoxin contamination in high-risk areas in 2019"

Fig. 3

Average content and rates of aflatoxin contamination in high-risk areas in 2020"

Fig. 4

Distribution of climate parameters in high-risk areas of aflatoxin from 2019 to 2020"

Fig. 5

Performance graph of the training set, validation set, and test set"

Table 2

Data set 1 (2019) classification matrix between model prediction and observation data"

分类矩阵
Classification matrix		 预测值为高风险地区
Predicted as
high-risk areas		 预测值为低风险地区
Predicted as
low-risk areas
观察值为高风险地区
Originally classified as high-risk areas		 8 (9.52%) 7 (8.33%)
观察值为低风险地区
Originally classified as low-risk areas		 6 (7.14%) 63 (75.00%)

Table 3

Data set 2 (2020) classification matrix between model prediction and observation data"

分类矩阵
Classification matrix		 预测值为高风险地区
Predicted as
high-risk areas		 预测值为低风险地区
Predicted as
low-risk areas
观察值为高风险地区
Originally classified as high-risk areas		 7 (10.14%) 2 (2.89%)
观察值为低风险地区
Originally classified as low-risk areas		 7 (10.14%) 53 (76.81%)

Fig. 6

The overall recognition results of the application"

[1] 李培武, 丁小霞, 白艺珍, 周海燕, 印南日. 农产品黄曲霉毒素风险评估研究进展. 中国农业科学, 2013, 46(12): 2534-2542.
LI P W, DING X X, BAI Y Z, ZHOU H Y, YIN N R. Advance in research on risk assessment of aflatoxin in agricultural products. Scientia Agricultura Sinica, 2013, 46(12): 2534-2542. (in Chinese)
[2] HELL K, FANDOHAN P, BANDYOPADHYAY R, KIEWNICK S, SIKORA R, COTTY P J, LESLIE J F, BANDYOPADHYAY R. Pre- and postharvest management of aflatoxin in maize: An african perspective. Mycotoxins: Detection Methods, Management, Public Health and Agricultural Trade, 2008: 219-229.
[3] HILL R A, BLANKENSHIP P D, COLE R J, SANDERS T H. Effects of soil moisture and temperature on preharvest invasion of peanuts by the Aspergillus flavus group and subsequent aflatoxin development. Schizophrenia Research, 1983, 45(2): 628-633. doi: 10.1128/aem.45.2.628-633.1983.
doi: 10.1128/aem.45.2.628-633.1983
[4] DORNER J W, COLE R J, SANDERS T H, BLANKENSHIP P D. Interrelationship of kernel water activity, soil temperature, maturity, and phytoalexin production in preharvest aflatoxin contamination of drought-stressed peanuts. Mycopathologia, 1989, 105(2): 117-128. doi: 10.1007/BF00444034.
doi: 10.1007/BF00444034
[5] CHAUHAN Y S, WRIGHT G C, RACHAPUTI R C N, HOLZWORTH D, BROOME A, KROSCH S, ROBERTSON M J. Application of a model to assess aflatoxin risk in peanuts. The Journal of Agricultural Science, 2010, 148(3): 341-351.
doi: 10.1017/S002185961000002X
[6] CHAUHAN Y, TATNELL J, KROSCH S, KARANJA J, GNONLONFIN B. An improved simulation model to predict pre-harvest aflatoxin risk in maize. Field Crops Research, 2015, 178: 91-99.
doi: 10.1016/j.fcr.2015.03.024
[7] THAI C N, BLANKENSHIP P D, COLE R J, SANDERS T H, DORNER J W. Relationship between aflatoxin production and soil temperature for peanuts under drought stress. Transactions of the ASAE, 1990, 33(1): 324-329.
doi: 10.13031/2013.31333
[8] CHAUHAN Y S, WRIGHT G C, RACHAPUTI N C. Modeling climatic risks of aflatoxin contamination in maize. Australian Journal of Experimental Agriculture, 2008, 48(3): 358-366.
doi: 10.1071/EA06101
[9] HELL K, MUTEGI C. Aflatoxin control and prevention strategies in key crops of Sub-Saharan Africa. African Journal of Microbiology Research, 2010, 5(5): 459-466.
[10] PAYNE G A, CASSEL D K, ADKINS C R. Reduction of aflatoxin contamination in corn by irrigation and tillage. Phytopathology, 1986, 76(7): 697-684.
[11] MCCOWN R L, HAMMER G L, HARGREAVES J N G, HOLZWORTH D P, FREEBAIRN D M. APSIM: A novel software system for model development, model testing and simulation in agricultural systems research. Agricultural Systems, 1996, 50(3): 255-271.
doi: 10.1016/0308-521X(94)00055-V
[12] PARMAR R, MCCLENDON R W, BLANKENSHIP P D, COLE R J, DORNER J W. Estimation of aflatoxin contamination in preharvest peanuts using neural networks. Transactions of the ASAE, 1997, 40: 809-813.
[13] BREIMAN L. Bagging predictors. Machine Learning, 1996, 24(2): 123-140.
[14] ADELE C. Neural networks. Technometrics, 2012, 42(4): 432.
[15] 武琳霞. 中国花生黄曲霉毒素污染风险预警模型研究[D]. 北京: 中国农业科学院, 2019.
WU L X. Researches on risk prediction model of aflatoxin contamination in China[D]. Beijing: Chinese Academy of Agricultural Sciences, 2019. (in Chinese)
[16] 马良, 李培武, 张文. 高效液相色谱法对农产品中黄曲霉毒素的测定研究. 分析测试学报, 2007, 26(6): 774-778.
MA L, LI P W, ZHANG W. Determination of aflatoxins in agricultural products by high performance liquid chromatography. Journal of Instrumental Analysis, 2007, 26(6): 774-778. (in Chinese)
[17] 国家气象信息中心. [2021-07-28] http://data.cma.cn.
National Meteorological Information Center. [2021-07-28] http://data.cma.cn. (in Chinese)
[18] WEBMASTER Climate Prediction Center. [2021-07-28] https://www.cpc.ncep.noaa.gov.
[19] 国家卫生和计划生育委员会, 国家食品药品监督管理总局. 食品安全国家标准食品中真菌毒素限量: GB 2761—2017[S]. 北京: 中国标准出版社, 2017.
National Health and Family Planning Commission of the People’s Republic of China and State Food and Drug Administration. National standards for food safety limit of mycotoxins in foods: GB 2761-2017[S]. Beijing: Standards Press of China, 2017. (in Chinese)
[20] DENG X Y, LIU Q, DENG Y, MAHADEVAN S. An improved method to construct basic probability assignment based on the confusion matrix for classification problem. Information Sciences, 2016, 340/341: 250-261.
doi: 10.1016/j.ins.2016.01.033
[21] 张安定. 遥感原理与应用题解. 北京: 科学出版社, 2016.
ZHANG A D. Remote Sensing Principle and Application Problem Solution. Beijing: Science Press, 2016. (in Chinese)
[22] 王云飞, 庞勇, 舒清态. 基于随机森林算法的橡胶林地上生物量遥感反演研究: 以景洪市为例. 西南林业大学学报, 2013, 33(6): 38-45, 111.
WANG Y F, PANG Y, SHU Q T. Counter-estimation on aboveground biomass of Hevea brasiliensis plantation by remote sensing with random forest algorithm-A case study of Jinghong. Journal of Southwest Forestry University, 2013, 33(6): 38-45, 111. (in Chinese)
[23] 李贞子, 张涛, 武晓岩, 李康. 随机森林回归分析及在代谢调控关系研究中的应用. 中国卫生统计, 2012, 29(2): 158-160, 163.
LI Z Z, ZHANG T, WU X Y, LI K. Methodology of regression by random forest and its application on metabolomics. Chinese Journal of Health Statistics, 2012, 29(2): 158-160, 163. (in Chinese)
[24] 曹振, 崔路瑶, 雷斌, 王婧旖, 曹双胜. 城轨列车滚动轴承智能诊断的特征降维与随机森林方法. 吉林大学学报(工学版), 2021: 1-7. doi: 10.13229/j.cnki.jdxbgxb20210312.
doi: 10.13229/j.cnki.jdxbgxb20210312
CAO Z, CUI L Y, LEI B, WANG J Y, CAO S S. Feature dimensionality reduction and random forest method in intelligent diagnosis of rolling bearings for urban rail trains. Journal of Jilin University (Engineering ang Technology Edition), 2021: 1-7. doi: 10.13229/j.cnki.jdxbgxb20210312. (in Chinese)
doi: 10.13229/j.cnki.jdxbgxb20210312
[25] 傅质馨, 孙宁新, 朱俊澎, 袁越. 基于输出功率预测的风电机组运行风险度评估. 电力信息与通信技术, 2021, 19(5): 14-22. doi: 10.16543/j.2095-641x.electric.power.ict.2021.05.003.
doi: 10.16543/j.2095-641x.electric.power.ict.2021.05.003
FU Z X, SUN N X, ZHU J P, YUAN Y. Risk assessment of wind turbine operation based on wind power output prediction. Electric Power Information and Communication Technology, 2021, 19(5): 14-22. doi: 10.16543/j.2095-641x.electric.power.ict.2021.05.003. (in Chinese)
doi: 10.16543/j.2095-641x.electric.power.ict.2021.05.003
[26] 方匡南, 吴见彬, 谢邦昌. 基于随机森林的保险客户利润贡献度研究. 数理统计与管理, 2014, 33(6): 1122-1131. doi: 10.13860/j.cnki.sltj-20141122-063.
doi: 10.13860/j.cnki.sltj-20141122-063
FANG K N, WU J B, XIE B C. Measurement of customer profitability of insurance company in China based on random forest. Journal of Applied Statistics and Management, 2014, 33(6): 1122-1131. doi: 10.13860/j.cnki.sltj-20141122-063. (in Chinese)
doi: 10.13860/j.cnki.sltj-20141122-063
[27] 方匡南, 吴见彬, 朱建平, 谢邦昌. 随机森林方法研究综述. 统计与信息论坛, 2011, 26(3): 32-38.
FANG K N, WU J B, ZHU J P, XIE B C. A review of technologies on random forests. Statistics ＆ Information Forum, 2011, 26(3): 32-38. (in Chinese)
[28] WU L X, DING X X, LI P W, DU X H, ZHOU H Y, BAI Y Z, ZHANG L X. Aflatoxin contamination of peanuts at harvest in China from 2010 to 2013 and its relationship with climatic conditions. Food Control, 2016, 60: 117-123.
doi: 10.1016/j.foodcont.2015.06.029
[29] 丁小霞. 中国产后花生黄曲霉毒素污染与风险评估方法研究[D]. 北京: 中国农业科学院, 2011.
DING X X. Study on post-harvest peanut aflatoxins contamination and risk assessment in China[D]. Beijing: Chinese Academy of Agricultural Sciences, 2011. (in Chinese)
[30] BATTILANI P, TOSCANO P, VAN DER FELS-KLERX H J, MORETTI A, CAMARDO LEGGIERI M, BRERA C, RORTAIS A, GOUMPERIS T, ROBINSON T. Aflatoxin B1 contamination in maize in Europe increases due to climate change. Scientific Reports, 2016, 6: 24328. doi: 10.1038/srep24328.
doi: 10.1038/srep24328
[31] DING X, WU L, LI P, ZHANG Z, ZHOU H, BAI Y, CHEN X, JIANG J. Risk assessment on dietary exposure to aflatoxin B1 in post-harvest peanuts in the Yangtze River ecological region. Toxins, 2015, 7(10): 4157-4174. doi: 10.3390/toxins7104157.
doi: 10.3390/toxins7104157
[32] COLE R J, SANDERS T H, HILL R A, BLANKENSHIP P D. Mean geocarposphere temperatures that induce preharvest aflatoxin contamination of peanuts under drought stress. Mycopathologia, 1985, 91(1): 41-46. doi: 10.1007/BF00437286.
doi: 10.1007/BF00437286
[33] 王海鸥, 陈守江, 胡志超, 谢焕雄. 花生黄曲霉毒素污染与控制. 江苏农业科学, 2015, 43(1): 270-273. doi: 10.15889/j.issn.1002-1302.2015.01.091.
doi: 10.15889/j.issn.1002-1302.2015.01.091
WANG H O, CHEN S J, HU Z C, XIE H X. Peanut aflatoxin contamination and control. Jiangsu Agricultural Sciences, 2015, 43(1): 270-273. doi: 10.15889/j.issn.1002-1302.2015.01.091. (in Chinese)
doi: 10.15889/j.issn.1002-1302.2015.01.091
[34] KLICH M A. Aspergillus flavus: The major producer of aflatoxin. Molecular Plant Pathology, 2007, 8(6): 713-722. doi: 10.1111/j.1364-3703.2007.00436.x.
doi: 10.1111/j.1364-3703.2007.00436.x
[35] 戴显红, 丁小霞, 李培武, 陈琳, 姜俊, 贾明明. 土壤交换性钙含量对花生黄曲霉毒素污染影响的研究. 农产品质量与安全, 2017(4): 11-17.
DAI X H, DING X X, LI P W, CHEN L, JIANG J, JIA M M. Effect of exchangeable calcium concentration in soil on aflatoxin pollution in peanuts. Quality and Safety of Agro-Products, 2017(4): 11-17. (in Chinese)
[36] 唐秀梅, 罗赛云, 钟瑞春, 唐荣华. 南方花生黄曲霉毒素污染防控栽培技术. 现代农业科技, 2016(22): 35-36.
TANG X M, LUO S Y, ZHONG R C, TANG R H. Cultivation techniques for the prevention and reduction of aflatoxin contamination of peanut in South China. Modern Agricultural Science and Technology, 2016(22): 35-36. (in Chinese)
[1] WU Yue,SUI XinHua,DAI LiangXiang,ZHENG YongMei,ZHANG ZhiMeng,TIAN YunYun,YU TianYi,SUN XueWu,SUN QiQi,MA DengChao,WU ZhengFeng. Research Advances of Bradyrhizobia and Its Symbiotic Mechanisms with Peanut [J]. Scientia Agricultura Sinica, 2022, 55(8): 1518-1528.
[2] BIAN NengFei, SUN DongLei, GONG JiaLi, WANG Xing, XING XingHua, JIN XiaHong, WANG XiaoJun. Evaluation of Edible Quality of Roasted Peanuts and Indexes Screening [J]. Scientia Agricultura Sinica, 2022, 55(4): 641-652.
[3] WANG Juan,CHEN HaoNing,SHI DaChuan,YU TianYi,YAN CaiXia,SUN QuanXi,YUAN CuiLing,ZHAO XiaoBo,MOU YiFei,WANG Qi,LI ChunJuan,SHAN ShiHua. Functional Analysis of AhNRT2.7a in Response to Low-Nitrogen in Peanut [J]. Scientia Agricultura Sinica, 2022, 55(22): 4356-4372.
[4] SHI XiaoLong,GUO Pei,REN JingYao,ZHANG He,DONG QiQi,ZHAO XinHua,ZHOU YuFei,ZHANG Zheng,WAN ShuBo,YU HaiQiu. A Salt Stress Tolerance Effect Study in Peanut Based on Peanut//Sorghum Intercropping System [J]. Scientia Agricultura Sinica, 2022, 55(15): 2927-2937.
[5] HAO Jing,LI XiuKun,CUI ShunLi,DENG HongTao,HOU MingYu,LIU YingRu,YANG XinLei,MU GuoJun,LIU LiFeng. QTL Mapping for Traits Related to Seed Number Per Pod in Peanut (Arachis hypogaea L.) [J]. Scientia Agricultura Sinica, 2022, 55(13): 2500-2508.
[6] FENG Chen,HUANG Bo,FENG LiangShan,ZHENG JiaMing,BAI Wei,DU GuiJuan,XIANG WuYan,CAI Qian,ZHANG Zhe,SUN ZhanXiang. Effects of Different Configurations on Nitrogen Uptake and Utilization Characteristics of Maize-Peanut Intercropping System in West Liaoning [J]. Scientia Agricultura Sinica, 2022, 55(1): 61-73.
[7] MENG XinHao,DENG HongTao,LI Li,CUI ShunLi,Charles Y. CHEN,HOU MingYu,YANG XinLei,LIU LiFeng. QTL Mapping for Lateral Branch Angle Related Traits of Cultivated Peanut (Arachis hypogaea L.) [J]. Scientia Agricultura Sinica, 2021, 54(8): 1599-1612.
[8] JING Dan, YUE XiaoFeng, BAI YiZhen, GUO Can, DING XiaoXia, LI PeiWu, ZHANG Qi. The Infectivity of Aspergillus flavus in Peanut [J]. Scientia Agricultura Sinica, 2021, 54(23): 5008-5020.
[9] GU BoWen,YANG JinFeng,LU XiaoLing,WU YiHui,LI Na,LIU Ning,AN Ning,HAN XiaoRi. Effects of Continuous Application of Biochar on Chlorophyll Fluorescence Characteristics of Peanut at Different Growth Stages [J]. Scientia Agricultura Sinica, 2021, 54(21): 4552-4561.
[10] XUE HuaLong,LOU MengYu,LI Xue,WANG Fei,GUO BinBin,GUO DaYong,LI HaiGang,JIAO NianYuan. Effects of Phosphorus Application Levels on Growth and Yield of Winter Wheat Under Different Crops for Rotation [J]. Scientia Agricultura Sinica, 2021, 54(17): 3712-3725.
[11] ZHANG MaoNing,HUANG BingYan,MIAO LiJuan,XU Jing,SHI Lei,ZHANG ZhongXin,SUN ZiQi,LIU Hua,QI FeiYan,DONG WenZhao,ZHENG Zheng,ZHANG XinYou. Genetic Analysis of Peanut Kernel Traits in a Nested-crossing Population by Major Gene Plus Polygenes Mixed Model [J]. Scientia Agricultura Sinica, 2021, 54(13): 2916-2930.
[12] TingHui HU,LiangQiang CHENG,Jun WANG,JianWei LÜ,QingLin RAO. Evaluation of Shade Tolerance of Peanut with Different Genotypes and Screening of Identification Indexes [J]. Scientia Agricultura Sinica, 2020, 53(6): 1140-1153.
[13] ZhiJun XU,Sheng ZHAO,Lei XU,XiaoWen HU,DongSheng AN,Yang LIU. Discovery of Microsatellite Markers from RNA-seq Data in Cultivated Peanut (Arachis hypogaea) [J]. Scientia Agricultura Sinica, 2020, 53(4): 695-706.
[14] HaiChao CAO,QingShun LIU,HaiXiu BAI,Jun HAN,ShiLing YANG,Ming XUE,Feng LIU. Development of 30% Clothianidin·Pyraclostrobin·Difenoconazole Flowable Concentrate for Seed-Coating and Its Application Effect in Peanut Field [J]. Scientia Agricultura Sinica, 2019, 52(20): 3595-3604.
[15] LIU Peng,TIAN YingZhe,ZHONG YongJia,LIAO Hong. Isolation and Application of Effective Rhizobium Strains in Peanut on Acidic Soils [J]. Scientia Agricultura Sinica, 2019, 52(19): 3393-3403.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
京ICP备09089781号-30
Copyright 2018 © Editorial office of Scientia Agricultura Sinica
Tel: 86-010-82106279    E-mail: zgnykx@mail.caas.net.cn
Supported by Beijing Magtech Co.Ltd