| OBJECTIVE:In order to study the differences in the quality of Poria cocos from different producing areas,explore the high suitable areas for Poria cocos planting,and evaluate the quality of Poria cocos,so as to provide theoretical basis for the selection of Poria cocos planting areas and the development and utilization planning of related industries.This study:(1)Predict the current and future potential distribution of Poria cocos,analyze the impact of climate change on the suitable distribution area of Poria cocos,determine the suitable planting area of Poria cocos,provide the basis for the development of Poria cocos cultivation activities,and meet market demand;(2)According to the index component content of the sample,the different quality areas of Poria cocos were divided to determine the high quality area of Poria cocos,which was convenient for the directional research and utilization of related products;(3)Based on hyperspectral imaging technology,the identification model of Poria cocos origin was established to realize the non-destructive,rapid and accurate identification of Poria cocos origin,so as to quickly judge the quality of Poria cocos by combining the results of quality zoning.METHODS:(1)The distribution point information of Poria cocos in recent years was obtained and screened through databases and literature materials.The climate,altitude and soil type data of different periods were downloaded from the WorldClim and the HWSD,and the unified accuracy was 2.5 m.Person correlation analysis was used to screen environmental factors,and the MaxEnt models of SS1-RCP2.6,SSP2RCP4.5 and SSP5-RCP8.5 representing low,medium and high CO2 emission concentrations in the current(1970-2000)and future 2041-2060 and 2061-2080 periods were established.The AUC value was used to evaluate the accuracy of the model.(2)A total of 40 batches of Poria cocos samples were collected from 18 counties in 8 provinces of Anhui,Fujian,Guangxi,Hubei,Henan,Hunan,Shaanxi and Yunnan.High performance liquid chromatography(HPLC)method was used to measure the content of dehydrotumulosic acid,tumulosic acid,pachymic acid C,pachymic acid and pachymic acid A.The difference of triterpene acid content in different provinces and counties was analyzed by one-way ANOVA,and the Person correlation analysis between triterpene acid and environmental factors was carried out.According to the results of difference analysis and correlation results,the index components were selected,and the regression model between triterpene acid content and environmental factors was established.The suitable areas of different quality of Poria cocos were predicted,and the quality zoning results of Poria cocos were obtained..(3)The samples of Poria cocos from 18 counties in 8 provinces were collected as the research object.Hyperspectral data were collected in the visible-near infrared band(V band,410-990 nm)and short-wave infrared band(S band,950-2 500 nm).The data were divided into training set and prediction set according to the ratio of 7:3.The original spectral data were divided into S-band,V-band and full-band.Based on the original data(RD)of different bands,S-G smoothing(SGS),multiple scattering correction(MSC),first derivative(FD),second derivative(SD),standard normal variation(SNV)and other pretreatments were carried out.Then,the data were classified according to the origin categories of provinces and counties with different accuracies.Combined with partial least squares discriminant analysis(PLS-DA)and linear support vector machine(LinearSVC)classification methods,the optimal parameters of the model were determined by ten-fold cross validation to establish the origin identification model.Finally,the confusion matrix is used to evaluate the best model,and the F1 score is used as an evaluation method.RESULTS:(1)The AUC value of MaxEnt model for predicting the distribution of Poria cocos was 0.928,indicating that the model had high accuracy and could be used to predict the potential distribution of Poria cocos in different periods.The most suitable distribution area needs to keep the rainfall dry or rainy in the driest season and keep wet in the wettest season.In winter,the temperature should be kept in a slightly cold or mild range.The cultivation area of Poria cocos should be located in the middle and high altitude(100-3 500 m).The most suitable producing areas of Poria are located in Yunnan,Dabie Mountains,Sanshengpo and central and southern Zhejiang.The most suitable planting areas for Poria cocos in different periods in the future can be located in Yunnan,Sanshengpo,Dabie Mountains and Zhejiang(except for SSP1-RCP2.6 scenario in 2061-2080).Comparing the suitable area of Poria cocos in the future period with the suitable area of Poria cocos in the current period,it is found that except for the decrease of the suitable area of Poria cocos under the SSP1-RCP2.6 scenario in 20612080,the total suitable area of Poria cocos in other future periods will increase.(2)The results of the difference analysis of the content of triterpene acid in Poria cocos from different producing areas showed that there was a significant difference between the content of Dehydrotumulosic acid and Polyporenic acid C in different provinces.There were significant differences in the contents of Dehydrotumulosic acid,Tumulosic acid,Polyporenic acid C and Pachymic acid in different counties.On the whole,the quality of Poria cocos in Yiling of Hubei,Danfeng of Shaanxi and Jingdong of Yunnan was the best.In addition,the quality of Poria cocos in Huoshan/Susong of Anhui,Macheng/Qichun/Yingshan of Hubei,Jingzhou of Hunan and Pingnan of Fujian was also better.For the five triterpene acids,the total content of the five triterpene acids in Guangxi,Shaanxi and Fujian was the highest.The triterpene acid content of Poria cocos in Anhui Susong,Fujian Pingnan,Hubei Yiling,Shaanxi Danfeng and Yunnan Jingdong was generally higher.The results of the quality regionalization of Poria cocos showed that the precipitation and temperature in the wettest season and the altitude of the origin all had an effect on the quality of Poria cocos.The contents of Dehydrotumulosic acid and Polyporenic acid C in southern Tibet,northern Yunnan,southwestern Gansu,Shaanxi and southern central Shanxi was higher.The content of Pachymic acid was higher in southern Tibet,northwestern Yunnan,western Sichuan and southwestern Gansu.On the whole,the quality of Poria cocos in the western part of southern Tibet,the central part of western Sichuan,the northern part of Yunnan,the southwestern part of Gansu,Shaanxi and the southern part of central Shanxi is better.(3)The results of different precision origin classification models of Poria cocos showed that the accuracy of the prediction set of the origin recognition model established by FD processing data with different origin accuracy combined with LinearSVC was the highest in the same band,and the highest accuracy of the classification model by province was FDA+LinearSVC combination,98.66%;the FDV+LinearSVC combination had the highest accuracy of 98.56%.The overall F1 scores of the two models are 98.59%and 98.62%,respectively,indicating that the model has good performance.CONCLUSIONS:According to the distribution point information and ecological factor data of Poria cocos,the maximum entropy model was used to predict the distribution of suitable habitats of Poria cocos,and the regional division of Poria cocos quality was carried out in combination with the content of index components,which had certain significance for the cultivation,processing and export of Poria cocos.Hyperspectral technology combined with machine learning algorithm can nondestructively,quickly and accurately identify the origin of Poria cocos samples,and has a good application prospect in the identification of the origin of edible fungi.Combined with the results of comprehensive quality zoning of Poria cocos,the model can be used to determine whether it is a high-quality origin.It provides experimental methods and theoretical basis for the directional research and production of Poria cocos food and drug products in specific producing areas,and provides feasibility for rapid,efficient and pollution-free quality analysis of Poria cocos medicinal materials in the market. |
PDF Full Text Request