Prediction of active ingredients in Salvia miltiorrhiza Bunge. based on soil elements and artificial neural network

PeerJ

article

Yu Liu¹  Ke Wang¹  Zhu-Yun Yan¹  Xiaofeng Shen³  Xinjie Yang⁴ 

[1] State Key Laboratory of Characteristic Chinese Medicine Resources in Southwest China, School of Pharmacy, Chengdu University of Traditional Chinese Medicine;School of Big Data and Artificial Intelligence, Chengdu Technological University;Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences;School of Pharmacy, Shaanxi University of Chinese Medicine

关键词: Salvia miltiorrhiza Bunge.;    Active ingredients;    Artificial neural network;    Soil elements;   

DOI  :  10.7717/peerj.12726

学科分类：社会科学、人文和艺术（综合）

来源: Inra

【 摘 要 】

The roots of Salvia miltiorrhiza Bunge. are commonly used in the treatment of cardiovascular diseases, and tanshinones and salvianolic acids are its main active ingredients. However, the composition and content of active ingredients of S. miltiorrhiza planted in different regions of the soil environment are also quite different, which adds new difficulties to the large-scale and standardization of artificial cultivation. Therefore, in this study, we measured the active ingredients in the roots of S. miltiorrhiza and the contents of rhizosphere soil elements from 25 production areas in eight provinces in China, and used the data to develop a prediction model based on BP (back propagation) neural network. The results showed that the active ingredients had different degrees of correlation with soil macronutrients and trace elements, the prediction model had the best performance (MSE = 0.0203, 0.0164; R2 = 0.93, 0.94). The artificial neural network model was shown to be a method that can be used to screen the suitable cultivation sites and proper fertilization. It can also be used to optimize the fertilizer application at specific sites. It also suggested that soil testing formula fertilization should be carried out for medicinal plants like S. miltiorrhiza, which is grown in multiple origins, rather than promoting the use of “special fertilizer” on a large scale. Therefore, the model is helpful for efficient, rational, and scientific guidance of fertilization management in the cultivation of S. miltiorrhiza.

【 授权许可】

CC BY