Spatial and temporal distribution characteristics of Paris polyphylla var. yunnanensis and the prediction of steroidal saponins content
Zhong Chen, Li Li, Yuanzhong Wang
Abstract
The dried rhizome of Paris polyphylla var. yunnanensis (PPVY) can be utilized in medicine, exhibiting the efficacies of clearing away heat and toxins and alleviating swelling and pain. However, as an important traditional medicinal plant, the ecological factors that promote distribution and its geographical distribution pattern are still unclear. To assess the influences of ecological determinants on suitable habitats, the species distribution ensemble platform (Biomod2) and maximum entropy (MaxEnt) models were established to project the spatial and temporal distribution patterns of PPVY under various climate scenarios. On this basis, machine learning was combined with ultra-performance liquid chromatography (UPLC) and Fourier transform infrared spectroscopy (FT-IR) to assess the nuances of PPVY chemical composition in different regions, and spectrochemical characterization of steroidal saponins was also modelled for the sustainable supply of PPVY. The experimental results showed that under the current climatic conditions, the main suitable habitats of PPVY were mainly concentrated in southwest China, especially in Yunnan, Sichuan, and Guizhou provinces. With the change in the climate, the suitable habitat area of PPVY will expand to higher elevations and latitudes in the future. The long short-term memory network (LSTM) model had better prediction performance for the five important steroidal saponins, and 1D-MSC-SG is the dominant preprocessing method for model building. In terms of practical application, the results of this study aim to provide guidance for the scientific cultivation and quality evaluation of PPVY. Agribusinesses can harness the predicted suitable habitats for optimal planting area selection to improve the yield and quality of PPVY. In the pharmaceutical industry, the developed spectrochemical characterization models can be used to rapidly assess the quality of PPVY raw materials, which plays an essential role in ensuring the efficacy and safety of PPVY. • First time to evaluate and predict this species’s distribution by Biomod2 model. • Exploring the main chemical components responsible for the differences between PPVYs. • Modelling spectrochemical signatures for non-destructive prediction of PPVY. • Temperature and elevation are the key factors.