Regulation of transcriptome networks that mediate ginsenoside biosynthesis by essential ecological factors

Ginseng, a valuable Chinese medicinal herb, is renowned worldwide for its effectiveness in alleviating certain conditions and promoting overall health. In this study, we performed weighted gene co-expression network analysis (WGCNA) on the accumulation of essential saponins under the influence of 13 essential environmental factors (including air temperature, air bottom temperature, surface mean temperature, soil temperature, surface shortwave radiation, soil moisture, soil water content, rainfall, total precipitation, elevation, soil type, soil pH, and soil water potential). We identified a total of 40 transcript modules associated with typical environmental factors and the accumulation of essential saponins. Among these, 18 modules were closely related to the influence of typical environmental factors, whereas 22 modules were closely related to the accumulation of essential saponins. These results were verified by examining the transcriptome, saponin contents, environmental factor information and the published data and revealed the regulatory basis of saponin accumulation at the transcriptome level under the influence of essential environmental factors. We proposed a working model of saponin accumulation mediated by the transcriptional regulatory network that is affected by typical environmental factors. An isomorphic white-box neural network was constructed based on this model and the predicted results of the white-box neural network correlated with saponin accumulation. The effectiveness of our correlation-directed graph in predicting saponin contents was verified by bioinformatics analysis based on results obtained in this study and transcripts known to affect the biosynthesis of saponin Rb1. The directed graph represents a useful tool for manipulating saponin biosynthesis while considering the influence of essential environmental factors in ginseng and other medicinal plants.