Evaluation of the impact of ecological factors on the habitat suitability and bioactive components accumulation of the medicinal holoparasitic plant Cynomorium songaricum using machine learning models

Guilin Chen^1*Jiacheng Ji¹Xinxin Wei¹Huan Guan¹Zikang Jin¹Xin Yue²Zhuoran Jiang¹Youla Su^1*Shuying Sun^1*

• ¹Inner Mongolia University, Hohhot, China
• ²College of Pharmacy, Inner Mongolia Medical University, Hohhot, Inner Mongolia Autonomous Region, China

The efficacy of traditional Chinese medicine is determined by its bioactive components, which exhibit variability depending on environmental conditions and hereditary influences. In this study, we focus on Cynomorium songaricum Rupr., a medicinally significant species facing sustainability challenges. However, the ecological drivers governing its distribution, as well as the relationship between environmental factors and bioactive components, remain unclear. Thus, we sampled 28 representative distribution areas of C. songaricum in China. Employing Maximum Entropy (MaxEnt) modeling, we projected current and future (2050s-2090s) habitat suitability under four emission scenarios. Notably, species distribution exhibited expansion (8.03%-29.06% range increase across scenarios) with precipitation of the wettest month (BIO13) and soil pH emerging as key drivers (combined contribution >49%). Ultra-performance liquid chromatography (UPLC) fingerprinting combined with machine learning regression was applied to quantify six key bioactive components in C. songaricum, 3,4-dihydroxybenzaldehyde, catechin, epicatechin, ursolic acid, total phenolics, and crude polysaccharides-revealing significant concentration variations among geographically distinct populations. Slope gradient (slope), min temperature of coldest month (BIO6), precipitation of coldest quarter (BIO19), sunshine duration in growing season(hsdgs), and isothermality (BIO3) were identified as key regulatory factors influencing the accumulation of multiple components. Specifically, slope acted as a key shared negative regulator for 3,4-dihydroxybenzaldehyde, catechin, and crude polysaccharides. BIO6 served as a key shared positive regulator for catechin and total phenolics, while functioning as a key negative regulator for ursolic acid. BIO19 was identified as a key shared negative regulator for catechin and epicatechin. Hsdgs acted as a key positive regulator for ursolic acid while negatively regulating crude polysaccharides. Additionally, BIO3 served as a key shared positive regulator for both ursolic acid and total phenolics. This study provides the scientific basis for enabling targeted cultivation zones that balance therapeutic compound yield with arid ecosystem conservation.