EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explainable machine learning for predicting glycyrrhizic acid accumulation in Glycyrrhiza glabra L. using environmental drivers and SHAP-based interpretation

Emran Dastres, Hassan Esmaeili

Journal of Food Composition and Analysis · 2026-10 · 158 · 109468

DOI: 10.1016/j.jfca.2026.109468 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Food Composition and Analysis, 2026-10. Verfasst von Emran Dastres und 1 weiteren Autor:innen. Die Arbeit nennt 117 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Glycyrrhizic acid, the principal bioactive triterpenoid saponin in licorice ( Glycyrrhiza glabra L.), exhibits pronounced spatial variability. Understanding its drivers is essential for stock conservation and cultivation. Root samples from 41 wild populations in Fars Province, Iran, were quantified via validated HPLC-DAD and evaluated against 18 climatic, edaphic, and topographic variables. Following Boruta selection, CatBoost, XGBoost, and Random Forest models were evaluated using repeated 5-fold cross-validation. CatBoost achieved superior predictive performance (R2 = 0.875, RMSE = 0.210, MAE = 0.155, MAPE = 10.30%, CCC = 0.925, NSE = 0.870). Annual rainfall, soil electrical conductivity (EC), and elevation were dominant drivers. SHAP analysis delineated critical physiological thresholds: glycyrrhizic acid accumulation was strongly promoted under annual rainfall > 300 mm, soil EC < 3.5 dS/m, and elevation < 2200 m, reflecting water stress, osmotic pressure, and high-altitude UV-B constraints. Spatial mapping identified Beyza, northern Shiraz, Marvdasht, and southern Abadeh as optimal production zones, showing high congruence with wild population densities. While natural thresholds constrain wild harvesting, agronomic practices (e.g., irrigation, soil reclamation) can buffer these limits in cultivation. Overall, this explainable machine-learning framework bridges geospatial modeling and phytochemical quality assessment, providing precise decision-support tools for medicinal plant conservation and targeted agricultural development.

Zitieren

APA

Dastres, E., & Esmaeili, H. (2026). Explainable machine learning for predicting glycyrrhizic acid accumulation in Glycyrrhiza glabra L. using environmental drivers and SHAP-based interpretation. Journal of Food Composition and Analysis, 158, 109468. https://doi.org/10.1016/j.jfca.2026.109468

Vancouver

Dastres E, Esmaeili H. Explainable machine learning for predicting glycyrrhizic acid accumulation in Glycyrrhiza glabra L. using environmental drivers and SHAP-based interpretation. Journal of Food Composition and Analysis. 2026;158:109468. doi:10.1016/j.jfca.2026.109468

BibTeX
@article{Dastres2026,
  author = {Dastres, Emran and Esmaeili, Hassan},
  title = {Explainable machine learning for predicting glycyrrhizic acid accumulation in Glycyrrhiza glabra L. using environmental drivers and SHAP-based interpretation},
  journal = {Journal of Food Composition and Analysis},
  year = {2026},
  volume = {158},
  pages = {109468},
  doi = {10.1016/j.jfca.2026.109468}
}
Kostenlos starten →