Logo image
Machine learning and molecular dynamics reveal Jatropha curcas phytochemicals as natural modulators of lipid metabolism enzymes for enhanced biodiesel production
Journal article   Open access   Peer reviewed

Machine learning and molecular dynamics reveal Jatropha curcas phytochemicals as natural modulators of lipid metabolism enzymes for enhanced biodiesel production

Mustafa Alhaji Isa, Esther O. Babatunde, Haruna Yahaya Ismail and Lukhanyo Mekuto
Industrial crops and products, Vol.239, p.122443
01/01/2026
Handle:
https://hdl.handle.net/10210/519721

Abstract

Life Sciences & Biomedicine Science & Technology Agricultural Engineering Agriculture Agronomy
Lipid metabolism in non-edible oilseeds such as Jatropha curcas L. represents a critical target for improving biodiesel yields. This study identified phytochemicals from Jatropha curcas L. as potential allosteric modulators of lipid-metabolizing enzymes. A total of forty-six phytochemicals were screened, and seventeen were retained for bioactivity relevance. Eight were considered after absorption, distribution, metabolism, excretion, and toxicity profiling. A machine learning ensemble using Random Forest, Support Vector Machine, and Extreme Gradient Boosting (XGBoost) prioritized vitexin and isovitexin with probabilities of 0.24 and 0.21. Docking into predicted allosteric sites of Acyl-CoA Synthetase, Diacylglycerol Acyltransferase, Glycerol-3-phosphate Acyltransferase, and Lipase revealed strong binding affinities between -7.46 and -9.16 kcal/mol. Molecular dynamics simulations over 400 nanoseconds confirmed stable complexes, with isovitexin showing lower root-mean-square deviations and more compact radius of gyration profiles. Binding free energy estimates reached -45.38 kcal/mol. Quantum chemical calculations indicated frontier orbital gaps of 4.27 eV for vitexin and 4.47 eV for isovitexin. Functional mapping of Glycerol-3-phosphate Acyltransferase suggested modulation by these compounds may enhance lipid flux, supporting biodiesel production. These results lay a computational foundation for future experimental validation and metabolic engineering approaches.
pdf
Research (19)10.81 MBDownloadView
Open Access CC BY V4.0
url
https://doi.org/10.1016/j.indcrop.2025.122443View
Published (Version of record) Open

Metrics

1 Record Views

Details

Logo image