Micropropagation of Scoparia dulcis L. through machine learning-assisted optimization and comparative LC-MS phytochemical profiling
Abstract
Scoparia dulcis L., a medicinally important herb widely recognized for its antidiabetic, anti-inflammatory, and antioxidant properties, was subjected to machine learning (ML) based optimization for micropropagation, followed by the comparative phytochemical evaluation of both field-grown and micropropagated plants in the present study. Leaf explants were cultured on Murashige and Skoog (MS) medium supplemented with varying concentrations of plant growth regulators (PGRs). The combination of MS + BAP 1mg/L + NAA 0.5mg/L yielded the highest shoot induction with a maximum of 10.33 ± 1.28 shoots per explant, while half-strength MS media supplemented with IAA 1.5mg/L proved most effective for root induction, producing 6.93 ± 0.61 roots. Well-rooted plantlets were successfully hardened and established under ex vitro conditions, confirming the reproducibility of the protocol for large-scale propagation of this medicinally valuable species. Experimental datasets of shoot and root induction were employed for training various machine learning algorithms; among them, Random Forest (RF) proved to be the most suitable for shoot number prediction, whereas Gaussian Process Regression (GPR) was the best choice for root number prediction. The experimentally tested optimal concentrations of hormones predicted by the machine learning models have outperformed conventional ones in two out of three shoot parameters and all three root parameters, with RF-predicted shoot numbers of 10.15 validated to 11.8 ± 0.92 shoots and GPR-predicted root numbers of 7.76 validated to 7.26 ± 0.35 roots. This study also aimed to perform an LC-MS metabolic profile comparison on both the field-grown and micropropagated plants of S. dulcis. It appears that there are both shared and unique phytochemical compounds between the two different sources of plant material (i.e., micropropagated and field-grown). Hence, this indicates that in vitro tissue culture may bring changes in the expression of metabolites, indicating that the conditions of the in vitro plant growth environment may play a role in regulating the biosynthesis of specific metabolites. These findings establish a validated, ML-assisted micropropagation protocol for S. dulcis and provide the first comparative phytochemical evidence in support of its use as a sustainable substitute for field-grown plants.
Related
- Comparative phytochemical profiling and bioactivity evaluation of three Indonesian pomegranate (Punica granatum) peel varieties: Antioxidant, antibacterial, and in vitro antiproliferative activities
- Phytochemical Profiling, In Vitro α-Amylase Inhibitory and Antioxidant Activities of Madhuca longifolia Leaf and Bark Extracts with Molecular Docking Analysis
- Comparative phytochemical characterization and in vitro biological activities of coffee (Coffea arabica) and loquat (Eriobotrya japonica) seed extracts
- Integrated Phytochemical Profiling, Bioactive Compound Identification and Functional Evaluation of Cissus quadrangularis L.
- Phytochemical Profiling, Molecular Docking Study, and Gastroprotective Effect of Matricaria chamomilla L. Aqueous Extract in Ethanol‐Induced Gastric Ulcer in Mice
BibTeX
@article{purohit2026micropropagation,
title = {{Micropropagation of Scoparia dulcis L. through machine learning-assisted optimization and comparative LC-MS phytochemical profiling}},
author = {Nikita Purohit and Deepa Sankar Parasurama},
journal = {Frontiers in Plant Science},
year = {2026},
month = nov,
doi = {10.3389/fpls.2026.1927416},
url = {https://www.frontiersin.org/articles/10.3389/fpls.2026.1927416},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Micropropagation of Scoparia dulcis L. through machine learning-assisted optimization and comparative LC-MS phytochemical profiling
AU - Nikita Purohit
AU - Deepa Sankar Parasurama
PY - 2026
DA - 2026/11/06
JO - Frontiers in Plant Science
DO - 10.3389/fpls.2026.1927416
UR - https://www.frontiersin.org/articles/10.3389/fpls.2026.1927416
AB - Scoparia dulcis L., a medicinally important herb widely recognized for its antidiabetic, anti-inflammatory, and antioxidant properties, was subjected to machine learning (ML) based optimization for micropropagation, followed by the comparative phytochemical evaluation of both field-grown and micropropagated plants in the present study. Leaf explants were cultured on Murashige and Skoog (MS) medium supplemented with varying concentrations of plant growth regulators (PGRs). The combination of MS + BAP 1mg/L + NAA 0.5mg/L yielded the highest shoot induction with a maximum of 10.33 ± 1.28 shoots per explant, while half-strength MS media supplemented with IAA 1.5mg/L proved most effective for root induction, producing 6.93 ± 0.61 roots. Well-rooted plantlets were successfully hardened and established under ex vitro conditions, confirming the reproducibility of the protocol for large-scale propagation of this medicinally valuable species. Experimental datasets of shoot and root induction were employed for training various machine learning algorithms; among them, Random Forest (RF) proved to be the most suitable for shoot number prediction, whereas Gaussian Process Regression (GPR) was the best choice for root number prediction. The experimentally tested optimal concentrations of hormones predicted by the machine learning models have outperformed conventional ones in two out of three shoot parameters and all three root parameters, with RF-predicted shoot numbers of 10.15 validated to 11.8 ± 0.92 shoots and GPR-predicted root numbers of 7.76 validated to 7.26 ± 0.35 roots. This study also aimed to perform an LC-MS metabolic profile comparison on both the field-grown and micropropagated plants of S. dulcis. It appears that there are both shared and unique phytochemical compounds between the two different sources of plant material (i.e., micropropagated and field-grown). Hence, this indicates that in vitro tissue culture may bring changes in the expression of metabolites, indicating that the conditions of the in vitro plant growth environment may play a role in regulating the biosynthesis of specific metabolites. These findings establish a validated, ML-assisted micropropagation protocol for S. dulcis and provide the first comparative phytochemical evidence in support of its use as a sustainable substitute for field-grown plants.
SN - 1664-462X
ER -