Assessing the accuracy of total antioxidant capacity methods using machine learning approaches
- 1Department of Food Science, College of Agriculture and Veterinary Medicine, United Arab Emirates University
- 2Independent Researcher
Abstract
Accurate estimation of antioxidant capacity is complicated by the interactions between the antioxidant molecules, the differences in redox assay chemistries, and inconsistencies in calibration. To assess the validity of four antioxidant assays—2,2′-azino-bis-3-ethylbenzthiazoline-6-sulphonic acid (ABTS), 1,1-diphenyl-2-picrylhydrazyl (DPPH), Ferric Reducing Antioxidant Power (FRAP), and Total Phenolic Content (TPC)—by integrating assay-specific analysis, supervised machine-learning modelling and residual-based classification of antioxidant standard-mixture responses. Replicate-level data from CB1–CB5 phenolic standard mixtures were used to train and compare Linear Regression (LR), Random Forest (RF) and tuned XGBoost (XGB) models. Models were evaluated using grouped cross-validation and an independent grouped test set to avoid leakage between technical replicates from the same experimental condition. The best-performing model was refitted on the full dataset, and residual deviations between observed and model-predicted gallic acid equivalent (GAE) values were used to assign operational interaction classes. Linear Regression showed weak predictive performance under grouped validation, whereas ensemble models improved prediction accuracy. Tuned XGBoost achieved the best overall performance, with the highest grouped cross-validated R2 and independent test-set R2, and the lowest prediction errors among the models tested. Residual-based classification revealed assay- and mixture-dependent response patterns, with most observations classified as additive and smaller subsets showing positive or negative deviations from model-predicted GAE values. Combining assay-specific analysis, grouped model validation and residual-based interaction classification provides a transparent framework for evaluating antioxidant assay behaviour in defined standard mixtures. The results highlight substantial assay- and composition-dependent variability and support the use of machine-learning models to identify mixture responses that deviate from expected patterns. These model-derived interaction classes should be interpreted as analytical indicators for further investigation rather than direct mechanistic proof of biochemical synergy or antagonism.
Related
- Association of dietary total antioxidant capacity, dietary inflammatory index, and dietary phytochemical index with sarcopenia in middle-aged and older Iranian women: a case-control study
- Sugar profile, phenolic compounds, and antioxidant capacity of the most preferred commercial fruit juices in a sample of school-age children: a combined laboratory and survey-based study
- Commercial Oat‐Based Cosmetic Ingredients Promote Antioxidant Capacity and Reduce Inflammatory Markers in Skin Cellular Stress Models: An Insight Into Molecular Mechanisms In Vitro
- Machined versus cast abutments for dental implants: a 1-year within-patient multicentre randomized controlled trial assessing marginal seal capacity and outcomes
- Next-generation non-Saccharomyces yeast derivatives: Fine tuning antioxidant capacity through inactivation process
BibTeX
@article{alam2026assessing,
title = {{Assessing the accuracy of total antioxidant capacity methods using machine learning approaches}},
author = {Muneeba Zubair Alam and Delphine Vincent and Nadiya Samad and Shamma Hamdan ALjnaibi and Hissa Abdulla Ali and Afaf Kamal-Eldin},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-63770-y},
url = {https://doi.org/10.1038/s41598-026-63770-y},
note = {CC BY},
}
RIS
TY - JOUR
TI - Assessing the accuracy of total antioxidant capacity methods using machine learning approaches
AU - Muneeba Zubair Alam
AU - Delphine Vincent
AU - Nadiya Samad
AU - Shamma Hamdan ALjnaibi
AU - Hissa Abdulla Ali
AU - Afaf Kamal-Eldin
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-63770-y
UR - https://doi.org/10.1038/s41598-026-63770-y
AB - Accurate estimation of antioxidant capacity is complicated by the interactions between the antioxidant molecules, the differences in redox assay chemistries, and inconsistencies in calibration. To assess the validity of four antioxidant assays—2,2′-azino-bis-3-ethylbenzthiazoline-6-sulphonic acid (ABTS), 1,1-diphenyl-2-picrylhydrazyl (DPPH), Ferric Reducing Antioxidant Power (FRAP), and Total Phenolic Content (TPC)—by integrating assay-specific analysis, supervised machine-learning modelling and residual-based classification of antioxidant standard-mixture responses. Replicate-level data from CB1–CB5 phenolic standard mixtures were used to train and compare Linear Regression (LR), Random Forest (RF) and tuned XGBoost (XGB) models. Models were evaluated using grouped cross-validation and an independent grouped test set to avoid leakage between technical replicates from the same experimental condition. The best-performing model was refitted on the full dataset, and residual deviations between observed and model-predicted gallic acid equivalent (GAE) values were used to assign operational interaction classes. Linear Regression showed weak predictive performance under grouped validation, whereas ensemble models improved prediction accuracy. Tuned XGBoost achieved the best overall performance, with the highest grouped cross-validated R2 and independent test-set R2, and the lowest prediction errors among the models tested. Residual-based classification revealed assay- and mixture-dependent response patterns, with most observations classified as additive and smaller subsets showing positive or negative deviations from model-predicted GAE values. Combining assay-specific analysis, grouped model validation and residual-based interaction classification provides a transparent framework for evaluating antioxidant assay behaviour in defined standard mixtures. The results highlight substantial assay- and composition-dependent variability and support the use of machine-learning models to identify mixture responses that deviate from expected patterns. These model-derived interaction classes should be interpreted as analytical indicators for further investigation rather than direct mechanistic proof of biochemical synergy or antagonism.
SN - 2045-2322
ER -