Predicting Antioxidant Synergism via Artificial Intelligence and Benchtop Data

Abstract

Lipid oxidation is a major issue affecting products containing unsaturated fatty acids as ingredients or components, leading to the formation of low molecular weight species with diverse functional groups that impart off-odors and off-flavors. Aiming to control this process, antioxidants are commonly added to these products, often deployed as combinations of two or more compounds, a strategy that allows for lowering the amount used while boosting the total antioxidant capacity of the formulation. While this approach allows for minimizing the potential organoleptic and toxic effects of these compounds, predicting how these mixtures of antioxidants will behave has traditionally been one of the most challenging tasks, often leading to simple additive, antagonistic, or synergistic effects. Approaches to understanding these interactions have been predominantly empirically driven but thus far, inefficient and unable to account for the complexity and multifaceted nature of antioxidant responses. To address this current gap in knowledge, we describe the use of an artificial intelligence model based on deep learning architecture to predict the type of interaction (synergistic, additive, and antagonistic) of antioxidant combinations. Here, each mixture was associated with a combination index value (CI) and used as input for our model, which was challenged against a test (n = 140) data set. Despite the encouraging preliminary results, this algorithm failed to provide accurate predictions of oxidation experiments performed in-house using binary mixtures of phenolic antioxidants and a lard sample. To overcome this problem, the AI algorithm was then enhanced with various amounts of experimental data (antioxidant power data assessed by the TBARS assay), demonstrating the importance of having chemically relevant experimental data to enhance the model’s performance and provide suitable predictions with statistical relevance. We believe the proposed method could be used as an auxiliary tool in benchmark analysis routines, offering a novel strategy to enable broader and more rational predictions related to the behavior of antioxidant mixtures.

Spectrofluorimetric determination of butylated hydroxytoluene and butylated hydroxyanisole in their combined formulation: application to butylated hydroxyanisole residual analysis in milk and butter

“…BHT and BHA combinations show synergistic antioxidant effect 2 , 3 , so they are added either individually or in combination with each other or other antioxidants such as gallates or tertiary butylhydroquinone (TBHQ) to beverage whiteners, breakfast cereals, butter oil, anhydrous milk fat, ghee, chewing gum, cocoa and chocolate products, dried vegetables, seaweeds, nuts, seeds, fat spreads, dairy fat spreads and blended spreads, food supplements, frozen fish, fish fillets, and fish products, including mollusks, crustaceans, and echinoderms. In pharmaceutical preparations and cosmetic formulations, they act as preservatives and antioxidants for those preparations and formulations containing fats and oils.…”

eutXG: A Machine-Learning Model to Understand and Predict the Melting Point of Novel X-Bonded Deep Eutectic Solvents

“…Aiming to address this gap in knowledge, it is worth mentioning two excellent reviews from Vega’s and Billen’s groups, which offer examples and critical insights about the advantages and limitations of thermodynamic modeling approaches to understand and predict the properties of DES. In addition to those, it is worth noting the emergence of methods based on artificial neural networks ,,− which, despite their increasing capabilities, are highly dependent on the quality of the database used for the training step, , the notation used to encode molecular structures, and the architecture of the network . Perhaps one of the biggest limitations of these methods is that many machine-learning models provide limited generalization capacity, potentially leading to inaccurate predictions when presented new chemical functionalities .…”

“…In addition to those, it is worth noting the emergence of methods based on artificial neural networks ,,− which, despite their increasing capabilities, are highly dependent on the quality of the database used for the training step, , the notation used to encode molecular structures, and the architecture of the network . Perhaps one of the biggest limitations of these methods is that many machine-learning models provide limited generalization capacity, potentially leading to inaccurate predictions when presented new chemical functionalities . In this regard, it is important to highlight a few DES-related databases recently published by Omar et al, Shi et al, Lavrinenko et al, Hou et al, Hopkins et al, Abdollahzadeh et al, and Makarov et al that can greatly facilitate the advancement of these methods.…”