Non-linearity of Metabolic Pathways Critically Influences the Choice of Machine Learning Model

Ophélie Lo-Thong-Viramoutou; Philippe Charton; Xavier F Cadet; Brigitte Grondin-Perez; Emma Saavedra; Cédric Damour; Frédéric Cadet

doi:10.3389/frai.2022.744755

Non-linearity of Metabolic Pathways Critically Influences the Choice of Machine Learning Model

Front Artif Intell. 2022 Jun 10:5:744755. doi: 10.3389/frai.2022.744755. eCollection 2022.

Authors

Ophélie Lo-Thong-Viramoutou^{1

2

3}, Philippe Charton^{1

2

3}, Xavier F Cadet⁴, Brigitte Grondin-Perez⁵, Emma Saavedra⁶, Cédric Damour⁵, Frédéric Cadet^{1

2

3}

Affiliations

¹ University of Paris, BIGR-Biologie Intégrée du Globule Rouge, Inserm, UMR_S1134, Paris, France.
² Laboratory of Excellence GR-Ex, Paris, France.
³ Laboratory DSIMB, UMR_S1134, BIGR, Inserm, Faculty of Sciences and Technology, University of La Reunion, Saint-Denis, France.
⁴ PEACCEL, Artificial Intelligence Department, Paris, France.
⁵ EnergyLab, EA 4079, Faculty of Sciences and Technology, University of La Reunion, Saint-Denis, France.
⁶ Departamento de Bioquímica, Instituto Nacional de Cardiología Ignacio Chávez, Mexico City, Mexico.

Abstract

The use of machine learning (ML) in life sciences has gained wide interest over the past years, as it speeds up the development of high performing models. Important modeling tools in biology have proven their worth for pathway design, such as mechanistic models and metabolic networks, as they allow better understanding of mechanisms involved in the functioning of organisms. However, little has been done on the use of ML to model metabolic pathways, and the degree of non-linearity associated with them is not clear. Here, we report the construction of different metabolic pathways with several linear and non-linear ML models. Different types of data are used; they lead to the prediction of important biological data, such as pathway flux and final product concentration. A comparison reveals that the data features impact model performance and highlight the effectiveness of non-linear models (e.g., QRF: RMSE = 0.021 nmol·min^-1 and R² = 1 vs. Bayesian GLM: RMSE = 1.379 nmol·min^-1 R² = 0.823). It turns out that the greater the degree of non-linearity of the pathway, the better suited a non-linear model will be. Therefore, a decision-making support for pathway modeling is established. These findings generally support the hypothesis that non-linear aspects predominate within the metabolic pathways. This must be taken into account when devising possible applications of these pathways for the identification of biomarkers of diseases (e.g., infections, cancer, neurodegenerative diseases) or the optimization of industrial production processes.

Keywords: Entamoeba histolytica glycolysis pathway; Trypanosoma cruzi detoxification pathway; artificial intelligence; drug target identification; machine learning; non-linear modeling; penicillin production.