Digital modeling of metformin and diet interactions on gut-microbiota metabolism in prediabetic patients
Open Access
- 6 January 2026
- journal article
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 31, 250-262
Abstract
Prediabetes confers a high risk of progressing to type 2 diabetes mellitus (T2DM). While metformin, a first-line T2DM therapy, improves glycemic control in prediabetes, its effects on the gut microbiota and host metabolic shifts remain poorly understood. Here, we applied a genome-scale community metabolic modeling to build a personalized “digital microbiota” for analyzing the metabolic activity of gut microbes in 106 samples of Mexican prediabetic patients, distributed among patients without treatment and patients treated with metformin over baseline, 6 and 12 months. To contrast microbial metabolic activity across groups and explore how diet modulates it, we simulated computationally the microbial metabolic fluxes under Western, Mediterranean, and traditional Milpa diets across the three groups. As expected, in general terms in silico dietary intervention changes the metabolic responses in the microbiota profiles among the stages, suggesting specific combinations of diets that favor the production of relevant metabolites for wellness, such as amino sugars, short-chain fatty acids, and bile acid exchange fluxes. Furthermore, by selecting two individuals across the entire time as case studies, we provide a proof of concept for in silico personalized diet design. These examples illustrate how the concept of personalized digital microbiota could be leveraged to optimize dietary strategies and potentially improve outcomes in prediabetic patients.Keywords
Funding Information
- Universidad Nacional Autónoma de México (IN213824)
- Universidad Nacional Autónoma de México (CVU 855825)
- Instituto Nacional de Medicina Genómica
- Secretaría de Ciencia, Humanidades, Tecnología e Innovación (CVU 702604)
This publication has 44 references indexed in Scilit:
- One path, two solutions: Network-based analysis identifies targetable pathways for the treatment of comorbid type II diabetes and neuropsychiatric disordersComputational and Structural Biotechnology Journal, 2024
- Secure and privacy-preserving automated machine learning operations into end-to-end integrated IoT-edge-artificial intelligence-blockchain monitoring system for diabetes mellitus predictionComputational and Structural Biotechnology Journal, 2023
- Diagnosis and Management of PrediabetesJAMA, 2023
- Exercise and microbiome: From big data to therapyComputational and Structural Biotechnology Journal, 2023
- The combination of linagliptin, metformin and lifestyle modification to prevent type 2 diabetes (PRELLIM). A randomized clinical trialMetabolism, 2020
- The HDL lipidome is widely remodeled by fast food versus Mediterranean diet in 4 daysMetabolomics, 2019
- Assessment of a Personalized Approach to Predicting Postprandial Glycemic Responses to Food Among Individuals Without DiabetesJAMA Network Open, 2019
- Aberrant intestinal microbiota in individuals with prediabetesDiabetologia, 2018
- Interaction of gut microbiota with bile acid metabolism and its influence on disease statesApplied Microbiology and Biotechnology, 2016
- Personalized Nutrition by Prediction of Glycemic ResponsesCell, 2015