The human gut microbiome significantly affects our health. Fiber intake plays an important role here, yet identifying the optimal fiber mix for gut health is challenging. Here we aim to conduct in vivo, ex vivo and in vitro studies on various fibers to study how specific cazymes present in the gut microbes break down fibers and modulate the gut microbiome. This information will be used to develop an AI model to predict fiber combinations that will modulate the gut microbiome towards health.

The proposed research aligns with the Kennis- en Innovatieagenda (KIA) and Meerjarige Missiegedreven Innovatie Programma’s (MMIP) of the Dutch Ministry of Agriculture, Nature and Food Quality (LNV) by addressing public health, sustainability, and innovation in food production. It focuses on the vital role of dietary fiber in promoting health and reducing diet-related diseases, such as obesity, type 2 diabetes, and cardiovascular disease, through the development of an AI tool. This tool will guide the food industry in creating healthier fiber-rich products tailored to microbiome responses, supporting the MMIP's mission to foster innovative, sustainable, and health-promoting food systems.
The project also addresses the KIA's sustainability goals by promoting the valorization of fiber-rich side streams from the plant-protein industry, reducing food waste, and encouraging a shift towards plant-based diets. By leveraging underutilized by-products, it enhances the value chain for plant-based protein production while contributing to environmental benefits and supporting farmers in adopting sustainable practices.
In summary, this research contributes to public health improvement, supports the development of innovative and sustainable food products, and aligns with the government's mission to advance smart technologies and sustainable diets for a healthier society.

The FIBERME project aligns with the governmental mission to develop healthier food products and processes by focusing on dietary fiber's role in promoting health. Increased fiber intake is linked to reduced risks of obesity, type 2 diabetes, and cardiovascular diseases. The project emphasizes microbiome-inspired fiber selection to create personalized, optimized, and effective dietary interventions, leveraging AI to predict the microbiome's response to fibers and maximize their impact.
FIBERME also addresses food waste reduction by transforming fiber-rich plant-based by-products into valuable ingredients. Its outcomes will guide the selection of bio-based glycans to selectively stimulate beneficial gut microbiome members, enhancing health and well-being. By integrating innovation and technology, the project fully supports the mission of fostering healthier food choices through advanced processes.

FIBERME is an initiative that aims to use smart technologies and data-mining innovation that includes an online artificial intelligence (AI) tool, to predict the positive effects of natural, plant-based, and sustainable fiber on the human gut microbiome. The specific deliverables of FIBERME include:
1. Analysis of the Fiber-Gut-Microbe Axis: This involves establishing a coherent and comprehensive in vivo dataset that contains information on how different fibers with distinct structures behave in realistic scenarios within the human gut-microbial-ecosystem. The envisioned understanding of the interactions between fibers and the gut microbiome is crucial for the prediction of their potential health effects. The in vivo experiments and their interpretation are expected to be finished by Q2, 2027
2. Creation of a Comprehensive Fermentation Database: FIBERME will compile a vast database containing information from over 2,000 fermentations using various fiber-rich sources. The database will consist of all the information gathered from the in vivo, ex vivo and in vitro trials and will encompass a wide range of information such as sugar composition of fibers used as input, their modulatory effects microbiome species and cazyme composition, bioconversion activities (e.g., formation of organic acids through fiber fermentation). This data will be essential for understanding how different fibers are metabolized and fermented by human gut microbes. This database will be ready by Q2, 2026.
3. Machine Learning Algorithms for Prediction: The project will employ machine learning techniques, to predict how specific microorganisms can consume particular fibers. The predictions will ideally be based on the genomic information of the microorganisms and by proxy the presence of specific carbohydrate-active enzymes (cazymes) that play a role in fiber degradation. This will result in an AI model capable of providing valuable information on how various fibers can positively modulate the human gut microbiome. The final model is expected to be finished by Q3, 2027
4. AI Tool Interface: As part of the FIBERME project, an interface will be developed to access and query the data generated by the AI model. This user-friendly interface will allow industrial partners and scientists to gain direct insights into the potential effects of different fibers on the human gut microbiome. The final interface is expected to be finished by Q1, 2028