ADToolbox: Incorporating Metagenomics Data for Improved Prediction of Anaerobic Digestion Dynamics

Abstract

Handling the global food waste requires sustainable solutions. Anaerobic digestion (AD) is a common approach for waste remediation and bioenergy production which has been traditionally used mainly for methane production. Shifting the focus from methane to volatile fatty acids (VFAs), such as acetic, propionic, and butyric acids, offers promising alternatives due to their diverse applications and higher added value. Achieving desired AD product distributions requires a comprehensive understanding of factors like temperature, pH, feedstock composition, and the complex microbial dynamics inherent in AD. Various AD modeling approaches exist, from simple equations to complex flux balance analysis (FBA) and machine learning (ML) model. The Anaerobic Digestion Model No. 1 (ADM 1) is a commonly used kinetic model, striking a reasonable balance between parameter requirements and biochemical details involved in the model. Yet, it falls short in capturing specific VFAs like caproic acid and integrating microbial information directly. We present ADToolbox, a Python package for modeling AD metabolism. ADToolbox incorporates metagenomic information into an enhanced ADM model. The model accommodates a more detailed feedstock degradation model and VFA and methanogenesis metabolism. ADToolbox provides a variety of interfaces such as command line interface and an interactive web interface line interface, and a Python API, facilitating large-scale metagenomic analyses and modeling simulations. In this article we indicate that prioritizing the microbial aspect of AD enhances flexibility and predictive power in terms of VFA production accuracy, contributing to sustainable waste management strategies. Explore ADToolbox at https://chan-csu.github.io/ADToolbox for detailed documentation.