This study investigated the effects of composition of carbohydrate, protein, and fiber fractions─individually and in mixtures─on anaerobic digestion (AD) performance using an integrated kinetic, statistical, and metagenomic framework. Batch AD experiments were designed according to the augmented simplex centroid protocol for three components. Time-resolved profiles of chemical oxygen demand (COD), volatile fatty acids (VFAs), and biogas yields (CH4 and CO2) were analyzed using a Contois-based kinetic model to quantify acidogenic and methanogenic parameters. Mixture regression modeling was employed to resolve the individual and interactive contributions of substrate fractions to process yields and kinetic constants. Microbial community dynamics were elucidated through metagenomic analysis of bacterial and archaeal populations. The results demonstrated that substrate composition exerts a statistically significant influence on both reaction kinetics and methane yield. Carbohydrate-rich systems exhibited high acidogenesis rates but were prone to VFA accumulation and reduced methane conversion efficiency. Protein-containing systems showed elevated VFA and CO2 production, consistent with amino acid fermentation and ammonia-related inhibition. In contrast, fiber-dominant and balanced ternary systems achieved superior COD removal, stable VFA turnover, and the highest methane yields. Regression analysis identified fiber as the primary positive contributor to methane yield, while negative interaction terms explained yield suppression in rapidly fermentable or nitrogen-rich mixtures. Metagenomic data revealed that fiber-rich systems favored syntrophic bacterial consortia and hydrogenotrophic methanogens, underpinning their kinetic stability and enhanced methane production. This study provides mechanistic insights and quantitative tools for rational feedstock design, supporting the development of robust, high-efficiency AD systems for diverse organic waste streams.
Madani et al. (2026) studied this question.