Computational framework for integrated biorefinery simulation (BPC)

Ruby Sandy Moreno Mejia, Arnulfo Villanueva-Castillo, C.F. Pastelín, Claudia Mancilla-Simbro, Briseida Lucía Blanco Alfonso, ERICK FERNANDEZ MENESES, Miguel Ángel Zambrano, Fabiola Rodríguez-Andrade, Juan Ricardo Cruz-Aviña, J. Jesús Hinojosa, Hermilo Lucio-Castillo

This repository contains the MATLAB scripts, input data, and model parameters for the computational simulation of an integrated multi-product biorefinery processing 2,740 t/day of mixed sugarcane bagasse and pineapple residues. The simulation encompasses four processing stages — pretreatment, delignification, enzymatic saccharification, and multi-organism fermentation — using an 8-member microbial consortium producing five bioproducts (ethanol, butanol, PHA bioplastic, hydrogen, and biodiesel). Kinetic models include Haldane (substrate inhibition), Monod (fermentation), and modified Gompertz (H2 production). Additionally, a bioinformatic pipeline for metabolic engineering of Saccharomyces cerevisiae is included, incorporating heterologous ABE pathway genes (7 genes for butanol) and PHA synthesis genes (4 genes from Cupriavidus necator), optimized through 6 CRISPR-mediated knockouts. CONTENTS:- matlab/ : 12 MATLAB R2025a scripts (master_generar_todo.m runs the full pipeline)- data/ : 13 CSV files with input parameters organized by category (composition, fermentation, consortium, metabolic engineering, costs) KEY FINDINGS:- Hemicellulose solubilization: 70-90% (k = 0.08/h)- Lignin extraction: 60-85% (k = 0.06/h)- Potential glucose production: 970-1,214 t/day- Butanol theoretical yield: 0.35-0.40 g/g glucose (realistic limit ~13.5 g/L due to yeast tolerance)- PHA yield: 0.38-0.42 g/g glucose- Strain construction cost: $1,410-1,600 USD All kinetic parameters were drawn from the published literature. Parameter uncertainty was propagated through Monte Carlo simulation (N = 10,000) with ±25% variation. A stoichiometric carbon balance and global sensitivity analysis (Morris method) are included. REQUIREMENTS:- MATLAB R2022a or later- Statistics and Machine Learning Toolbox FUNDING:Vicerrectoría de Investigación y Estudios de Posgrado (VIEP), Benemérita Universidad Autónoma de Puebla, grant VIEP reg. 00435-PV/2024.