Improving xylose-fermenting yeast for 2G ethanol production via constraint-based modeling combined with omics data analysis and machine learning

Abstract

The study of an organism's metabolism provides information relevant to energy production and phenotype. Constraint-based modeling (CBM) is an approach widely used to study biochemical networks, in particular the reconstruction of the metabolic network through genome-scale metabolic models (GSMM), which contains the reactions, genes that encode each enzyme and its metabolites in an organism. Nowadays, the integration of omics data (transcriptomics, proteomics and metabolomics) with GSMMsis still underdevelopment and it is a challenge on bioinformatics studies. Recently, Dr. Rocha has developed aset of platforms to perform different tasks regarding CBM: model reconstruction (merlin-merlin-sysbio.org), metabolic simulations and strain optimization (OptFlux(http://www.optflux.org/) and MewPy(https://github.com/BioSystemsUM/mewpy)), as well as a framework of omics integration with GSMM simulation, called Troppo. These assays allow improving the flux of reactions present in metabolic models and understanding contrasting conditions in-silico. However, in addition to being informative by themselves, metabolic fluxes can be improved through the features extracted from omics data and GSMM simulations to train machine learning (ML) models. In this project, we propose the development and application of methods based on omics data analysis supported by ML models, integrated with genome-scale metabolic models to understand the mechanisms behind the adaptative regulation of xylose-fermeting yeasts in 2G ethanol fermentation, a technology that produced bioethanol from non-food lignocellulosic biomass as a renewable alternative. We aim to build predictive models trained with omics data and metabolic simulation features that will enable us to learn how to improvethe engineered yeast to improve ethanol production. (AU)