Funding informationGenome-scale metabolic models are powerful tools to understand and engineer cellular systems facilitating their use as cell factories. This is especially true for microorganisms with known genome sequences from which nearly complete sets of enzymes and metabolic pathways are determined, or can be inferred. Yeasts are highly diverse eukaryotes whose metabolic traits have long been exploited in industry, and although many of their genome sequences are available, few genome-scale metabolic models have so far been produced. For the first time, we reconstructed the genomescale metabolic model of the hybrid yeast Zygosaccharomyces parabailii, which is a member of the Z. bailii sensu lato clade notorious for stress-tolerance and therefore relevant to industry. The model comprises 3096 reactions, 2091 metabolites, and 2413 genes. Our own laboratory data were then used to establish a biomass synthesis reaction, and constrain the extracellular environment. Through constraint-based modeling, our model reproduces the co-consumption and catabolism of acetate and glucose posing it as a promising platform for understanding and exploiting the metabolic po-1 2 MARZIA DI FILIPPO ET AL tential of Z. parabailii.
K E Y W O R D SGenome-scale metabolic model/ constraint-based modeling / hybrid yeast / stress tolerance / Zygosaccharomyces parabailii 1 | INTRODUCTION Current genome sequencing technologies allow a fast and cheap overview into the genetic composition of virtually any organism, but connecting such genotypes to observed phenotypes remains a challenge. The reconstruction of genome-scale metabolic networks provide structured frameworks to represent the biochemical transformations within a target organism as complex genotype-phenotype relationships. Afterwards, different modeling approaches can be used to simulate, understand, predict and eventually control the behavior of such genome-scale metabolic networks.Flux Balance Analysis (FBA) is a widely used constraint-based modeling approach that relies on linear programming and the optimization of a given objective function (e.g., maximization of growth) for the determination of the metabolic model flux distribution [1,2]. This approach is based on the assumption that organisms operate under a series of constraints limiting their possible functions, and leading to the definition of allowable cell phenotypes from defined metabolic networks [3].In the last two decades, genome-wide reconstructions of metabolism have been produced for a plethora of model organisms, spanning from bacteria to higher eukaryotes [4]. The original versions of these models typically undergo incremental improvements. In particular, 10 different versions of the Saccharomyces cerevisiae metabolic network have been produced to date, by implementing cellular compartments, curated reactions, standard nomenclature, and even transcriptional regulation, as reviewed in [5]. However, less extensive efforts have been dedicated to other so-called non-conventional or non-Saccharomyces yeast species, de...