Medusa: Software to build and analyze ensembles of genome-scale metabolic network reconstructions

Abstract

Uncertainty in the structure and parameters of networks is ubiquitous across computational biology. In constraint-based reconstruction and analysis of metabolic networks, this uncertainty is present both during the reconstruction of networks and in simulations performed with them. Here, we present Medusa, a Python package for the generation and analysis of ensembles of genome-scale metabolic network reconstructions. Medusa builds on the COBRApy package for constraint-based reconstruction and analysis by compressing a set of models into a compact ensemble object, providing functions for the generation of ensembles using experimental data, and extending constraint-based analyses to ensemble scale. We demonstrate how Medusa can be used to generate ensembles and perform ensemble simulations, and how machine learning can be used in conjunction with Medusa to guide the curation of genome-scale metabolic network reconstructions.

Novel Drivers of Virulence in Clostridioides difficile Identified via Context-Specific Metabolic Network Analysis

“…Coefficients within the formulations of DNA replication, RNA replication, and protein synthesis component reactions were adjusted by genomic nucleotide abundances and codon frequencies to yield strain-specific biomass objective functions ( 60 ). To successfully simulate growth, we next performed an ensemble-based parsimonious flux balance analysis (pFBA) gap-filling approach ( 61 , 62 ), utilizing a metabolic reaction database centered on Gram-positive anaerobic bacterial metabolism (see Materials and Methods). Gap-filling refers to the automated process of identifying incomplete metabolic pathways due to an absence of genetic evidence that are necessary for in silico growth and addition of the minimal functionality needed to achieve flux through these pathways ( 63 ).…”

Enzyme-constrained models predict the dynamics of Saccharomyces cerevisiae growth in continuous, batch and fed-batch bioreactors

“…Despite of the substantially improved predictive power of the model, however the protein availability constraint was not enough to yield accurate predictions of all intracellular fluxes due to the highly dimensional solution space and the absence of regulatory information in the model. Predictions could improve considering space limitation in cell membranes as additional constraint or by the creation of an ensemble model [41], [42]. Ultimately, when ecModels are applied to strain design, this limitation can be overcome using proteomic data instead of a single constraint on the protein content of the cells [9] .…”

XomicsToModel: Omics data integration and generation of thermodynamically consistent metabolic models

“…Last, model feasibility check results are given, including any reaction that was relaxed if the model was not feasible. Find thermodynamically consistent subset 18 | A thermodynamically consistent draft model. Additionally, a message is given with the parameters used to identify the thermodynamically consistent subset.…”

“…Moreover, supplementary functions are provided to estimate the predictive capacity of an extracted model, given independent data. These features will facilitate analysis by complementary software with capabilities for machine learning from model ensembles [18]. The application of the XomicsToModel pipeline to extract an ensemble of dopaminergic neuronal metabolic models is presented elsewhere [24].…”

“…Find thermodynamically consistent subset 18 | A thermodynamically consistent draft model. Additionally, a message is given with the parameters used to identify the thermodynamically consistent subset.…”