-
This repository contains the work on enzyme-constrained genome-scale model of oleaginous, non-conventional yeast Rhodotorula toruloides.
-
Repository consists of code, data, model files and result files. Result visualisation is not provided.
- As a starting point for this work, a metabolic genome-scale model, named rhto-GEM, developed by Tiukova et al. 2019, was used.
- The initial model files used in this work are available here.
- The scripts were written in Matlab.
- Enzymatic constraints were applied and models were generated using the GECKO toolbox.
- The modelling was done with the RAVEN toolbox.
- Metabolic network maps are available upon request. They were constructed for internal use using Escher software.
- Script
edit_rhtoGEM.m
introduces reactions for alternative xylose assimilation pathway. - Script
reconstruct_ecRhtoGEM.m
generates condition-specific enzyme-constrained models with integrated absolute proteomics data. Models are reconstructed based on optimized parameters, including manually curated enzymatic kcat values, as provided inmanualModifications.m
. After downloading the GECKO folder, model reconstruction involves the following steps: 1)geckomat/enhanceGEM.m
pipeline, which creates the first version of ec-model with an enzyme pool that is used for screening required kcat modifications; 2)geckomat/utilities/integrate_proteomics/generate_protModels.m
pipeline, which sets constraints on high-quality measured individual enzymes from the provided proteomics dataset. Folderresults/generate_protModels_pipeline
provides the track of enzyme abundances which were automatically modified in order to reach the experimental conditions. - Script
analyze_ecRhtoGEM.m
runs random sampling, calculates flux mean value and standard deviation, calculates enzyme usage, extracts ATP, NADPH, and NADH production balances from flux results, and converts fluxes to original, non-ecModel version.
2022-09-15