arendma/Crekapp

Proteomics and constraint-based modelling reveal enzyme kinetic properties of C. reinhardtii on a genome-scale

Crekapp

This repository contains code to produce the results presented in the publication "Proteomics and constraint-based modelling reveal enzyme kinetic properties of Chlamydomonas reinhardtii on a genome scale"

Dependencies

• R (tested on version 4.2.1)
• matlab (tested on version 2020b & 2022b)
• COBRA toolbox (tested on 2022 release)
• [RAVEN] (https://github.com/SysBioChalmers/RAVEN) (tested on version 2.5.0)
• GECKO toolbox (tested on version 2.0.2)
• Gurobi solver (tested on version 9.5.2)

Setup (~ 30min)

After making sure all dependencies are installed and properly setup (see links above). Switch working directory to this git repository and open an R console and from within run source("Rsetup.r") to install nescessary R packages. Edit the Matlab_startup.m file to add the paths of the installed dependencies.

Reproducing results (on linux machine - scripts produce verbose output to comman line no reason to worry)

1. Set this repository as working directory

2. Run Rscript Program/fit_chemostatdat.r to obtain a model for maximum acetate uptake. (On windows it is sometimes only possible to source the scripts from within the R console using source("Program/fit_chemostat.r"))(~2-3s on Ryzen5 4000 16 GB RAM)

3. Run Rscript Program/QCC_smy_main2.r to process the raw QCC data into and generate plots of QconCat proteomics overview statistics. (Also here on windows alternatively you can use source("Program/QCC_smy_main2.r"))(~90s on Ryzen5 4000 16 GB RAM)

4. In Matlab run Matlab_startup to set path for depenencies (edit as mentioned above) (1s on Ryzen5 4000 16 GB RAM)

5. In Matlab run GECKO_startup to generate pcGEMs from autotrophic, mixotrohpic and heterotrophic chlamydomonas models. This will create a log file with GECKO output in the working directory. (~3h on Ryzen5 4000 16 GB RAM)

6. In Matlab run iCreNIDLE to estimate kapps with NIDLE and generate obtain comparison values for pFBA and BRENDA/SABIORK (~10 min on Ryzen 5 4000 16 GB RAM)

7. In Matlab run comp_ecModel_rescale to generate metabolic model predictions from the GEM and pcGEMs. (~2m on Ryzen5 4000 16 GB RAM)

8. Run Rscript Program/manuscript_fig.r to generate figures and statistics presented in the paper. (60s on Ryzen5 4000 16 GB RAM)