Legacy tutorials#

Five hands-on tutorials were part of the original RAVEN 1 paper (Agren et al., 2013). The scripts run with the current RAVEN version. The exercises are otherwise unchanged from the original. They use the MATLAB toolbox and build up from running a simulation on an existing model to reconstructing a genome-scale model from sequence data. The scripts and all the data files they use are in the tutorials/ folder of the RAVEN repository.

#

Tutorial

Content

1

Importing a GEM and running FBA

Load a model, set constraints and an objective, run FBA, visualise fluxes

2

Constructing a functional small model

Build a model from scratch; exchange reactions and the steady-state assumption

3

Knockouts and omics data

Gene deletions, and using a GEM as a scaffold for microarray data

4

Fixing an erroneous model

Systematic quality control: find and fix mass-balance and naming errors

5

Reconstructing a GEM from KEGG

De novo reconstruction from protein sequences using KEGG

Prerequisites#

  • The tutorials use the MATLAB toolbox. RAVEN must be installed and checkRaven must pass; see Installation.

  • Tutorial 2 loads its model with importModel (SBML), and Tutorials 3 and 4 with readYAMLmodel (YAML). curateModelFromTables curates an existing model from .tsv files; see Excel I/O.

  • To run a section of a script in MATLAB, highlight it, right-click, and choose "Evaluate selection".

  • Tutorials 2, 3 and 4 ship with a *_solutions.m companion script containing the completed exercise.

The reconstruction concepts carry over to raven-toolbox. The snake_case form of a RAVEN function name matches the raven-toolbox function name for some functions. Some RAVEN functions map to cobrapy (solveLP → model.optimize()), and some have no Python counterpart (there is no Excel reader). The API reference lists the functions, and Comparison of RAVEN and raven-toolbox shows what maps where.

On this page