GraphNet is a principled approach for using gene expression profiles to improve predictions of metabolic reaction rates. This new computational framework formulates probabilistic signatures by incorporating gene expression profiles in to the regulatory network linking with genomic-scale metabolic models. With the framework, cellular growth rates influenced by deletion of a regulatory gene or in combination of regulatory genes can be predicted in silico.

GraphNet Guide
Installation requirements
  • GraphNet is a Python-package and therefore the installation of Python is required
  • Install Ubuntu
  • Install gurobi, including the Python extension
  • Download python-bpfba and run './'
  • Done, now you can run the scripts.
  • In order to run the algorithm it is just needed to provide a genome regulatory network of the studied organism and a set of gene expression profiles as .txt files (tab delimited text table) along with a GEM in .xml format (SBML format). These files should be stored in the folder named “data” and should be called in the Python file as:
    Run: run_fendt(4.0, 6.0, 'data/gene-expression-profiles.txt', 'data/regulatory-network.txt', 'data/GEM.xml')
  • To run the algorithm, run Python in the command prompt; go to the folder where the algorithm and data are stored and type:
  • Download

    Latest version

    RAVEN Github page

    Pre trained HMMs:      

    Older versions: