Skip to content

Latest commit

 

History

History
115 lines (85 loc) · 4.06 KB

INSTALL.rst

File metadata and controls

115 lines (85 loc) · 4.06 KB
Raw

Installation of cobrapy

For installation help, please use the Google Group. For usage instructions, please see the documentation.

All releases require Python 2.7+ or 3.4+ to be installed before proceeding. Mac OS X (10.7+) and Ubuntu ship with Python. Windows users without python can download and install python from the python website. Please note that though Anaconda and other python distributions may work with cobrapy, they are not explicitly supported (yet!).

Stable version installation

cobrapy can be installed with any recent installation of pip. Instructions for several operating systems are below:

Mac OS X or Linux

  1. install pip.
  2. In a terminal, run sudo pip install cobra

We highly recommend updating pip beforehand (pip install pip --upgrade).

Microsoft Windows

The preferred installation method on Windows is also to use pip. The latest Windows installers for Python 2.7 and 3.4 include pip, so if you use those you will already have pip.

  1. In a terminal, run C:\Python27\Scripts\pip.exe install cobra (you may need to adjust the path accordingly).

To install without pip, you will need to download and use the appropriate installer for your version of python from the python package index.

Installation for development

Get the detailed contribution instructions for contributing to cobrapy.

Installation of optional dependencies

Optional Dependencies

On windows, these can downloaded from [this site] (http://www.lfd.uci.edu/~gohlke/pythonlibs/). On Mac/Linux, they can be installed using pip, or from the OS package manager (e.g brew, apt, yum).

  1. libsbml >= 5.10 to read/write SBML level 2 files
  2. lxml to speed up read/write of SBML level 3 files.
  3. numpy >= 1.6.1 for double deletions
  4. scipy >= 0.11 for ArrayBasedModel and saving to *.mat files.

Other solvers

cobrapy comes with bindings to the GNU Linear Programming Kit ([glpk] (http://www.gnu.org/software/glpk/)) using its own bindings called "cglpk" in cobrapy. In addition, cobrapy currently supports these linear programming solvers:

ILOG/CPLEX, MOSEK, and Gurobi are commercial software packages that currently provide free licenses for academics and support both linear and quadratic programming. GLPK and clp are open source linear programming solvers; however, they may not be as robust as the commercial solvers for mixed-integer and quadratic programming. QSopt_ex esolver is also open source, and can solve linear programs using rational operations, giving exact solutions.

Testing your installation

  1. Start python
  2. Type the following into the Python shell
from cobra.test import test_all
test_all()

You should see some skipped tests and expected failures, and the function should return False.