Install ProtParCon¶

ProtParCon runs only on Python 3.4 or above, if you are already familiar with installation of Python packages, you can easily install ProtParCon and its dependencies from PyPI with the following command on all major platforms (Windows, MacOS, or Linux):

pip install ProtParCon

It is strongly recommended that you install ProtParCon in a dedicated virtualenv, to avoid conflicting with your system packages.

Install ProtParCon to a virtual environment (recommended)¶

We recommend installing ProtParCon inside a virtual environment on all platforms.

Python packages can be installed either globally (a.k.a system wide), or in a user specified space. We do not recommend installing ProtParCon system wide.

Instead, we recommend that you install ProtParCon within a so-called “virtual environment” (virtualenv). Virtualenvs allows users not to conflict with already-installed Python system packages (which could break some of your system tools and scripts), and still install packages normally with pip (without sudo and the likes).

To get started with virtual environments, see virtualenv installation instructions.

Things that are good to know¶

ProtParCon is written in pure Python and depends only on standard libraries and one third party library: Biopython.

Note

NumPy is required by biopython, however, if biopython can be successfully installed as a dependency package of ProtParCon, NumPy should not be a problem.

Although ProtParCon itself is platform independent, the built-in supports for many programs, e.g. MAFFT for sequence alignment, FastTree for phylogenetic tree inference, and Seq-Gen for sequence simulation may not be platform independent, it is strongly recommended that ProtParCon is used on MacOS or Linux platforms.

ProtParCon is designed for providing a easy and common interface for various programs related to phylogenetic analysis and analysis of molecular parallel and convergent amino acid replacements, therefore, users are recommended to use ProtParCon inside their scripts for efficiently calling external programs and building their pipeline by chaining several external programs.