.. include:: global.rst

.. _installation:


Installation
============

The easiest way to install *ipyrad* and all of its dependencies is with conda_,
a command line program for installing Python packages. If you do not have *conda* installed, follow these instructions_ to install either *Anaconda* or *Miniconda* 
for Python2.7. If you're working on an :ref:`HPC <HPC_installation>` system you can install *conda* in your home directory without needing administrative privileges by following the same basic directions. 

The only difference between *Anaconda* and *Miniconda* is that *Anaconda* 
installs a large suite of commonly used Python packages along with the base
installer, whereas *Miniconda* is a bare bones version that includes only 
the framework for installing new packages. Unless you're really hard
up for disk space I recommend installing *Anaconda*. 

To install *ipyrad* using *conda* simply type the following into a terminal ::

    $ conda update conda            ## updates conda 
    $ conda install ipyrad          ## installs the latest release

If you wish to install a specific version of ipyrad, or to upgrade to the 
latest release from an older version, you could use one of the following commands::

    $ conda install ipyrad=0.7.0    ## install ipyrad v.0.7.0
    $ conda update ipyrad           ## update to the latest


Dependencies
------------
conda will install all of the following required dependencies during the
installation of ipyrad. 

Python Packages:  
* Numpy -- Scientific processing  
* Scipy -- Scientific processing  
* Pandas -- Used for manipulating data frames  
* Sphinx -- Used for building documentation  
* IPython2 -- Interactive version of Python 2.7
* ipyparallel -- Parallel, threading, MPI support
* jupyter -- Creating reproducible notebooks  
* Cython -- C bindings for Python 
* H5py -- Database and HDF5 headers
* Dill -- Store pickle objects of complex Classes  
* Toyplot -- [optional].

Executables:  
* vsearch
* muscle
* smalt
* samtools


.. _HPC_installation:

HPC installation
----------------
One of the benefits of using *conda* for installation is that it 
creates a Python package directory in your home directory called 
``~/anaconda/`` (or ``~/miniconda/``) where new packages are installed.
Make sure you follow the installation instructions_ so that Python
scripts will look in this directory by default. Because these Python packages 
are not stored in a system-wide directory you will not need administrator 
privileges to install new packages, nor will you have to load these modules 
from the system before using them.