Bio-python

Biopython is a set of freely available tools for biological computation written in Pythonby an international team of developers.

It is a distributed collaborative effort to develop Python libraries and applications which address the needs of current and future work in bioinformatics. The source code is made available under theBiopython License, which is extremely liberal and compatible with almost every license in the world.

This post will help you download and install Biopython, and start using the libraries and tools.


Current Release - 1.52 - 22 September 2009

See also What's new.

Files

Installation Instructions

For Windows we provide click-and-run installers (specific to your version of python), but you will first need to install some prerequisite software (listed below, in particular, NumPy).

Most Linux distributions will include an optional Biopython package (described below), and will take care of any prerequisite software automatically.

Otherwise you typically install from source by downloading and uncompressing the archive, then running the commands:

python setup.py build python setup.py test sudo python setup.py install

If you have trouble, see the full installation instructions:

If you are still stuck, sign up to the Biopython mailing list and ask for help there.

Required Software

  • Python 2.4, 2.5 or 2.6(Biopython 1.50 was the last release of Biopython to support Python 2.3)
  • C compiler (if compiling from source) You need a C compiler supported by distutils, gcc will work fine on UNIX-like platforms. This is not needed on Windows if using the install programs provided above.
  • NumPy (Numerical Python). Note that until BioPython 1.49, Biopython used the older Numeric library. We have tested NumPy 1.0, 1.1, 1.2, and 1.3 with Biopython. Note NumPy 1.2 onwards requires at least Python 2.4.

Optional Software

For compiling Biopython:

Some parts of Biopython use the following additional python libraries:

In addition Biopython includes wrapper code for calling a number of third party command line tools including:

  • Wise2 -- for command line tool dnal
  • NCBI Standalone BLAST -- command line tool for running BLAST on your local machine
  • Clustalw -- command line tool for building sequence alignments
  • SIMCOAL2 and FDist -- command line tools for population genetics
  • EMBOSS -- lots of useful command line tools.

Easy Install

We don't officially sanction this option, but it has been reported to work fine. If you haveeasy_install installed on your computer, you can download and install the latest Biopython distribution by simply executing this command:

easy_install -f http://biopython.org/DIST/ biopython

You will have to have administrator's rights to do this. On a Unix style system this is normally done by:

sudo easy_install -f http://biopython.org/DIST/ biopython

Packages

For those of you using Linux, the easiest way to install Biopython is through your distribution's package management system. However, unless you are running a recent release of your Linux Distribution, you may find that the Biopython packages available to be a little out of date. You might want to see if there is a backport available, otherwise you will have to install Biopython from source.

Ubuntu or Debian

You should be able to install Biopython and its dependencies using the Synaptic GUI tool (on the main menu under System / Administration / Synaptic Package Manager), or at the command line using:

sudo apt-get install python-biopython

If you want the documentation and unit tests,

sudo apt-get install python-biopython-doc

And if you want to use BioSQL,

sudo apt-get install python-biopython-sql

However, this will probably not be the latest release (seeUbuntu listing here, andDebian listing here). If you want the latest version of Biopython, you will need to install it from source. However, you should be able to automatically install the build dependencies with the following command:

sudo apt-get build-dep python-biopython

Note: You may need to additionally install the NumPy package by hand, as an out of date repository may still expect Biopython to use Numeric instead.

Fedora

Biopython is an official Fedora package (since Fedora 5). The package is named python-biopython, and can be installed using yum as root:

yum install python-biopython

or via one of the GUI package management systems such as pirut and PackageKit (available in F-9 and later).

Gentoo Linux

Gentoo's portage tree contains an ebuild (sci-biology/biopython) which builds from source. To install it, open a terminal as root and run:

emerge -va biopython

Here is a link to Biopython atGentoo-Portage which shows the latest versions in Gentoo's Portage tree.

Ports

FreeBSD

The most easy way of installing Biopython in FreeBSD is through the Ports Collection. If you're new to this procedure please take a look at this document. Supposing that you're familiar with this method and that you have an up-to-date ports tree, all you need to do is to execute the following commands as root:

# cd /usr/ports/biology/py-biopython # make install clean

Due to the great architecture of the ports system, this simple commands will automatically fetch and install Biopython (as well as its necessary dependencies).

Old Releases

Recent releases of Biopython require NumPy (and not Numeric):

Please note that Biopython 1.48 and older require the Numeric library, not its replacement NumPy. Windows installers for Python 2.4 and older are available from the Numerical Pythonwebsite. A Windows installer for Numeric 24.2 for Python 2.5 is available here:Numeric-24.2.win32-py2.5.exe446 Kb

Please note that Biopython 1.48 and older used mxTextTools 2.0 in some of the parsers. There were a few niggles with mxTextTools 3.0, so ideally install the older mxTextTools 2.0.
Share/Bookmark
Posted by Amit Vikram Singh at 5:04 PM
Labels: , ,

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)

Powered by MyPagerank.Net

LinkWithin