Difference between revisions of "Quick installation instructions for STAMP"

From Bioinformatics Software
Jump to navigationJump to search
 
(90 intermediate revisions by the same user not shown)
Line 3: Line 3:
 
== Binary release ==
 
== Binary release ==
  
Precompiled binaries are available for [[Media:STAMP_1.01_Setup.exe‎|Microsoft Windows]] and [[Media:STAMP_1_01_Mac_OSX.zip|Mac OS X]]. The Windows binary has been tested under Windows XP SP3 and Windows 7, but will likely also run under Windows Vista. The OS X binary has been tested under Leopard and Snow Leopard. Please note that the first time your run STAMP it may take a minute or more to load. After this, it will start up far more quickly. If either of these binaries fail to work for you please let us know. 
+
An [https://github.com/dparks1134/STAMP/releases/download/v2.1.3/STAMP_2_1_3.exe executable binary] is available for Microsoft Windows. It has been tested under Windows XP SP3 and Windows 7, but will likely also run under Windows Vista. The Windows binary can also be run on Linux or OS X using [http://www.virtualbox.org/ Virtual Box]. Due to differences in how the user interface appears under Windows, Linux and OS X we recommended using Virtual Box to run STAMP. Please note that the first time your run STAMP it may take a minute or more to load. After this, it will start up more quickly.  
  
 
Note: if you have a pristine copy of Microsoft Windows installed, you may need to install the Microsoft Visual C++ 2008 Redistributable Package ([http://www.microsoft.com/downloads/details.aspx?FamilyID=9b2da534-3e03-4391-8a4d-074b9f2bc1bf&displaylang=en 32-bit],[http://www.microsoft.com/downloads/details.aspx?familyid=bd2a6171-e2d6-4230-b809-9a8d7548c1b6&displaylang=en 64-bit]). This package contains a number of commonly required runtime components and as such you likely already have them. STAMP will fail with a message indicating the "configuration is incorrect" if you require this package.
 
Note: if you have a pristine copy of Microsoft Windows installed, you may need to install the Microsoft Visual C++ 2008 Redistributable Package ([http://www.microsoft.com/downloads/details.aspx?FamilyID=9b2da534-3e03-4391-8a4d-074b9f2bc1bf&displaylang=en 32-bit],[http://www.microsoft.com/downloads/details.aspx?familyid=bd2a6171-e2d6-4230-b809-9a8d7548c1b6&displaylang=en 64-bit]). This package contains a number of commonly required runtime components and as such you likely already have them. STAMP will fail with a message indicating the "configuration is incorrect" if you require this package.
Line 13: Line 13:
 
* [http://www.python.org/download/ Python 2.5 or 2.6] (we recommend Python 2.6.4)
 
* [http://www.python.org/download/ Python 2.5 or 2.6] (we recommend Python 2.6.4)
 
* [http://www.riverbankcomputing.co.uk/software/pyqt/download PyQt4]
 
* [http://www.riverbankcomputing.co.uk/software/pyqt/download PyQt4]
* [http://www.scipy.org/ NumPy and SciPy] (Note: NumPy 1.4.0rc1 is not stable, we recommend using NumPy 1.3.0)
+
* [http://www.scipy.org/ NumPy and SciPy]  
* [http://code.google.com/p/mpmath/ mpmath]
 
 
* [http://matplotlib.sourceforge.net/ matplotlib]
 
* [http://matplotlib.sourceforge.net/ matplotlib]
 +
* [http://biom-format.org/ biom-format]
 +
* [https://pypi.python.org/pypi/pyqi/ pyqi]
  
After you have installed Python and the above libraries, extract the [[Media:STAMP_1.0_source.zip|STAMP source code files]] to a directory (e.g., C:\STAMP). You can now run STAMP by moving to the source directory and entering <tt>'python STAMP.py'</tt> on your command line. If you get an error message indicating Python cannot be found you likely need to add it to your system path (check out this [http://docs.python.org/using/windows.html#excursus-setting-environment-variables website] for instructions). If you have trouble installing STAMP, please feel free to contact us.
+
After you have installed Python and the above libraries, extract the [https://github.com/dparks1134/STAMP/archive/v2.1.3.zip STAMP source code files] to a directory (e.g., C:\STAMP). You can now run STAMP by moving to the source directory and entering <tt>'python ./bin/STAMP'</tt> on your command line. If you get an error message indicating Python cannot be found you likely need to add it to your system path (check out this [http://docs.python.org/using/windows.html#excursus-setting-environment-variables website] for instructions). If you have trouble installing STAMP, please feel free to contact us.
  
== Source code on Apple's OS X ==
+
== Source code on Linux ==
  
Running STAMP from source on Apple's OS X is straight-forward. It requires minimal effort on your part, although it does require a lot of compiling to be done by your computer. There are many ways to get STAMP running from source, but we have found the following the simplest (if you have a better way, please let us know):
+
STAMP can be install on Linux (tested under Ubuntu 14.04) using the following commands:
 +
  sudo apt-get install libblas-dev liblapack-dev gfortran
 +
  sudo apt-get install freetype* python-pip python-dev python-numpy python-scipy python-matplotlib
 +
  sudo pip install STAMP
  
* Install [http://darwinports.com/ DarwinPorts]
+
This installs STAMP for use with the default version of Python on your system. You can now run STAMP by entering <tt>'STAMP'</tt> on your command line.
* '''Update DarwinPorts''' by typing the following into a console:
 
  sudo port -d selfupdate
 
  
From your console, install the necessary dependencies for STAMP using port:
+
Note: STAMP requires [http://www.riverbankcomputing.co.uk/software/pyqt/intro PyQt4]. This is install by default on Ubuntu, but may need to be installed independently for different flavours of Linux.
  sudo port install python26
 
  sudo port install python_select
 
  sudo python_select python26
 
  sudo port install py26-macholib
 
  sudo port install py26-sip
 
  sudo port install py26-pyqt4
 
  sudo port install py26-numpy
 
  sudo port install py26-scipy
 
  sudo port install py26-mpmath
 
  sudo port install py26-matplotlib
 
  
Our first attempt to install py26-pyqt4 failed, but was successful on our second attempt. Here are some rough estimates on the time required to install each of these package: python26 = 1 hour, python_select = 30 seconds, py26-macholib = 1 min, py26-sip = 10 min, py26-pyqt4 = 3 hours, py26-numpy = 3 hours, py26-scipy = 15 min, py26-mpmath = 30 seconds, py26-matplotlib = 5 min.
+
== Source code on Apple OS X ==
  
To test that all dependencies where installed correctly, start a python session and import each package:
+
Installation of STAMP on OS X is being driven by community feedback. Please let me know how your experience with installing STAMP of OS X goes so I can update these instructions.
  import PyQt4
 
  import numpy
 
  import scipy
 
  import mpmath
 
  import matplotlib
 
  
After you have installed Python and the above libraries, extract the [[Media:STAMP_1.0_source.zip|STAMP source code files]] to a directory (e.g., ~\STAMP). You can now run STAMP by moving to the source directory and entering <tt>'python STAMP.py'</tt> on your command line.  If you have trouble installing STAMP, please feel free to contact us.
+
Our thanks to Tom Battaglia for suggesting the following approach:
 
+
  1. Install Xcode/Command line Tools
== Command-line interface install ==
+
  2. Install the [http://continuum.io/downloads Anaconda] package for Python v2.7
 
+
  3. Install PIP (if not already on your system): sudo easy_install pip (skip if you already have pip installed)
If you wish to use STAMP strictly from the command-line (e.g., as typical of a cluster environment) only the following 3rd-party dependencies are required:
+
  4. Install numpy: pip install numpy
+
  5. Install STAMP: pip install STAMP
* [http://www.python.org/download/ Python 2.5 or 2.6] (we recommend Python 2.6.4)
+
  5. Open new terminal session, enter STAMP to run.
* [http://www.scipy.org/ NumPy and SciPy] (Note: NumPy 1.4.0rc1 is not stable, we recommend using NumPy 1.3.0)
 
* [http://code.google.com/p/mpmath/ mpmath]
 
 
 
For details on running STAMP from the command-line please see the [[Media:STAMP_Users_Guide.pdf|User's Guide]].
 
  
 
== Our build environment ==
 
== Our build environment ==
  
STAMP was developed on Microsoft Windows XP and ported to Apple's Mac OS X. [http://www.py2exe.org/ py2exe] was used to create the Windows binary. It is generally best to download the latest version of each dependency. Our build environment currently uses the following versions of each dependency:
+
STAMP was developed on Microsoft Windows XP and ported to Apple's OS X. [http://www.py2exe.org/ py2exe] was used to create the Windows binary. It is generally best to download the latest version of each dependency. Our build environment currently uses the following versions of each dependency:
  
* Python 2.6.4
+
* Python 2.7.3
* PyQt4 4.6.2
+
* PyQt4 4.8.4
* NumPy 1.3.0
+
* NumPy 1.9.1
* SciPy 0.7.1
+
* SciPy 0.15.1
* mpmath 0.13
+
* matplotlib 1.4.2
* matplotlib 0.99.1.1
+
* biom-format 2.0.1
 +
* pyqi 0.3.2
 +
* six 1.8.0
 +
* h5py 2.3.0 (optional)
 
* py2exe 0.6.9
 
* py2exe 0.6.9

Latest revision as of 19:31, 25 February 2016

Binary release

An executable binary is available for Microsoft Windows. It has been tested under Windows XP SP3 and Windows 7, but will likely also run under Windows Vista. The Windows binary can also be run on Linux or OS X using Virtual Box. Due to differences in how the user interface appears under Windows, Linux and OS X we recommended using Virtual Box to run STAMP. Please note that the first time your run STAMP it may take a minute or more to load. After this, it will start up more quickly.

Note: if you have a pristine copy of Microsoft Windows installed, you may need to install the Microsoft Visual C++ 2008 Redistributable Package (32-bit,64-bit). This package contains a number of commonly required runtime components and as such you likely already have them. STAMP will fail with a message indicating the "configuration is incorrect" if you require this package.

Source code on Microsoft Windows

STAMP is implemented in Python and running it from source is easy. Any operating system that supports Python should be able to run STAMP from source. Running from source is the best way to fully exploit and contribute to STAMP. STAMP is dependent on the following libraries:

After you have installed Python and the above libraries, extract the STAMP source code files to a directory (e.g., C:\STAMP). You can now run STAMP by moving to the source directory and entering 'python ./bin/STAMP' on your command line. If you get an error message indicating Python cannot be found you likely need to add it to your system path (check out this website for instructions). If you have trouble installing STAMP, please feel free to contact us.

Source code on Linux

STAMP can be install on Linux (tested under Ubuntu 14.04) using the following commands:

 sudo apt-get install libblas-dev liblapack-dev gfortran
 sudo apt-get install freetype* python-pip python-dev python-numpy python-scipy python-matplotlib
 sudo pip install STAMP

This installs STAMP for use with the default version of Python on your system. You can now run STAMP by entering 'STAMP' on your command line.

Note: STAMP requires PyQt4. This is install by default on Ubuntu, but may need to be installed independently for different flavours of Linux.

Source code on Apple OS X

Installation of STAMP on OS X is being driven by community feedback. Please let me know how your experience with installing STAMP of OS X goes so I can update these instructions.

Our thanks to Tom Battaglia for suggesting the following approach:

 1. Install Xcode/Command line Tools
 2. Install the Anaconda package for Python v2.7
 3. Install PIP (if not already on your system): sudo easy_install pip (skip if you already have pip installed)
 4. Install numpy: pip install numpy
 5. Install STAMP: pip install STAMP
 5. Open new terminal session, enter STAMP to run.

Our build environment

STAMP was developed on Microsoft Windows XP and ported to Apple's OS X. py2exe was used to create the Windows binary. It is generally best to download the latest version of each dependency. Our build environment currently uses the following versions of each dependency:

  • Python 2.7.3
  • PyQt4 4.8.4
  • NumPy 1.9.1
  • SciPy 0.15.1
  • matplotlib 1.4.2
  • biom-format 2.0.1
  • pyqi 0.3.2
  • six 1.8.0
  • h5py 2.3.0 (optional)
  • py2exe 0.6.9