Install SEA: Difference between revisions

From DISI
Jump to navigation Jump to search
No edit summary
No edit summary
 
(16 intermediate revisions by 4 users not shown)
Line 1: Line 1:
This is indened for Shoichet lab setup.  But we hope other find it useful.
This is intended for Shoichet lab setup.  But we hope other find it useful. See [[SEA]]
 
 
The full SEA installation requires a few Python packages:
 
* [http://numpy.scipy.org/ NumPy]
* [http://scipy.org/ SciPy]
* [http://math-atlas.sourceforge.net/ ATLAS]
 
These are best installed using your distribution's package manager:
 
    yum install scipy numpy atlas atlas-devel mysql mysql-server
 
On CentOS systems you may need to add an additional repository to make all of these packages available:
 
    wget http://dl.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-7.noarch.rpm
    sudo rpm -Uvh epel-release-6-7.noarch.rpm
    yum update
    yum install scipy numpy numpy-f2py atlas atlas-devel mysql mysql-devel mysql-server
 
The additional web-based results viewer requires:
 
* Apache
* PHP
* PHP MySQL Libraries
 
    sudo yum install php php-mysql php-cli
 
You will need to checkout the most recent version of the SEA source code. From within the BKS network this can be done by:
 
    svn co file:///raid4/svn/sea/trunk /tmp/sea
    svn co file:///raid4/svn/sea.molecule/trunk /tmp/sea
    svn co file:///raid4/svn/tanimoto/trunk /tmp/sea
 
Users outside of the network should see the [http://sea.bkslab.org/ SEA website] for source code availability.
 
Notes about checkout:
 
* If you will be installing SEA into the system's Python site-packages the source code will need to be checked out into a location with write access for both the active user account '''and''' the local root user, hence `/tmp/sea`.
* It is important that you are using the version of python into which you NumPy and SciPy have been added and which you intent to install SEA to. For example if you wish to install SEA to the local python instance `which python` should return `/usr/bin/python`. If it does not, the easiest way to remidy this is `alias python /usr/bin/python`. This can be the source of many headaches.
 
After checking out and installing these, you'll need to set up you public_html to view the SEA runs results on the web viewer (in your home directory):
 
    mkdir public_html
    cp -r /usr/arch/share/sea/www/ public_html/sea
    cp ~keiser/.sea.cfg . 
    cp ~keiser/.sea_databases.cfg .
    cp ~keiser/.sea_molecule* .
    rm .sea_molecule.cfg
 
    # Change .sea_databases.cfg file
      >> SEA database login information
      >> [scratch]
      >> host=scratch.bkslab.org
      >> user=your_username
      >> pass=your_password
      >> db=your database
 
You will also need to create the MySQL database:
    sea-database
        create
 
The recommended database settings are in /etc/my.cnf
 
Note that there is a useful manunal distributed with the source code located in `${seapath}/sea/data/doc/manual.html`. 
 




[[Category:SEA]]
[[Category:SEA]]
[[Category:Scaffold hopping]]
[[Category:Off target effects]]
[[Category:Off target effects]]
[[Category:Pharmacology]]
[[Category:Pharmacology]]
[[Category:Sysadmin]]

Latest revision as of 06:42, 13 March 2014

This is intended for Shoichet lab setup. But we hope other find it useful. See SEA


The full SEA installation requires a few Python packages:

These are best installed using your distribution's package manager:

    yum install scipy numpy atlas atlas-devel mysql mysql-server

On CentOS systems you may need to add an additional repository to make all of these packages available:

   wget http://dl.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-7.noarch.rpm
   sudo rpm -Uvh epel-release-6-7.noarch.rpm
   yum update
   yum install scipy numpy numpy-f2py atlas atlas-devel mysql mysql-devel mysql-server

The additional web-based results viewer requires:

  • Apache
  • PHP
  • PHP MySQL Libraries
   sudo yum install php php-mysql php-cli

You will need to checkout the most recent version of the SEA source code. From within the BKS network this can be done by:

   svn co file:///raid4/svn/sea/trunk /tmp/sea
   svn co file:///raid4/svn/sea.molecule/trunk /tmp/sea
   svn co file:///raid4/svn/tanimoto/trunk /tmp/sea

Users outside of the network should see the SEA website for source code availability.

Notes about checkout:

  • If you will be installing SEA into the system's Python site-packages the source code will need to be checked out into a location with write access for both the active user account and the local root user, hence `/tmp/sea`.
  • It is important that you are using the version of python into which you NumPy and SciPy have been added and which you intent to install SEA to. For example if you wish to install SEA to the local python instance `which python` should return `/usr/bin/python`. If it does not, the easiest way to remidy this is `alias python /usr/bin/python`. This can be the source of many headaches.

After checking out and installing these, you'll need to set up you public_html to view the SEA runs results on the web viewer (in your home directory):

   mkdir public_html
   cp -r /usr/arch/share/sea/www/ public_html/sea
   cp ~keiser/.sea.cfg .   
   cp ~keiser/.sea_databases.cfg .
   cp ~keiser/.sea_molecule* .
   rm .sea_molecule.cfg
   # Change .sea_databases.cfg file
     >> SEA database login information
     >> [scratch]
     >> host=scratch.bkslab.org
     >> user=your_username
     >> pass=your_password
     >> db=your database

You will also need to create the MySQL database:

   sea-database
       create

The recommended database settings are in /etc/my.cnf

Note that there is a useful manunal distributed with the source code located in `${seapath}/sea/data/doc/manual.html`.