LIGSITEcsc is a web server for the automatic identification of pockets on protein surface using Connolly surface and degree of conservation!


Requrement:
To install LIGSITEcsc in your own machine, you need

  1. Linux system with g++ compiler
  2. BALL library
  3. PyMol for visualization (option)


INSTALL:

  1. Download BALL from the BALL website www.ball-project.org/Downloads.
  2. Save BALL-1.1.1.tgz and unpack it: tar zxvf BALL-1.1.1.tgz
  3. change into the source directory: cd BALL-1.1.1/source
  4. run ./configure --help to see the options available
  5. run ./configure --disable-VIEW
  6. make (wait and wait,finally successed)
  7. make install
  8. If you do not get any errors, you have successfully installed the Ball library, which is a requirement for LIGSITEcsc. If you have problems in compiling BALL, please send e-mail to the BALL mailinglist at ball-discuss-l@postino.mpi-sb.mpg.de.

  9. Before running LIGSITEcsc, you have to export the library path of BALL. Add the following two lines to your ~/.bashrc file. Replace PUT THE BALL DIRECTORY HERE by the appropriate directoy path($BALL: the folder where BALL is installed). Replace PUT THE RIGHT DIRECTORY HERE by the folder you find under $BALL/lib/. It is something like Linux-i386-g++_3.3.3 (the last three numbers (3.3.3) will be different depending on your system).

    export BALL=PUT THE BALL DIRECTORY HERE
    export LD_LIBRARY_PATH=$BALL/lib/PUT THE RIGHT DIRECTORY HERE

    To make the updated changes in the bashrc file visible in your shell, type in the shell
    source ~/.bashrc
    To verify that your path settings are correct, type in the shell
    ls $LD_LIBRARY_PATH
    and you should get as output libBALL.so
  10. Now we are ready to progress to the installation of LIGSITEcsc

  11. Download the file "lcs.tgz" and put it to the $BALL directory
  12. Change into the Ball directoy:
    cd $BALL
  13. Unpack the LIGSITE code:
    tar zxvf lcs.tgz
  14. Change into the lcs directory:
    cd $BALL/lcs
  15. Type:
    make lcs



Usage:
lcs -i pdb_file_name [-s grid_space] [-n number_of_pockets] [-t SSS_threshold] [-d surface_density]
Options:
-i: pdb_file_name, the file name of the protein, no default
-s: grid space, default: 1.0 angstrom
-n: number of pockets,default: 3
-t: SSS_threshold, the threshold for SSS event, from 3 to 7, default: 6
-d: surface density, the density (dots/A^2 ) to calculate the surface vertex, default: 0.5

For example, you want to identify the pockets on "1dwd.pdb", just type:
"./lcs -i 1dwd.pdb"
you will get three standard pdb files:
"pocket.pdb", the identified pocket sites in pdb format
"pocket_all.pdb", all the pocket sites in pdb format
"pocket_r.pdb", all the grid points before clustering in pdb format

and one python script file called "pocket.py" to visualize the pocket sites using PyMol.
If you have pymol on your machine, just type:
pymol pocket.py

Note:
Please note that there is no conservation re-ranking in this source code. The pockets are ranked by their sizes. If you want to use conservation re-ranking, please use the webserver. Thanks.

Reference:
Bingding Huang and Michael Schroeder (2006), LIGSITEcsc: Predict protein binding sites using the Connolly surface and degree of conservation, 6:19. link.


Home
Bioinformatics group Biotec TU Dresden