Example of the usage of the program.
PH domains



Running the usage example

Testing similarity calculations with already computed interaction fields

How to prepare pdb/pqr files
 



9 protein models of PH domains are compared on the basis of their electrostatic potentials.  This is a subset of the 104 PH domains analyzed in the paper 1.  PIPSA performs an automated comparison, from which it is clear that there are 2 clusters, one formed by proteins having a mostly positive potential, and the other formed by proteins having negative potentials.  In this case, this classification could be done by simply looking at electrostatic potential contours.  However, PIPSA does the classification in an automated, objective and quantitative fashion, and allows more subtle classification, as found amongst the full 104 domain dataset 1. Indeed, PIPSA shows that for this 9 protein model example, the cluster of proteins having positive potentials can be further divided into two subclusters.



Running the usage example:

Given that all programs to use in conjunction with PIPSA are available, running the usage example can be done by 1 meta script using PIPSA's low level scripts.  For this example and for the local installations in my own computer this script looks like this.  You need to adjust your local assignments and skip the calculations you can not do due to absence of the programs  to use in conjunction with PIPSA.  Below is the detailed description of all PIPSA calculation steps.

-1.   Uncompress and untar pipsa distribtion package ( gunzip -c pipsa.tar.gz  | tar xvf - ). This will give you pipsa distribution directory pipsa/ .  Chdir here and note the directory name (using pwd) - this directory will be mentioned below as $pipsa_distr_dir, i.e. all occurencies of  $pipsa_distr_dir below should be replaced by this directory path.  (Under Linux, you can define an environmental variable pipsa_distr_dir as the path to the distribution directory, then the command lines below can be typed in (or pasted) as they appear.)

Note.    Executables in the distribution were compiled under Linix (see the file Makefile in src/ subirectory) and you might need to recompile them in order to adjust to your OS (see Installation instructions) .

0.  Unzip the files needed to run the usage example from $pipsa_distr_dir/exa/PH.zip to any directory.  This will give you the subdirectory PH.  Chdir there and note the directory name (using pwd) - this directory will be mentioned below as $pipsa_wrk_dir, i.e. all occurencies of  $pipsa_wrk_dir below should be replaced by this directory path).  This directory has the pdb files of proteins and the listings of pdb file names, needed for running PIPSA scripts.

1.  Type (from any directory)
$pipsa_distr_dir/scr/do_pipsa_analyt_sim     $pipsa_distr_dir/       $pipsa_wrk_dir/pdbs/
This will generate similarity matrix, distance matrix and its visualisation as a kenemage in the directory $pipsa_wrk_dir/pdbs/ .

2.1.  If you have UHBD installed and its executable with its path is $path_to_uhbd_exe then you can compute UHBD grids and perform similarity analysis using electrostatic potential computed by UHBD running the following:
$pipsa_distr_dir/scr/do_pipsa_UHBD_prep      $pipsa_distr_dir/    $pipsa_wrk_dir/
$pipsa_distr_dir/scr/do_pipsa_UHBD_grids     $pipsa_distr_dir/     $pipsa_wrk_dir/uhbd/      $path_to_uhbd_exe
$pipsa_distr_dir/scr/do_pipsa_sim                  $pipsa_distr_dir/     $pipsa_wrk_dir/uhbd/

2.2.  If you have APBS installed and its executable with its path is $path_to_apbs_exe then you can compute APBS grids and perform similarity analysis using electrostatic potential computed by APBS running the following:
$pipsa_distr_dir/scr/do_pipsa_APBS_prep     $pipsa_distr_dir/     $pipsa_wrk_dir/
$pipsa_distr_dir/scr/do_pipsa_APBS_grids     $pipsa_distr_dir/     $pipsa_wrk_dir/apbs/     $path_to_apbs_exe
$pipsa_distr_dir/scr/do_pipsa_sim                  $pipsa_distr_dir/     $pipsa_wrk_dir/apbs/

2.3.  If you have GRID installed and its executables grin, grid and k2a can be found in $path_to_grid_bin then you can compute GRID grids and perform similarity analysis using interaction field computed by GRID running the following:
$pipsa_distr_dir/scr/do_pipsa_GRID_prep    $pipsa_distr_dir/    $pipsa_wrk_dir/
$pipsa_distr_dir/scr/do_pipsa_GRID_grids    $pipsa_distr_dir/    $pipsa_wrk_dir/grid/    $path_to_grid_bin/
$pipsa_distr_dir/scr/do_pipsa_sim               $pipsa_distr_dir/    $pipsa_wrk_dir/grid/

The result of these steps will be the similarity matrix (sims.log), distance matrix (sims.mat) based on (dis)similarity and kinemage representation of distribution of proteins in 3D (sims.kin)
$pipsa_wrk_dir/pdbs/sims.log, sims.mat, sims.kin
$pipsa_wrk_dir/uhbd/sims.log, sims.mat, sims.kin
$pipsa_wrk_dir/apbs/sims.log, sims.mat, sims.kin
$pipsa_wrk_dir/grid/sims.log, sims.mat, sims.kin

Compare these with precomputed results in pipsa distribution:

$pipsa_distr_dir/exa/PHR/pdbs/sims.log, sims.mat, sims.kin
$pipsa_distr_dir/exa/PHR/uhbd/sims.log, sims.mat, sims.kin
$pipsa_distr_dir/exa/PHR/apbs/sims.log, sims.mat, sims.kin
$pipsa_distr_dir/exa/PHR/grid/sims.log, sims.mat, sims.kin

If you have Mage installed, you can visulaize representation of how the proteins can be distributed in three dimensions on the basis of their interaction potential similarity opening by Mage the files sims.kin

If you have NMRCLUST installed, you can cluster proteins on the basis of their interaction potential similarity using the files sims.mat .  For that, after executing nmrclust, answer "no" to the question "Use a PDB file for input?", and enter sims.mat as a Matrix filename.

3. If you have Phylip package installed and its executables can be found in $path_to_phylip_bin then you can generate phylogenic trees from interaction field similarities by executing one of the followings:

$pipsa_distr_dir/scr/do_pipsa_phylip    $pipsa_distr_dir/    $pipsa_wrk_dir/pdbs/    $path_to_phylip_bin
or
$pipsa_distr_dir/scr/do_pipsa_phylip    $pipsa_distr_dir/    $pipsa_wrk_dir/uhbd/    $path_to_phylip_bin
or
$pipsa_distr_dir/scr/do_pipsa_phylip    $pipsa_distr_dir/    $pipsa_wrk_dir/apbs/    $path_to_phylip_bin
or
$pipsa_distr_dir/scr/do_pipsa_phylip    $pipsa_distr_dir/    $pipsa_wrk_dir/grid/      $path_to_phylip_bin

The results of this step are phylogenic tree or diagram in the files (for example)
$pipsa_distr_dir/exa/PHR/uhbd/phylip_tree.ps
$pipsa_distr_dir/exa/PHR/uhbd/phylip_gram.ps
 



Testing similarity calculations with already computed interaction fields

If you do not have any of UHBD, APBS or GRID programs, you can test similarity calculations by pipsa downloading interaction potential grid files for this example.  Note that the binary grid files are generated under Linux and may be not readable under other OS-s.

Download tarfile with grids and pdbs here (30 MB file) and untar it - the subdirectory PHR will appear and will contain grid and pdb files needed for similarity calculation.  Chdir there and note the directory name (using pwd) - this directory will be mentioned below as $pipsa_wrk2_dir, i.e. all occurencies of  $pipsa_wrk2_dir below should be replaced by this directory path).
Run :
$pipsa_distr_dir/scr/do_pipsa_sim     $pipsa_distr_dir/     $pipsa_wrk2_dir/uhbd/
$pipsa_distr_dir/scr/do_pipsa_sim     $pipsa_distr_dir/     $pipsa_wrk2_dir/apbs/
$pipsa_distr_dir/scr/do_pipsa_sim     $pipsa_distr_dir/     $pipsa_wrk2_dir/grid/

The results of pipsa similarity analysis can now be found in the files sims.log, sims.mat and sims.kin in pipsa_wrk2_dir/uhbd, apbs, or grid.



How to prepare pdb/pqr files

pdb and pqr files for this example are already prepared - subdirectory pdbs/ has UHBD readable pdb files, gpdbs/ - GRID radable pdb files and pqs/ - APBS readable pqr files.  Original pdb files without any hydrogens are in subdirectory opdbs/ and, in principle, only they are needed.

1.   If you have WHATIF installed and path to its executable is $path_to_whatif_exe, then you can generate UHBD readable files using the script do_pipsa_WHATIF :
$pipsa_distr_dir/scr/do_pipsa_WHATIF    $pipsa_distr_dir/     $pipsa_wrk_dir/     $path_to_whatif_exe
this will make a directory $pipsa_wrk_dir/whatif , generate and put there UHBD readable pdb files.
N.B. This script was tested on Whatif version 19991018-1516 and should work with compatible versions of Whatif, working with gromos 87 / opls polar hydrogens.  It won't work with Whatif (version 5) generating gromos 96 polar hydrogens.
2.  PQR files for APBS run can be optained from PDB2PQR server .  For this example these files were generated from uhbd output files running scr/do_pipsa_UHBD_chk and aux/do_pipsa_uho2pqr .


[Back to Index]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

     
Imprint/Privacy