Tutorial on running Molecular Dynamics for GIST grid generation

Revision as of 18:34, 21 January 2017 by TBalius (Talk | contribs)

Jump to: navigation, search

Tutorial written by Trent Balius (Jan. 9, 2017).


This is foremost for training Shoichet lab members. But we hope that the community finds this useful.

To use this tutorial you will need the following:

  • AMBER12 or higher.
  • GPUs to run AMBER on. In our case we uses an SGE queuing system to manage these jobs.
  • It is helpful if you have DOCK3.7 the latest version.


Grid inhomogeneous Solvation Theory (GIST) is a method for calculating thermodynamic properties of water on GIST lattice.

Here, we will use these grids for DOCKing (although, there are many other uses).

Set up environment

you can add the following to your environment file (.cshrc) for just issue them on the command line.

source python with bio_python package.

source /nfs/soft/python/envs/complete/latest/env.csh


setenv DOCKBASE "/nfs/home/tbalius/zzz.github/DOCK"

Put msms in your path

set path = ( /nfs/home/tbalius/zzz.programs/msms $path )

Set up directories

cd /nfs/work/yourusername/
mkdir ambergisttutorial
cd ambergisttutorial 

Prepare for AMBER

  • download 4NVE form the PDB brake it into receptor and ligand pieces.
  • download pdb 4NVA form the PDB align it to 4NVE and out put waters that are close to the ligand.
  • here is one way to do it:

(1) Break up the protein from the ligand. run be_blasti.py

 $DOCKBASE/proteins/pdb_breaker/be_blasti.py --pdbcode 4NVE nocarbohydrate original_numbers | tee -a pdbinfo_using_biopython.log
 $DOCKBASE/proteins/pdb_breaker/be_blasti.py --pdbcode 4NVA nocarbohydrate original_numbers | tee -a pdbinfo_using_biopython.log
 cd 4NVE 
 sed -e "s/HETATM/ATOM  /g" lig.pdb > xtal-lig.pdb
 cd ../
 cd 4NVA
 grep HOH $mountdir/workingdir/4NVA/4NVA_A.pdb > $workdir/water.pdb
 cd ../

(2) Next align the two receptors (4NVE to 4NVA) and find those waters close to the ligand.

 mkdir aligned
 cd aligned

create file named "chimera.com" and write the following into the file:

 # Gist template #0
 open ../4NVA/rec.pdb
 # 4NVE rec
 open ../4NVE/rec.pdb
 # 4NVE lig
 open ../4NVE/lig.pdb
 # 4NVA waters
 open ../4NVA/water.pdb
 mmaker #0 #1
 matrixcopy #1 #2
 sel #3:HOH & #2 z<8
 del #3 & ~sel
 write format pdb  0 template.pdb
 write format pdb  1 aligned.rec.pdb
 write format pdb  2 aligned.lig.pdb
 write format pdb  3 close.waters.pdb
 /nfs/soft/chimera/current/bin/chimera --nogui chimera.com > & chimera.com.out
 cd ../aligned

(3) Now lets run tleap.


Run GIST post processing

Combining GIST grids

Run convergence analysis