next up previous
Next: Bibliography Up: VMD Tutorial Previous: Semi-empirical parameter generation: SPARTAN


Minimization with new parameters

In this unit, you'll use our state of the art molecular dynamics program NAMD2 with the Charmm22 force field. If you did the initial copy command at the beginning of the tutorial, you should have the following files in your directory:
Topology file: top_all27_prfar_cyg_nh4.inp
Parameter file: par_all27_prot_na_lipids_full.inp
NAMD config file: HisHmini.namd
PDB file: name of your MOE generated PDB file: filename.pdb
The parameter file was generated with the same protocol used in the previous section of this tutorial. However, rather than performing a semi-empirical calculation, the parameters in par_all27_prot_na_lipids_full.inp were calculated by full ab-initio quantum mechanics using the 6-31G** basis set, a process that takes several hours on a fast machine. First we have to generate a NAMD2-compatible PDB file. You can generate a clean PDB file using the VMD console. Load the PDB file HisHmoe_docked.pdb into VMD and run the following commands in the VMD console:
set sel [atomselect top "all"]
$sel set segname HISH
$sel writepdb HisHmoe_cyg.pdb
Now open the VMD generated file HisHmoe_cyg.pdb in a text editor and change the name of residue 84 from CYS to CYG. Otherwise PSFgen will not recognize the correct topology for the residue. We also must add hydrogens with the topology file using the program PSFgen in VMD. In the topology file the connectivity and therefore also the correct protonation state for each individual residue is defined. This allows us in contrast to the hydrogen adding procedure in MOE to set up the system with exact biochemical properties resembling glutamine bound to CYS84. Start PSFgen in the VMD console by running:
!Generate protein structure file with PSFgen

package require psfgen
topology top_all27_prfar_cyg_nh4.inp

segment HISH {
first NTER
last CTER
pdb HisHmoe_cyg.pdb
}

coordpdb HisHmoe_cyg.pdb HISH

guesscoord

writepsf HisH_gln.psf
writepdb HisH_gln.pdb
When you are finished you have PDB file containing all coordinates including hydrogens and PSF file containing all information about connectivity, mass and charge of each individual atom in the structure. Just have a look at both files with an editor and get familiar with the format.
\framebox[\textwidth]{
\begin{minipage}{.2\textwidth}
\includegraphics[width...
...f{Question:} Does the PSF file resemble the topology file?}
\end{minipage}
}
Finally you will solvate the system for running a minimization in a NPT ensemble with a timestep of 1 femtosecond, a uniform dielectric constant of 1 and periodic boundary conditions. The electrostatic interactions will calculated with the Particle Mesh Ewald Method. Start Solvate in the VMD console by running:
!Solvate system
package require solvate
solvate HisH_gln.psf HisH_gln.pdb

\framebox[\textwidth]{
\begin{minipage}{.2\textwidth}
\includegraphics[width...
...toms does the file now contain before and after solvation?}
\end{minipage}
}
Run the minimization on the solvated system using the configuration file HisHmini.namd. We will be running all the minimization jobs on the NCSA Platinum Supercomputing Cluster. Be sure you are logged into the platinum.ncsa.uiuc.edu domain before running the minimization job. Once logged in, at the unix command prompt, type:
namd2 HisHmini.namd > HisHmini.namd.out

\framebox[\textwidth]{
\begin{minipage}{.2\textwidth}
\includegraphics[width...
...ng used? What is temperature
during the minimization run?}
\end{minipage}
}
The output of your NAMD job can be found in HisHmini.namd.out. You can use namdplot to look at energy changes during the minimization. We will also provide you with the final DCD file generated during the minimization run. It can be found in felix/Forcefield (note: if you did the cp -r command in the Introduction, you already have the file). All the minimization files are located in the directory minData (to get there, type "cd minData" at the command line). Load the DCD file in VMD and check the relevant distances between glutamine and the catalytic triad. You can determine how much the structure has changed during the minimization by using the RMSDscript.
\framebox[\textwidth]{
\begin{minipage}{.2\textwidth}
\includegraphics[width...
...e? Has the structure converged during your minimization?
}
\end{minipage}
}


next up previous
Next: Bibliography Up: VMD Tutorial Previous: Semi-empirical parameter generation: SPARTAN
vmd@ks.uiuc.edu