C687 Tutorial:
Molecular Dynamics and Simulated Annealing
The purpose of this tutorial is to learn how to do basic molecular dynamics
using the Discover module.
The guidelines given in this tutorial are intentionally very loose.
You will need to figure out for yourself all the details of how you do these
exercises. Refer to the InsightII Simulated Annealing notes.
The overall aim is to do a simulated annealing run on the zinc finger for
which we built a homology based model in an earlier tutorial, and then
to analyze the results of the run.
This tutorial will be performed in the style of popular cooking shows,
i.e. you set up the run (but don't actually run it), then you analyze the
results of a run that I did earlier (there is not enough time to do the run
and analyze it in one tutorial session).
All files relevant to the run are in the directory /ruser/instruct1/stone/C687/MD and
are labeled zf24.### (various files).
Step 1: Getting Started
Duration: ~5 minutes
- Start up Insight and read in the file /ruser/instruct1/stone/C687/MD/zf1.psv
This is the homology modeled zinc finger, energy minimized with the backbone
fixed.
Step 2: Forcefields, Potentials, and Restraints
Duration: ~15 minutes
- Choose a Forcefield
- Check all potentials and charges
- Open Discover
- Fix the positions of the four zinc ligands (2 Cys and 2 His).
Color fixed residues
- Tether all backbone atoms with a force constant of 5.
Step 3: Setting Up the Command File
Duration: ~20 minutes
- Click on Run/Run, and select the local machine, the molecule name, COMMAND FILE,
auto-assign parameters, minimize, and Reduce Output. Then click EXECUTE .
- Now go into a UNIX shell and read through the command file which has
the suffix '.inp'. To understand each line in the file, follow
through the example (zf24.inp).
- At this stage, if you were actually going to perform a dyamics run you
would edit the command file to make it look like the example file. However,
for now it will suffice just to read through the example file and make sure
you understand each command. Spend ~10 minutes doing this.
Pay particular attention to the following features of the command file:
- Comment lines begin with an exclamation mark (!)
- Parameters for changing the non-bond cutoff, the dielectric, and
the decay time constant for heating and cooling.
- Specification of tethered and fixed atoms
- Initial quick minimization to reduce very high energies
- Second, longer minimization before starting the dynamics
- The equilibration step
- The beginning and end of the dynamics loop
- The steps within the dynamics loop
- Heating and high temperature dynamics
- Cooling and lower temperature dynamics (annealing)
- First, quick minimization
- Second, longer minimization
- Recording the results
- Also look at the files zf24.car and zf24.mdf which were created by
Discover at the same time as the command file.
Step 4: Running the Dynamics
Duration: ~5 minutes
Read through the instructions on how you would submit this dynamics job and check on its progress.
I already ran this job on pchem03 with a "nice" number of 40. It took
1 day, 10 hours, 34 minutes, 34.54 seconds of real time, starting at
9.25 am on a Wednesday.
The results are in the directory /ruser/instruct1/stone/C687/MD and
are labeled zf24.### (various files).
Step 5: Analyzing the Results
Duration: ~30 minutes
Follow through the directions on how to evaluate the simulated annealing.
Spend some time viewing the trajectory, trying out a few different ways of
graphing the results and identifying structural features of the most
stable structures. Also look at the secondary structures and whether the
hydrophobic groups cluster together.
Back to | C687 Summer 1997 | Courses
& Instruction | MolViz
Home |
Send comments to chemvis@indiana.edu
Last updated: 01/23/2001