Analysis


Making sense of random values


Visualizing

The most pleasurable part of the entire process is visualizing the entire simulation. The wiggling and ziggling of molecules is at times fascinating. To visualize the simulations, we will make use of the psf file and the dcd files generated during the simulations. The dance of molecules can be viewed in the Visual Molecular Dynamics (VMD) software. There are a number of commands that will help you in this process.

To start with, load the final.psf file into vmd. Next, load the dcd file you want to watch. There is a dcd file generated for each namd configuration file. That means a dcd file for minimization, 7 for heating (1+2*3), 1 for equilibration and finally "n" for production runs. You can load them manually, or automate it using a tcl script. The automation works based on the fact that you can load a molecule in vmd using the tk console window.

To load the psf file, type: mol new final.psf
Next to load the dcd file type: mol addfile min.dcd
Easy peasy? Here is a tcl script that loads all the dcd files in your simulation.

Here is an interactive video to help you out with visualizing the molecules more effectively in VMD.


Viscosity

This one requires some extra effort from the user. You need to write code to calculate the auto time correlation functions. The method is based on the Green-Kubo formula and isn't something you should try unless you are adamant in getting these values. I have tried to explain the process in this link. Have a look. There are three different aspects of the code. Calculating the time-correlation function, averaging it over n time origins, and finally block averaging to get the values. Though it might appear a bit hard (as I have refrained from providing the code) but a viscosity calculation will really change the nature of your publication ;)


Previous Page                   Main Page