[gmx-users] Re: gromacs tutorial

[Tsjerk Wassenaar]

Hi Michal,

I think Mark has explained this well enough on the gromacs user list.
But, since a tutorial is intended to learn something, I'll try to be
more elaborate and give some additional suggestions and exercises.

First of all, in the tutorial you're encouraged to look at the files
(.pdb, .gro, .top, .mdp). These are human readable files which are not
too difficult to understand. I hope you've had a look at them,
otherwise, I'd suggest you to do that first.

Now, the .pdb or .gro files contain coordinates only, together with
the designation of which particular atom of which particular residue
the coordinates belong to. There's nothing telling you which atoms are
connected and in which way. That is where the .top file comes in. You
should have a look in the Gromacs manual and read Chapter 5 to
understand more about topologies. The .top file maps the particles to
atom types and lists the bonds, angles and dihedrals between atoms in
the molecule(s).

Thus, together a .gro/.pdb file and a .top file are together
sufficient to fully characterize a molecular system at a given time.
Then from that system you want to start a simulation. There it comes
down to telling the md program what it should do with the molecular
system (as fully characterized by the .gro/.pdb and .top file). It
needs to no how to perform the integration, for how many steps, what
to save and at which interval, the temperature, etc, etc. These are
all, in essence, completely independent of your particular molecular
system. This part of the input, the run parameters, are listed in the
.mdp file. You should read through the Gromacs manual Chapter 7 to get
a feel for the contents of that file. Also check the .mdp files
provided in the tutorial and don't do anything before you understand
what each line actually does.

Finally, note that the parameters in your .mdp file, though
independent of your system, depend heavily on the question you are
trying to answer. That is where you should spend a sufficient amount
of thought, determining what particular parameters you need, rather
than just using a .mdp file you scored from a tutorial, which most
likely had other objectives than you had.

I hope that by now, you've acquired enough knowledge about these
things to revisit your questions and provide sensible answers
yourself.

Best,

Tsjerk


On 10/24/06, Michal Walczak <meehow.w at gmail.com> wrote:
> Hello!
>
> I've been doing your Gromacs tutorial and I found myself in
> difficulties with fullmd_sol.mdp. It is mentioned that I can download
> that file from underlined link, but is this file the same for all
> simulations or only for that particular hen-egg lisozyme ? If I need
> different file for different molecules how can I obtain such a file
> for 3 aa peptide which I'm currently investigating? How it works with
> fullmd_sol_1ps.mdp and minim.mdp? Do I have to create them on my own
> or can I download them for any molecule?
>
>
> Best regards,
>
> Michal Walczak
>


-- 
Tsjerk A. Wassenaar, Ph.D.
Groningen Biomolecular Sciences and Biotechnology Institute (GBB)
Dept. of Biophysical Chemistry
University of Groningen
Nijenborgh 4
9747AG Groningen, The Netherlands
+31 50 363 4336