[gmx-users] Including itp files

[Justin A. Lemkul]

Volker Wirth wrote:
> Hi all,
> maybe these are stupid questions, but in the moment I'm a bit in a
> files' mess, so I apologize for that...
>
> I want to use a lipid bilayer pdb from the Tieleman group, let's say
> http://moose.bio.ucalgary.ca/files/dppc64.pdb, and the first goal is
> to get the structure "into" GROMACS and to see if it works. If I run
> pdb2gmx -ignh -ff G53a6 -f dppc64.pdb -o lipids.pdb -p lipids.top
> I get a problem with the rtp file, "Atom N4 in residue DPPC 1 not
> found in rtp entry with 50 atoms while sorting atoms", , but it works
> with gmx, so
>
> 1) may I just replace the [DPPC] part of ffG53a6.rtp with the
> equivalent part of ffgmx.rtp or have I to edit the G53a6 file so that
> the first columns are the same?
>
>   

Well, if you replace parameters from ffG53a6 with those of the 
*deprecated* ffgmx, you won't be using the ffG53a6 force field any 
more.  Besides, the Gromos parameters for lipids do not represent the 
lipid properties as well as the Berger parameters from Tieleman's site, 
so I don't know exactly why you are doing this...

> It works with the gmx force field and I can run editconf and genbox.
> But when I use
> grompp_d -f em.mdp -c lipids_b4em.pdb -p lipids.top -o lipids_em.tpr
> the program is determined by too many warnings like "WARNING 1 [file
> "lipids_A.itp", line 69]: No default Bond types, using zeroes" so I
> think, this is the place, where the itp files (lipid.itp, dppc.itp),
> offered besides the pdb should do some job here. But
>
> 2) how will I get these files involved into my simulation?
>   

What you need to do is read a bit about including .itp files (Chapter 5 
of the manual), and search the list archive, since membrane questions 
are quite common.  Here are the basics:

1. Append the lipid parameters to your desired force field *nb.itp and 
*bon.itp files.  I believe the lipid.itp atom types come from the Gromos 
force fields, so you can append them to the appropriate force field 
there (and I have seen many people use the Gromos96 FF's in literature 
simulations of membrane proteins), or you can follow Chris Neale's 
elegant setup for use with OPLS parameters (search the list archive).

2. Then, you will need a simple topology file, calling (for example) 
"modified_FF.itp" as the force field topology to which you have appended 
your lipid parameters.

#include "modified_FF.itp"

#include "dppc.itp"

#include "spc.itp"

[ molecules ]
DPPC   x
SOL    y

Hopefully that makes sense.  Do check the archive, there are lots of 
messages about this type of thing.

-Justin

> I hope, you understand my problem, as for myself it is not yet really
> clear, what e.g. the itp file will actually do within the process...
>
> Best
> Volker
>
>   

-- 
========================================

Justin A. Lemkul
Graduate Research Assistant
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================