[gmx-users] g_order for lipid bilayers

[Justin Lemkul]

On 2/20/13 10:31 AM, Felipe Pineda, PhD wrote:
> Hi again,
>
> my topology is non-consecutively numbered. I was looking for a tool to renumber
> it, but didn't find it yet. A hint will be really appreciated. I was not sure

I doubt one exists.  Many people have asked for such a thing, but no one has had 
a desire to create one.  It should be a matter of simple find and replace 
though, so a quick Perl script should probably do the trick.

> either whether or not is it possible to use the xtc generated with the original
> topology with the renumbered one. My main doubt is how to properly generate the
> index file for g_order. If each group contain an atom type in the chain, let's
> say all C21, another all C22, etc., I get many warnings of the type:
>

The .xtc file has no knowledge of atom names or numbers.  It simply stores 
coordinates for atoms in the order that they appear in the coordinate file 
(which inherently must correspond to the topology).

> WARNING: distance between atoms 6767 and 6751 > 0.3 nm (1.406577). Index file
> might be corrupt.
>

Can you verify using any tools that the distance between these atoms should in 
fact be < 0.3 nm?  There are 16 atoms in between here, so it seems as if they 
should not be consecutive, even if you are using an all-atom model.

> If all atoms belonging to one chain are put together in a group, then the xvg
> files are empty.
>

Indeed, because that's not how g_order works.

-Justin

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

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

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