[gmx-users] Coordinate file for lipid bilayer

[Christopher Neale]

Sounds dangerous to me. Here's a possible methods section from your paper:

"we obtained a popc bilayer somewhere on the internet. Unfortunately, this structure did not match our topology file. Nevertheless, it has the same number of atoms and so we assumed that the difference was simply a matter of reordering ..." (not too appealing I think).

Peter's advice was good and I recommend avoiding a reordering based on an assumption that it will be correct. Using the method that Peter outlined, there will be some issues with equilibration at the new periodic boundary, but 100 ns of MD simulation should take care of that just fine (note that the Tieleman 128 lipid file is pretty old and, computers being slower then, I imagine that it was equilibrated with fewer than 100 ns itself).

Chris.


On 2012-05-28 08:03, James Starlight wrote:
> Peter,
>
> Thanks for advise.
>
> I've found already pre-equilibrated POPC bilayers with 200 lipids.
> I've examined that lipids and found that they are very similar to the
> berger's lipids (it consists of equal nymber of atoms ) but the atom
> order in each lipid is slightly different than in Tieleman's popc.itp
> file so during processing of that lipids I've got error of
> non-matching atoms. Is there any trivial way to make new popc.itp
> based on existing gro file with correct atom order ?
>
>
> James
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120528/528bb0d0/attachment.html>