VMD-L Mailing List

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Sat Dec 03 2011 - 18:18:38 CST

dear tim,

On Fri, Dec 2, 2011 at 12:02 PM, Timothy Sirk <tsirk_at_vt.edu> wrote:
> apologies if this is a double post, it was dropped last time. maybe the
> attachment was too big.
> ======================================================
>
> Dear All,
>
> I am using topotools 1.2 with VMD 1.9 to write gromacs topology files from
> lammps data files. With most of my datafiles this seems to work great. The
> one exception is attached.
>
> When I write a top file from this particular data file, the 'atoms' section
> is numbered incorrectly and out-of-order (see sniplet below). This numbering
> is fatal for grompp. Can someone make a suggestion why this is happening, or
> even try to reproduce the problem using my datafile? This is 60 molecules of

i can reproduce it with my development version.
i will look into it. johns is getting ready for the
next release and we are currently finalizing features
and chasing down known bugs that we want to
have sorted out for this release.

> pe at 78 atoms/molecule. I could produce a simplified system from your
> suggestions.
>
> The system is a molecular crystal, where bonds are wrapped completely around
> the box. It was orginally PCFF forcefield, but the class2 terms are stripped
> off (BondBond, BondAngle etc) to work with topotools.

this should no longer be needed with the next topotools update.
i have implemented a fix for that just a few minutes ago.

> There is also a minor issue with the number of atoms per charge group. For
> many of my datafiles, the top output still has one too many atoms in some
> charge groups (33 instead of 32).  Its much improved from topotools 1.1, but
> having 32 or less would be really convenient.

it works for this example, but can you please send me an input
that still has problems? i can just arbitrarily reduce the constant
for the maximal number of charge groups, but i'd rather understand
the reason why you are getting the off-by-one error.

more later,
     axel.

>
> The datafile and entire .top file are attached.
>
> Thanks,
>
> Tim
>
>
> [ atoms ]
> ; nr  type  resnr residue atom cgnr charge  mass
>      1       type1      1      RES      1      1     0.0530     1.0080
>      2       type2      1      RES      2      1    -0.1060    12.0112
>      3       type1      1      RES      1      2     0.0530     1.0080
>      7       type1      1      RES      1      2     0.0530     1.0080
>      8       type2      1      RES      2      2    -0.1060    12.0112
>      9       type1      1      RES      1      3     0.0530     1.0080
>     13       type1      1      RES      1      3     0.0530     1.0080
>     14       type2      1      RES      2      3    -0.1060    12.0112
>     15       type1      1      RES      1      4     0.0530     1.0080
>     19       type1      1      RES      1      4     0.0530     1.0080
>     20       type2      1      RES      2      4    -0.1060    12.0112
>     21       type1      1      RES      1      5     0.0530     1.0080
>     25       type1      1      RES      1      5     0.0530     1.0080
>     26       type2      1      RES      2      5    -0.1060    12.0112
>     27       type1      1      RES      1      6     0.0530     1.0080
>
> 

-- 
Dr. Axel Kohlmeyer
akohlmey_at_gmail.com  http://goo.gl/1wk0
College of Science and Technology
Temple University, Philadelphia PA, USA.