[gmx-users] PR for CG lipid and CG protein

[chris.neale at utoronto.ca]

Hi Justin,

I just double checked and certainly the confout.gro from my EM run in  
parallel with domain decomposition in mdrun is broken over periodic  
boundaries.

I'm running gromacs-4.0.5 on a nehalem under openmpi. Here is my .mdp file

gpc-f101n084-$ cat em_pos.mdp
define=-DPOSRES_INDO
constraints = none
integrator = steep
tinit = 0
nsteps = 5000
nstcomm = 1
nstxout = 5000
nstvout = 5000
nstfout = 5000
nstxtcout = 5000
nstenergy = 5000
ns_type = grid
rlist = 1
rcoulomb = 1
rvdw = 1
coulombtype = PME
ewald-rtol = 1e-5
optimize_fft = yes
fourierspacing = 0.12
fourier_nx = 0
fourier_ny = 0
fourier_nz = 0
pme_order = 4
gen_seed = -1

; COM PULLING
pull                     = umbrella
pull_geometry            = position
pull_dim                 = N N Y
pull_start               = no
pull_nstxout             = 500
pull_nstfout             = 500
pull_ngroups             = 1
pull_group0              = POPC
pull_pbcatom0            = 0
pull_group1              = Protein
pull_pbcatom1            = 0
pull_init1               = 0 0 0.0
pull_rate1               = 0
pull_k1                  = 500.0
pull_vec1                = 0 0 0


Chris.

-- original message --

On 1/9/10 10:49 AM, chris.neale at utoronto.ca wrote:
> Dear Sunny,
>
> I replied a while ago, but it seems to have not made it to the list.
> When using inflategro with gromacs3, you can run cycles like this:
>
> 1. inflategro.pl
> 2. mdrun (EM)
> 3. goto 1
>
> When running with gromacs4, however, you must run cycles like this:
>
> 1. inflategro.pl
> 2. mdrun (EM)
> 3. trjconv -pbc mol
> 4. goto 1
>
> since mdrun now writes broken molecules.

Are you sure that recent Gromacs versions still do this?  I requested  
a fix (for
final output configurations only) with the original version 4.0, which  
I thought
(if memory serves) was fixed as of version 4.0.2, released over a year  
ago.  Is
this not the case?  I get intact output structures from everything I've done
since 4.0.2.

Trajectories certainly still contain "broken" frames, the confout.gro  
should not
be an issue.

-Justin

>
> The developer of inflategro.pl, Christian Kandt, is aware of this and
> may have modified the script to allow broken molecules in inflategro,
> but if you have a version that is more than 2 months old then it
> certainly can not handle molecules that are broken over PBC, as lipids
> usually are near the boundary.
>
> Chris.