[gmx-users] trjconv -pbc nojump

[Justin Lemkul]

On 9/2/15 9:36 AM, Reisser, Sabine (IPC) wrote:
> Well, in the end I would like to have an average value for the P atoms of
> each leaflet. I know which lipids belong to which leaflet, but I need to
> avoid box jumps along z, otherwise I will obviously get a wrong average.
> Using first trjconv  -pbc whole and then -pbc nojump actually seems to work
> in the way that molecules don't jump in z anymore. However, they are not
> whole anymore. Also, I would like all lipids of one leaflet on one side of

Once made whole, I don't understand how they would have further problems.  Make 
sure your reference structure in the .tpr passed to -s has whole molecules that 
are all in the box as a proper reference.

> the box. Now, the lipids upper edge is at the box edge, and some have jumped
> to the lower box edge. I thought maybe with -pbc cluster I could fix this.
> Otherwise I have to shift the molecules which are jumped (already in the
> starting structure) manually by the box height.
>

This can be overcome with the -trans (in concert with -pbc mol) and/or -center 
options.

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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