[gmx-users] Re: Pbc and Com in simulation video

[janne.hirvi at joensuu.fi]

Thanks for your help Alan, Yang, and Belquis!

Now I managed to get trajectory in which a droplet stays in the middle  
of the simulation box while a surface moves below the droplet. First I  
used trjconv to remove periodicity (-pbc nojump) and then performed  
translational fit for the droplet (-fit translational) and finally  
transferred all the translating surface atoms back to the simulation  
box (-pbc -inbox).

However, there is still a problem with fitting procedure which is  
performed in three dimension and hence the surface translates upwards  
during the video. It should be quite easy to modify trjconv so that  
the fitting is performed only for the xy coordinates (not for the z  
coordinate perpendicular to the surface) but with my programming  
skills I wasn't able to do that (actually I wasnt able to find the  
correct lines). So I would like to know if someone has made a such  
modification to trjconv (trjconv_d) or could help me to do it?

Thanks,

Janne


> Hello Janne,
>
> this is what I usually do when my molecules cross the box or dance all
> around and it works for me:
>
> I first do this with the "original" unmodified or fitted xtc or trr file:
>
> trjconv -f  -o  -pbc nojump
>
> then i take the modified trr or xtc and do a fit:
>
> trjconv -f -s -fit rot+trans
>
>
> Belquis
>
>
>> Hello gmx-users!
>>
>> I am trying to make a simulation video in which a water droplet is
>> rolling or sliding on a surface. Due to a periodic boundary conditions
>> a droplet is not whole all the time but it disappears from the right
>> and appears from the left in cycles.
>>
>> I think that I could create more visual video by removing the lateral
>> motion of the center of mass of a droplet so that the droplet would
>> stay whole
>> in the middle of simulation window and the surface would translate
>> below the droplet. I am not exactly sure how realistic this kind of
>> video would be and would like to know if someone has better ideas?
>>
>> If there are no better ideas I would need advices to be able to create
>> a such trajectory or video. Right now I am struggling with trjconv
>> which might be suitable for this purpose. However, if I use the
>> droplet as a group for translational fit (-fit translation) I get
>> trajectory in which the whole system wanders strangly (laterally (the
>> surface moves first to the left and then back to the right) and also
>> in perpendicular to the surface due to the lowering of the height of
>> the center of mass of the droplet during the simulation) and the
>> droplet wont stay whole in the middle of the box. Any ideas what I am
>> doing wrong or how I would be able to get rid of even this lateral
>> wandering so that I could at least see how realistic this procedure is?
>>
>>
>> Thanks for your time and help in advance,
>>
>> Janne
>>
>>
>> ------------------------------------------------------------------------------
>> Janne Hirvi, MSc(Physical Chemistry), Researcher
>> University of Joensuu, Department of Chemistry, P.O.Box 111 80101 Joensuu,
>> FI
>> Tel: +358 13 2514544 & +358 50 3474223
>> E-mail: Janne.Hirvi at joensuu.fi & hirvi at cc.joensuu.fi
>> ------------------------------------------------------------------------------