[gmx-users] trjconv, pbc and breaking of multiple peptides

[ms]

On 09/11/10 21:36, Justin A. Lemkul wrote:
>> I am doing REMD simulations of multiple homopolymeric peptides in a
>> PBC box, in vacuum (it's a custom coarse grain). GROMACS 4.0.5. I want
>> to analyze features of the system that require to use g_gyrate to find
>> out the moments of inertia of the system, for example.
>>
>> I understand g_gyrate is one of the few tools that is *not* PBC aware.
>> I am trying therefore to correct my REMD trajectories using several
>> variants of trjconv, with simple command lines like:
>>
>> trjconv -f xac_12855257520.xtc -o xac_12855257520_pbcatom.xtc -pbc nojump
>>
>> ...but alas, when I see them with ngmx, the peptides tend to break. I
>> tried both pbc nojump and pbc atom, with same results (pbc nojump
>> perhaps is a bit worse). I can't use pbc cluster because they're not
>> always clustered and I am interested in both the clustered and
>> non-clustered states.
>>
>> I'd like to give more information about that but I don't really know
>> what is relevant to debug this kind of issue. I looked in the mailing
>> list but (despite multiple woes about pbc nojump etc.) I didn't find a
>> helpful thread. Any hint?
>>
>
> There are numerous -pbc options with trjconv; have you tried others? I
> have never had luck with -pbc nojump actually working, and -pbc atom
> provides no guarantee that molecules will be properly reconstructed.
> Using -pbc mol -center is often a much better approach, in my experience.

Excellent advice; it seems to do exactly what I want. Thanks a lot!!

Massimo


> -Justin
>
>> thanks!
>> Massimo
>>
>


-- 
Massimo Sandal, Ph.D.
http://devicerandom.org