[gmx-developers] Heads up! trjconv broken
Anton Feenstra
feenstra at few.vu.nl
Tue Jun 15 09:38:17 CEST 2010
On 14/06/10 19:25, Daniel Larsson wrote:
> On Jun 14, 2010, at 18:56 , Tsjerk Wassenaar wrote:
>
>> Hi Daniel,
>>
>> -fit and -pbc should not be combined, and -pbc should never follow a
>> call with -fit. Note that fit rotates the coordinates with respect to
>> the box, breaking the relation with the PBC. Then think whether the
>> mangling you see follows logically from this, or not,
>
> Yes, the settings that I've tried now that causes the unexpected behavior do all involve the -fit flag.
>
> So, if I understand this correctly: -fit with either rot+trans, rotxy+transxy or progressive will/might not wrap atoms that passes a periodic boundary "correctly".
>
> This is not something that was obvious to me, and I think it might be good to add a note about this in the documentation as a service to the user. Perhaps trjconv should even disallow any PBC flag together with these above options? And perhaps also display a warning in conjunction with the use of these flags (since one can have made multiple passes over a trajectory).
Do the trajectory formats allow the storage of rotated boxes? In that
case any operation that rotates the coordinates should also rotate the
box. (But then the PBC routines should also work with arbitrary rotated
boxes!)
If not, it might be best to simply remove the box altogether after
rotation; that would effectively make it impossible to do PBC afterwards.
--
Groetjes,
Anton
_____________ _______________________________________________________
| | |
| _ _ ___,| K. Anton Feenstra |
| / \ / \'| | | IBIVU/Bioinformatics - Free University Amsterdam |
|( | )| | | De Boelelaan 1083A - 1081 HV Amsterdam - Netherlands |
| \_/ \_/ | | | Tel +31 20 59 87783 - Fax +31 20 59 87653 - Room P136 |
| | Feenstra at few.vu.nl - www.few.vu.nl/~feenstra/ |
| | "Pump Up the Volume Along With the Tempo" (Jazzy Jeff)|
|_____________|_______________________________________________________|
More information about the gromacs.org_gmx-developers
mailing list