On Tue, Sep 8, 2015 at 10:55 AM, Hannes Loeffler <Hannes.Loeffler.stfc.ac.uk
> wrote:
> Hi,
>
> I was wondering if index and sort order are guaranteed in a prmtop file.
> E.g. if I have a bond A-B or an angle A-B-C, can I rely on A < B or A
> < C? Also would bonds be sorted in ascending order on A, angles on
> A then B, etc.?
>
I think the answer to this is 'no'. In practice, LEaP might be
implemented this way (but I doubt it), but sander and pmemd don't require
it and other tools that generate prmtop files (e.g., ParmEd) may not
implement things that way (e.g., when converting from CHARMM or GROMACS).
The only ordering which *is* guaranteed in the prmtop file (to my
knowledge) is that larger atom indices are always excluded from smaller
atom indices in the exclusion list (and never the other way around).
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Sep 08 2015 - 13:30:05 PDT
