Comments on comments...
> Do I understand correctly that tleap will in fact (silently) change the explicit improper definitions based on the alphabetical rule?
> Leap expects the force field developer to enter torsion terms in the proper order, and (to my knowledge) it doesn't try to "fix" any non-alphabetical entries.
Yes re-ordering will happen in the prmtop!... I already faced this issue while implementing SIRAH (it is checked with ParmEd).
My guess is that re-ordering is a trick to speed up the search and setting of these torsions...
> For purely planar terms I don't know off hand if this actually matters (probably not for energies, maybe for forces?)
As Bill Ross pointed out, planar centers won't be affected in terms of energy or forces, because the parameters are the same regardless the atom order, if the central atom is correctly set.
On the other hand, defining chiral centers with improper terms will be an issue, but this is not used in all-atoms force fields, so you shouldn't have a problem... however it's my nightmare for using coarse-grained models :-(
Best,
MatÃas
----- Mensaje original -----
De: "David A Case" <david.case.rutgers.edu>
Para: "AMBER Mailing List" <amber.ambermd.org>
Enviados: Jueves, 24 de Septiembre 2020 16:20:56
Asunto: Re: [AMBER] How does tleap detect improper torsions?
On Thu, Sep 24, 2020, Brian Radak wrote:
>
>Do I understand correctly that tleap will in fact (silently) change the
>explicit improper definitions based on the alphabetical rule? For purely
>planar terms I don't know off hand if this actually matters (probably not
>for energies, maybe for forces?)
Here's my understanding: tleap figures out what impropers there should
be in this way: for each central atom that has exactly three bonded
neighbors, look in the force field for an improper in which the three
non-central atoms obey the alphabetical rule.
If it finds a match, it uses it;
If it fails to find a match, that's OK: it assumes that there is not
supposed to be such an improper, because it is not found in the
force field. For things with sp2 hybridization, it will print a
warning.
I'm not sure exactly what you mean by "change the explicit improper
definitions", because I'm not sure what is meant by "explicit improper
definitions". Leap expects the force field developer to enter torsion
terms in the proper order, and (to my knowledge) it doesn't try to "fix" any
non-alphabetical entries.
If there ever is any uncertainty, use parmed to inspect your prmtop file
to see what terms are really there.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Sep 24 2020 - 21:00:03 PDT
