On Wed, Aug 3, 2011 at 11:08 AM, Massimiliano Porrini <M.Porrini.ed.ac.uk>wrote:
> Dear Jason,
>
> I'm very thankful for the answers to my questions.
>
> Eventually I recreated my seh.lib file from scratch, applying the correct
> directives and it works perfectly now.
>
> As far as the frcmod file is concerned you were right. The thing is
> that I had erroneously defined
> the -CH_2 hydrogens as HC, while they are H1 type.
> So the OH - CT - H1 angles parameters are now included in 99SBildn force
> field.
>
> Just for curiosity and to totally clarify the issue I have one more
> question:
>
> In AmberTools Users' Manual, pag. 54, the LEaP command "bond" allows one to
> define also the bond order (single, double, triple and aromatic).
> I am wondering why, when (as you wrote previously) amber FF just needs to
> know
> what is bonded to what and the bond order will be considered via force
> constants and
> equilibrium distances.
>
I believe (but I'm not sure here) that tleap will actually do some basic
structure refinement and relaxation with its own rudimentary force field
derived via bond orders. Furthermore, some file formats do in fact have
bond order information present in them (i.e. the mol2 file, and even the
amber OFF file -- see
http://ambermd.org/doc/OFF_file_format.txt for
details). I believe what leap does is assign different equilibrium
lengths/force constants based on the bond order when constructing its basic
force field parameters for coarse relaxation. However, none of this
actually translates into the prmtop, and since I've never used any feature
requiring proper bond orders, I've just ignored them.
If you look through the amber prmtop format (
http://ambermd.org/formats.html),
you'll see no reference anywhere to "bond order". Just which bonds exist,
their equilibrium values, and their force constants. Same for angles.
HTH,
Jason
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Aug 03 2011 - 09:30:02 PDT
