Hi Brett,
Antechamber will make you one ligand library file (ideally get a .lib, but
a .prepi is sufficient) and this should be sufficient to describe all four
copies of the ligand in your system. It is probably preferable to have one
description of the ligand that spans all the contexts in which it appears:
there is, after all, only one description for each amino acid (well,each
distinct protonation state of every amino acid), no matter where it appears
in the protein. AmberTools and antechamber can easily take care of your
problem; if your ligand is a common biopolymer subunit, you might consult
the libraries in ${AMBERHOME}/dat/leap/lib/ to see if it's already there.
Your question gets at a fundamental aspect of non-polarizable simulations,
though: even if the electronic structure is expected to change upon
binding, most simulations and the available implementations for methods
such as steered MD or umbrella sampling will demand that the ligand retain
its properties over the entire reaction coordinate. If you really think
that the properties of the ligand are a function of its place around the
receptor, you may have some luck using thermodynamic integration with
multiple models to estimate this contribution. However, if you do that,
understand that the MD program doesn't know that there is a cost to making
each model, e.g. if the charge distribution changes then it costs energy to
get it that way, and in the classical simulation this fee is waived. Some
metric for estimating the relative cost of each solute would be needed to
complete the thermodynamic cycle--these metrics exist, but are not
currently part of AmberTools.
Dave
On Fri, Apr 8, 2016 at 1:48 AM, Brett <brettliu123.163.com> wrote:
> Dear All,
>
> If the protein complex containing several identical ligands (for example 4
> ADP), can the antechamber made the ligand files (should 4 different sets of
> ligand files be made, for each ADP)? Can AmberTools 15 make this MD
> workable?
>
> Brett
>
> _______________________________________________
> 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 Apr 07 2016 - 23:30:03 PDT
