AMBER Archive (2004)

Subject: Re: AMBER: Titratable amino acid termini

From: John Mongan (jmongan_at_mccammon.ucsd.edu)
Date: Tue Apr 27 2004 - 18:28:15 CDT

You can do this, and it shouldn't be too much more difficult than
constructing any other new titratable group.

As described in the manual, you'll need to calculate charge sets for
each of your protonation states. You'll probably also want to build a
new C-terminal residue that has 4 acidic hydrogens, as has been done for
the AS4 and GL4 groups.

Next you'll need to construct a reference compound (something for which
you have a reasonable reference pKa for) and conduct thermodynamic
integration to determine the electrostatic transition energy between the
protonation states.

One thing to be aware of here is that it is important for charges that
change between the different protonation states to see the same 1-4
interactions in the reference compound as they do in your system of
interest. (This is why the backbone charges do not change in the
titratable groups distributed with AMBER 8.) We're working on a general
solution for terminal residues, but for now the simplest way to deal
with this issue is to use RES1 - RES2 - NME for your N terminal
reference compound and ACE - RES(n-1) - RES(n) for your C terminal
reference compound. (This assumes that none of these residues have
titrating sidechains.)

When you have the transition energy, you can add the definition to
CPin.pm as described in the manual.

Finally, you'll either want to hack cpinutil.pl so it recognizes
terminal residues as being different than interior residues, or edit the
PDB file produced by ambpdb to change the names of the terminal resiudes
so that your terminal residue definitions are applied only to the
terminal residues.

Hope this helps,

John

On Tue, 2004-04-27 at 02:43, Sarah Harris wrote:
> Dear AMBER experts
>
> We have recently obtained AMBER8 and are particularly interested in
> running constant pH simulations of an 11 amino acid peptide. However,
> this peptide has titratable C and N termimi as well as titratable side
> chains. Is there any way that we can include these in our calculations
> by obtaining suitable parameters to define two new titratable groups?
> How difficult would this be?
>
> Thanks very much for your help
> Many regards
> Sarah Harris
>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu