AMBER Archive (2009)

Subject: Re: [AMBER] Zinc ion, help

From: FyD (fyd_at_q4md-forcefieldtools.org)
Date: Wed Jul 08 2009 - 04:43:26 CDT

Dear Ibrahim,

> I am working on metalloproteins (MMP) and they have Zinc ion in its
> active site. As you know, Zinc ion parameter is a problem. In the
> following paper:
> http://www.mmvsl.farm.unipi.it/downloads/free-downloads/amber-ff-parameters-for-the-zinc-hydroxammate-interaction/
> the author derived the partial charges depending on a model of Zinc
> ion with three methylimidazoles and one acetate ion. The three
> imidazoles here take place of three histidine residues in the active
> site.
> Sorry, I am a little confused. According to AMBER manual, we should
> cap the cut-off residues with NME or ACE in order to maintain the
> overall charge, i.e. each histidine residue should be capped with
> two NME (or ACE). By this, three imidazoles model will not give an
> accurate charge.
> Am I right? I am so sorry, I feel I misunderstand something. I need
> a help, please.

In this type of work, you could follow different approaches:

-1- The first one is as you said 'use a capped amino-acid for each
residue involved in your complex'. In this case, you 'simply' have to
set an intra-molecular charge constraint set to zero during the
fitting step for each capping group to make your complex fully
compatible with the other force field libraries/parts of your system.

-2- Another approach would be to only use each amino-acid side chain
(this is what is done, I guess). Here you need to find other charge
constraints to make your complex compatible with the other parts of
your system. The big advantage here is that your number of atoms
involved in QM computation is smaller than in the case -1-. However,
the charge constraints to be used might be less obvious; I mean if you
want to avoid manual 'adaptations'.

regards, Francois

_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber