Re: minimization temperature and constraints

From: Michael A. Wilson (mwilson_at_mail.arc.nasa.gov)
Date: Tue Dec 05 2006 - 18:55:24 CST

(1) From a cynical point of view, we keep proteins fixed while doing various
things to equilibrate the rest of the system so as to give the protein every
opportunity to behave in precisely the manner we wish it to behave...

 From a more practical view, you're usually starting with a crystal or NMR
structure, or someone's best guess as to the structure of the protein. When you
construct water/protein or water/membrane/protein systems, you usually insert a
protein into a hole that you have created in an equilibrated water or
water/membrane box. Byt the resulting configuration is going to be a system that
is far from equilibrium. If you simply try to start MD from this configuration,
bad contacts between the protein and the rest of the system can lead to extreme
local heating that will cause the protein to unfold before the system has time
to equilibrate. Since (a) this local heating is not really something the protein
would likely experience out in the wild and (b) MD timescales are not usually
adequate to re-fold the protein (if it would re-fold), equilibrating the rest of
the system to the protein structure is the the best way to start.

Basically, you're holding the protein fixed, at first, because you believe that
stucture to be relevant, so you want to create a system that is equilibrated
subject to the condition that it contains that structure.

(2) It probably doesn't matter. Especially if you're just setting up a box of
water. I think its more typical to run NPT to equilibrate things and run NVT
once you're happy that that box size doesn't have any systematic drift.

Mike

AYTUG TUNCEL wrote:
> Dear all,
>
> I have some basic questions about minimization procedure,
>
> 1) What is the difference between keeping the protein (at least the CA atoms) fixed or relaxed during minimization? For ex. What is the purpose of keeping the portein fixed while heating up the system (say from 0K to 310K)?
>
> 2)If I want to provide a uniform density for water molecules in the box, should I couple NPT with heating up process or do the NPT after reaching the desired temperature (310K) ?
>
> 3)In NAMD usersguide there is this parameter "reinitvels". What is the difference between using this parameter or manualy increasing the temperature (heating up the system) after minimization?
>
> Thanks in advance.
>
> Aytug
>
>
>

-- 
  _____________________________________________________________________________
  Mike Wilson, MS 239-4                                      (650) 604-5496 day
  NASA Ames Research Center                                  (650) 604-1088 fax
  Moffett Field, CA   94035

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:54 CST