AMBER Archive (2009)

Subject: Re: [AMBER] Question on heating for homology based model.

From: Carlos Simmerling (
Date: Sun Feb 01 2009 - 11:47:02 CST

the temperature of crystallization isn't really relevant, you still need
heating because it is an average structure, and likely minimized in a force
field during structure determination. also, you need a very careful
procedure when you equilibrate a structure that does not come from an
experiment. in your case, you likely should restrain all "known" parts while
fully equilibrating those that you built in, such as loops or side chains.
only after these are fully relaxed should you begin to allow the parts that
match the known structure to significantly move.

whether you only minimize or do full MD depends a lot on what you want the
model for, and whether the aprts important to you changed from the template.
there's really no way for us to answer that for you, you need to think about
your goals and use judgement based on reading the literature about other
peer-reviewed work that is similar to what you want to do.

On Sat, Jan 31, 2009 at 1:43 PM, drugdesign <> wrote:

> Dear Amber users,
> I've made a homology based model using template structure with high
> resolution(80% identity with template structure), which was crystallized
> under 298K. Do I need a heating for it from 0 to 300K while making
> molecular dynamics simulation? How much can be difference with and without
> heating for such a model? Do I need to make molecular dynamics at all or
> just to make local minimization?
> Best regards,
> Alex
> _______________________________________________
> AMBER mailing list
AMBER mailing list