Re: Basic question.

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Sat May 27 2006 - 15:20:56 CDT

The answer to this is really dependent on your system. The fundamental
purpose of equilibration is to get rid of any artifacts from your
initial system setup, so you just need to run this until your system
behvior stabilizes (this can be judged, for example, by water diffusion
beginning to follow the diffusion law, or protein RMSD stabilizing).
Around 100ps is generally adequate for this, but you should check for
your particular system.
Best,
Peter

Taeho Kim wrote:

>Thank you peter for you reply..
>
>If so, how long is it proper time for equilibration?
>Now, I running 1ns and 1.5ns for equilibration, but it is very long I
>think..
>
>
>
>Peter Freddolino ¾´ ±Û:
>
>
>>Hi Taeho,
>>to answer (2) first, equilibration is a short molecular dynamics
>>simulation designed to relax the initial configuration; for example, it
>>lets dipoles orient properly and water find more appropriate positions.
>>The only technical difference is that it is sometimes carried out with
>>restraints on part of the system (often the protein, for example). This
>>step is necessary to avoid any artifacts that might show up from
>>starting to take data from a system that is going to undergo quick
>>relaxations such as this. I would highly recommend to you a textbook on
>>molecular modelling techniques (such as Leach's /Molecular Modelling:
>>Principles and Applications/) to have a look at the need for this step
>>in more detail.
>>Best,
>>Peter
>>
>>Taeho Kim wrote:
>>
>>
>>
>>
>>>Dear all.
>>>
>>>I have some basic question.
>>>
>>>1. Why do we have to do equilibration step?
>>>2. What is different between equilibration and molecular dynamics?
>>>
>>>I have stuied about that by using Namd tutorial. I couldn't understand
>>>clearly.
>>>
>>>Thank you.
>>>
>>>Best regards...
>>>
>>>
>>>
>>>
>>>
>>
>>
>>
>>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:43:38 CST