[gmx-users] Umbrella sampling

[Justin Lemkul]

On 8/10/12 1:14 PM, Paula Andrea Delgado Pinzon wrote:
> Good morning,
> I am trying to run an umbrella sampling simulation, my system is a complex of two proteins.  I have a couple of  questions:
> 1. i don't know which is the purpose in defining the charge of each protein end

A net charge is a consequence of the constituent residues of a protein.  The net 
charge is defined in the topology as a matter of bookkeeping.

> 2. When i check the topol.itp file of the reference molecule i realized that it already has
>
> #ifdef POSRES_B
> #include "posre_B.itp"
> #endif
>
> and when i check the file of the other protein it has it too

If you're trying to draw an analogy to my umbrella sampling tutorial, be 
careful.  Position restraints in that case are introduced for a very specific 
reason (see our associated paper) and are not generally necessary for umbrella 
sampling.

> 3. I don't know how to decide the size of the unit cell and which is the criterion to center the molecules

The unit cell must be of sufficient size to satisfy all of the following:

(a) The minimum image convention is always satisfied
(b) The restraint distance is less than half the shortest box vector

> 4. I am not sure why in the equilibration step and NVT it hasn't been done

Is this again in reference to the tutorial?  My system represents a very robust 
system for which a single NPT equilibration phase is sufficient.  This is not 
necessarily true for other systems.  Equilibrate as you see fit.

> 5. In the neutralization part in this kind of simulation ones has to specified the concentration of the solution or i can do it just with the number of ions i want to add
>

You can set whatever amount of ions you like.  A net neutral system is generally 
required.  Additional ionic strength can be added depending upon the nature of 
the system you wish to model.

-Justin

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================