Melanie - Your system is pretty compact, so I would not expect you to see
any air bubbles if you equilibrate your system well. The general idea
would be;
1. First create a system (without water) with a reasonable structure
(get rid of bad contacts).
2. Solvating the structure with some solvent + ions.
3. Minizing the solute+solvent system in case there are still bad
contacts. For a system like yours, I would not expect any bad contacts at
this stage if you have used leap to add water and ions, but it is not a
bad idea to minimize the solute+solvent system in 2 steps; i) Minimizing
only the water molecules using positional restraints on solute, ii)
minimizing the solute+solvent all together. Positional restraints
basically means freezing the coordinates of atoms. You do not want to use
positional restrains in and MD run, but it is a useful way in order to
prepare/equilibrate the system. Otherwise during minimization and/or
equilibration, the initial structure of the solute (which would resemble
the experimental structure) might get lost. In case positional restraints
would not work, you can use NMR distance/angle/torsional restraints on the
system to keep it close to the experimental structure during
minimization/equilibration (Check out the tutorials).
3. Equilibrating the system: Again, for a system like this, equilibration
would not be hard. First, use positional restraints on solute and increase
the temperature from 0 to 300 K gradualy using NVT. This will add
velocities to the water molecules. Then, run another MD in NPT without
using any restraint on the system. This will decrease the box size and
will create a pressure around 1 atm. The density would be equilibrated,
too. You can take a look at the tutorials for reference.
4. Start your production runs using NPT at 300 K (if you want to do MD at
another temperature, use that temperature in both equilibration and
production runs). For a system like this, I would guess that the system
will be equilibrated pretty soon.
If you use constant pressure dynamics (NPT), you should not see air
bubbles. Now the question is this: Would the MD simulations sample
physical or unphysical regions in the conformational space? I am not a
protein person, so people who had experiences on this type of systems can
comment more on it. Your system is a unique system; some sort of a cyclic
peptide system. As a result, the force field can do a good or bad job on
sampling. If you have enough computer time, I would suggest you to do 10
or more independent simulation on the structure with different random seed
numbers. At the end, you can combine your independent simulations to get a
feeling on the prediction.
What I wrote above assumes that you just want to do MD on this system to
get some structural prediction. If you want to see/analyze some
conformational transitions, there are methods in AMBER package you might
want to use.
Hope this helps. Good luck.
Best regards,
Ilyas Yildirim, Ph.D.
-----------------------------------------------------------
= Department of Chemistry - 2145 Sheridan Road =
= Northwestern University - Evanston, IL 60208 =
= Ryan Hall #4035 (Nano Building) - Ph.: (847)467-4986 =
=
http://www.pas.rochester.edu/~yildirim/ =
-----------------------------------------------------------
On Mon, 4 Apr 2011, Melanie wrote:
> Hi Ilyas,
> I'm trying to do both minimization and MD. I'm trying to do this in explicit
> solvent (water), because the protein is solvated under normal conditions.
> Thus, the minimization is done to try and compact the water molecules and
> get rid of air bubbles. I'm not sure if that is what I am actually
> accomplishing, but that is the general idea.
> Since I will be doing MD, how do I set those restraints? Regarding the
> disulfide bond, that was fixed by changing the CYS residues to CYX, so that
> error was resolved.
>
> Thanks,
> Melanie
> PS. Sorry about the name confusion there; my coworker was helping me send my
> emails, and accidentally signed her name in her hurry.
>
> On Mon, Apr 4, 2011 at 7:23 AM, Ilyas Yildirim
> <i-yildirim.northwestern.edu>wrote:
>
>> Hi Audrey -
>>
>> What kind of calculation do you want to do on this structure? Just a
>> minimization or MD? It is really difficult to find the best parameter set
>> from analogy, but I would guess that it will not make a too big difference
>> for this structure. Generally, minimization is done to get rid of close
>> contacts. When I minimize nucleic acids in gas phase, for instance, I will
>> see some distortions on the planes but they will vanishe in an MD run. The
>> same thing would probably happen in your case. Also, how do you minimize the
>> structure? In vacuum or using implicit solvent model? You might want to try
>> to minimize the structure using one of the implicit solvent models
>> implemented in AMBER if you did not try already. That might resemble the
>> real environment more than gas phase.
>>
>> If the aim of the minimization is to prepare the best coordinates for the
>> MD simulations, you might want to put some restraints on the structure (this
>> can be positional and/or NMR distance/angle/torsional restraints).
>>
>> Another thing - I do not see any parameters defined for the sulphur region
>> in the frcmod file (resid 1 and 9). After creating the bonds did leap give
>> any error message for that sulphur bond?
>>
>>
>> Best regards,
>>
>> Ilyas Yildirim, Ph.D.
>> -----------------------------------------------------------
>> = Department of Chemistry - 2145 Sheridan Road =
>> = Northwestern University - Evanston, IL 60208 =
>> = Ryan Hall #4035 (Nano Building) - Ph.: (847)467-4986 =
>> = http://www.pas.rochester.edu/~yildirim/ =
>> -----------------------------------------------------------
>>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Apr 04 2011 - 23:00:03 PDT
