[gmx-users] certain number of solvent molecules with genbox WITHOUT "ci"/"nmol" options

[Justin A. Lemkul]

Vedat Durmaz wrote:
> 
> Am 09.01.2012 12:37, schrieb Mark Abraham:
>> On 9/01/2012 9:00 PM, Vedat Durmaz wrote:
>>> hi guys,
>>>
>>> i have a molecular system that i want to simulate many times with 
>>> differing initial conditions, but each with the same number of 
>>> (solvent) atoms! i'm using gromacs 4.5.4 ...
>>
>> Use different values for gen_seed. By the time you equilibrate the 
>> positional correlations will be lost.
>>
>> Mark
>>
> thanks for your help, mark. but you didn't get the point. let me try to 
> describe it again:
> 
> the positions of the solvent atoms are not important. i just need the 
> same number of solvent molecules within any of the 50 systems. the 
> differences of these systems are already given by different 
> conformations of the ligand.
> 
> 
> 
> as written below, genbox -cs doesn't yield in the same number of solvent 
> molecules for the 50 occurrences of the same system. in contrast, 
> "genbox -ci ... -nmol 23000" would result in the same number of solvent 
> molecules, however it runs into the error "Cannot allocate memory" 
> having reached memory limit (which has 40 GB size!).
> 
> what i would like to know is whether there is a way of telling genbox to 
> put a certain number of solvent molecules without using the -ci option 
> (which requires more memory than my system offers).
> 
> if there is no other way, i will have to change each of the 50 
> topologies and pdb files manually in order to keep an equal number of 
> solvent molecules for each system.
> 

Use the -maxsol option to limit the number of water molecules to some specified 
value.  In conjunction with the normal -cs, the program is limited to whatever 
value you specify.

-Justin

>>>
>>> genbox with "-cs" gives me different numbers of solvent molecules.
>>>
>>> putting more than 20,000 tip4p molecules with the "-ci" and "-nmol" 
>>> combination of genbox instead runs into memory problems (40 GB is not 
>>> sufficient for one single system!).
>>>
>>> does anyone know another way of solving that issue avoiding "manual" 
>>> straightening?
>>>
>>> thanks and take care
>>>
>>> vedat
>>

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

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

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