[gmx-users] Error with MARTINI depending by the box size

[Justin Lemkul]

On 12/10/12 5:45 PM, francesco oteri wrote:
> For Justin,
> I need this water for one simple reason: less then 20nm doesn't work....As
> I said before
>

It seems you have identified the source of the problem, and it is independent of 
box size.  I questioned the box size because it seemed rather random and you had 
not shown any data for box sizes less than 19 nm, so I was curious how you 
arrived at the need for 20 nm, more than double the size of your solute.

It would be interesting to see if you could identify a minimum box size that 
does not require large numbers of solvent configurations to be stacked within 
the unit cell.  The only reason I could see for what you're reporting is if 
neighboring solvent blocks somehow get crossed to produce overlap when they 
should simply be next to one another.  The larger the box, the greater the 
probability that this happens.

>
> 2012/12/10 francesco oteri <francesco.oteri at gmail.com>
>
>> Hi Mark,
>> you are right respect the -vdwd 0.4: In MARTINI tutorials they suggest to
>> use 0.21. Since I still got errors with this procedure, I decided to remove
>> water manually through vmd.
>>
>> Looking carefully at the configurations, I found that the water molecule
>> originating the error is exactly superimposed to an other molecule so I
>> simply deleted it and the same error is reported for an other water
>> molecule.
>>
>> Although I could scan all the pdb to detect all the superimposed water
>> molecules, I believed genbox checked for this. Of course the original box
>> does't contain superimposed molecules.
>>
>>

It is highly unusual for genbox to produce overlapping waters, but per the help 
description, only solute-solvent overlaps are checked, not solvent-solvent, 
which would likely require enormous amounts of memory (and genbox already has 
memory issues).

-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

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