From: Vermaas, Joshua (
Date: Mon May 13 2019 - 13:35:32 CDT

One method where I've had some success determining "inside" and "outside" are to use algorithms in computer vision applied to water densities. That can define an "inside" and an "outside", which can be converted into a volumetric density field that you can use to select atoms with (eg volume > 0 for inside and <0 for outside). Its a fun problem, but not one where I'm aware of a built-in solution.


On 2019-05-13 12:15:19-06:00 wrote:

Hello -
Depends on what type of micellar molecules you have, but generally there is a head group and a tail group. If the micelles are stable on your simulation timescale, head groups should always be positioned to be lining the interior, so you could develop distance based selection criteria for counting waters that are touching head groups, (first internal layer) and then the waters that are touching those (second internal layer), and so forth until you've counted all the waters on the inside. I suppose it will be sensitive to tail length, lipid dynamics, and water infiltration though.

On Mon, May 13, 2019 at 12:56 PM Nick Palmer <<>> wrote:
Hello everyone,
I am simulating two reverse micelles and I need to be able to count the number of water molecules that are in each reverse micelle and out in the solvent for each frame of the simulation. What would be the best way to determine whether or not a water is inside one of these reverse micelles or not? I had tried to find all the waters within a certain amount of distance of the center of mass, but this only works when the two reverse micelles are not next to each other. Also, they don't stay the same shape the whole time so I would have to change the distances at every frame. Is there a way I could use the SASA of the reverse micelle to define a layer, so that I could count the number of water molecules inside and outside of each reverse micelle?
Another issue I have been having is that sometimes the reverse micelles move across periodic boundary conditions, which seems to shift the center of mass for the micelles. How could I use pbc tools to solve this problem since these molecules are not directly attached to each other?
Thank you very much in advance!

Nicholas J. Palmer