[gmx-users] Computational electrophysiology (compEL) setup issues

[Kutzner, Carsten]

Dear Francesco,


> On 27. May 2019, at 16:31, Francesco Petrizzelli <francescopetri at hotmail.com> wrote:
> 
> Dear gmx users,
> 
> I'm trying to simulate change in the ion flux upon mutations using Computational electrophysiology (compEL).
> I have used packmol-memgen in order to generate the double bilayer using a salt concentration of 1 M (3216 Cl- and 3168 K+).
It is even easier to just generate a single bilayer with a channel, and then duplicate
the whole system by stacking two copies on top of each other, e.g. with the script 
found on page www.mpibpc.mpg.de/grubmueller/compel under 
www.mpibpc.mpg.de/15388676/makeSandwich.tgz

> Then, I have setted up the simulation following the compEL protocol. However, after 50 ns of simulation, I still have two big issues:
> 
> 1) In the swapions logfile I've got a huge number of warnings regarding ions that moved from Domain_B to Domain_A probably throught the membrane (checking the movie with VMD it seems they move across the boundaries). I thought to tackle this issue by increasing the water layer.
The .xvg output file also lists the z-position of your split group (=channel) centers over
time. Can you check in a molecular viewer whether these make sense? They should be in the
middle (regarding z coordinate) of each of the two membranes. This is important, because these
z-positions define the compartment boundaries.

A cyl0-r (and cyl1-r) of 0.7 nm is too small for a pore radius of 3 A to reliably track the ions, 
some might sneak through your channel without being recorded in the cylinder. Rather choose this 
value a bit too large than too small. It should be *at least* half of the pore radius.

> 2) In the same way, watching  the movie I have notice that ions enter the channel (the pore radius is about 3 A) but they do not cross from one domain to the other. I think I may have misunderstood the protocol, because in numerous article they talk about  setting an ionic imbalance between compartments. However, I have not performed it in any step. Do I have to manually set an ionic imbalance (and, in this case there are a tool to perform it?) or it should be set by the input compEL file? I
> ; Names of the ion types that can be exchanged with solvent molecules
> ; -1 means fix the numbers as found in time step 0. These numbers have to add up to the total number of ions present in the swap group.
> iontype0-name = Cl-_Cl-
> iontype0-in-A = -1   ; requested number of Cl ions in compartment A
> iontype0-in-B = -1   ; requested number of Cl ions in compartment B
> iontype1-name = K+_K+
> iontype1-in-A = -1   ; requested number of K+ ions in compartment A
> iontype1-in-B = -1   ; requested number of K+ ions in compartment B
You set -1, so if ions move through the channels, the protocol restores the numbers found
at the start of the simulation, which may or may not give you an ionic imbalance
and a voltage across the membrane. 

You need to put positive numbers there, which add up to the actual number of
ions in the simulation, e.g. let’s say you have a total of 20 Cl- and 24 K+ :

iontype0-name = Cl-
iontype0-in-A = 10  # means 10 Cl- in compartment A
iontype0-in-B = 10  # means 10 Cl- in compartment B as well
iontype1-name = K+
iontype1-in-A = 12  # 13 K+ in A
iontype1-in-B = 12  # 11 K+ in B

This will set and keep an imbalance of two elementary charges between the compartments, that
will be set at the first time step by exchanging an appropriate number of ions and water
between A and B.

Best,
  Carsten


--
Dr. Carsten Kutzner
Max Planck Institute for Biophysical Chemistry
Theoretical and Computational Biophysics
Am Fassberg 11, 37077 Goettingen, Germany
Tel. +49-551-2012313, Fax: +49-551-2012302
http://www.mpibpc.mpg.de/grubmueller/kutzner
http://www.mpibpc.mpg.de/grubmueller/sppexa