[gmx-users] »Computational Electrophysiology«

[Kutzner, Carsten]

Hi Harutyun,

> On 3. Apr 2019, at 00:08, Harutyun Sahakyan <sahakyanhk at gmail.com> wrote:
> 
> Dear Gromacs users,
> 
> I am trying to simulate membrane potential using computational
> electrophysiology protocol.
> 
> I modeled protein-membrane system with charmm-gui, equilibrate it and run
> MD simulation. After some steps, I get this error.
> 
> 
> =======================================================================
> starting mdrun 'Title'
> 50000000 steps, 100000.0 ps.
> SWAP: Performed 35 swaps in step 100 for iontype POT.
> SWAP: Performed 14 swaps in step 100 for iontype SOD.
> step  200: timed with pme grid 72 72 108, coulomb cutoff 1.200: 2495.3
> M-cycles
> 
> SWAP: Warning: 12 atoms were detected as being in both channels! Probably
> your split
> SWAP:          cylinder is way too large, or one compartment has collapsed
> (step 200)
> SWAP: Performed 1 swap in step 200 for iontype SOD.
> 
> SWAP: Warning: 8 atoms were detected as being in both channels! Probably
> your split
> SWAP:          cylinder is way too large, or one compartment has collapsed
> (step 300)
> SWAP: Performed 1 swap in step 300 for iontype POT.
> SWAP: Performed 1 swap in step 300 for iontype SOD.
> step  400: timed with pme grid 64 64 100, coulomb cutoff 1.327: 2610.4
> M-cycles
> 
> SWAP: Warning: 8 atoms were detected as being in both channels! Probably
> your split
> SWAP:          cylinder is way too large, or one compartment has collapsed
> (step 400)
> SWAP: Performed 2 swaps in step 400 for iontype POT.
> SWAP: Performed 2 swaps in step 400 for iontype SOD.
> 
> SWAP: Warning: 9 atoms were detected as being in both channels! Probably
> your split
> SWAP:          cylinder is way too large, or one compartment has collapsed
> (step 500)
> SWAP: Performed 1 swap in step 500 for iontype POT.
> SWAP: Performed 2 swaps in step 500 for iontype SOD.
> step  600: timed with pme grid 56 56 84, coulomb cutoff 1.538: 3583.0
> M-cycles
> 
> SWAP: Warning: 8 atoms were detected as being in both channels! Probably
> your split
> SWAP:          cylinder is way too large, or one compartment has collapsed
> (step 600)
> SWAP: Performed 2 swaps in step 600 for iontype POT.
> SWAP: Performed 1 swap in step 600 for iontype SOD.
> 
> SWAP: Warning: 6 atoms were detected as being in both channels! Probably
> your split
> SWAP:          cylinder is way too large, or one compartment has collapsed
> (step 700)
> SWAP: Performed 3 swaps in step 700 for iontype POT.
> 
> -------------------------------------------------------
> Program:     gmx mdrun, version 2018.1
> Source file: src/gromacs/swap/swapcoords.cpp (line 1911)
> 
> Fatal error:
> Could not get index of SOL atom. Compartment contains 3 SOL molecules before
> swaps.
> 
> For more information and tips for troubleshooting, please check the GROMACS
> website at http://www.gromacs.org/Documentation/Errors
> ==================================================================
> 
> 
> 
> The number of steps before the error depends on ion concentrations in
> compartments A and B. Moreover, the number of SOL molecules in the error
> message can be different as well.
The thing is, that there should be *way* more than 3 SOL molecules in each
compartment, otherwise something is severly broken. How many waters do you
have in the single-membrane system that you built with charmm-gui? This is
the number of waters that should be in each of the two compartments after
you build your double-membrane system (e.g. by applying the makeSandwich
script found at https://www.mpibpc.mpg.de/grubmueller/compel).

> 
> I have tried to change cylinder parameters however it does not solve the
> problem.
Note that the definition of the cylinders just serves to count how many
ions go through each of the two channels, but it does not affect whether
ions are swapped or not.

> 
> Do you have any ideas? Have you faced this error before?
> 
> Channels are defined as several amino acids in the central parts of the
> channel.
Please check the output .xvg - it should list the z-positions of the
two channels, which define the compartment boundaries. Do these make
sense if you look at the system in Pymol or VMD? Do they partition the
simulation box in about equal volumes (in PBC)? 

> 
> I attached my md.mdp file for more detailed information.
Didn’t go through, attachments won’t work but you could paste the 
CompEL .mdp parameter section in here directly.

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