COM motion removal during NVT simulation with electric field

User discussions
,
1

Hi,

I am studying the motion of a positively charged protein in the presence of electric field at 300 K. I have observed a conflict between maintaining the physical drift of the protein under the influence of electric field and preventing the COM motion of the system.

I have performed simulations with and without the COM removal: (a) comm-mode = Linear and (b) comm-mode = None.

The following are the observations:

  • In case (a), the displacement of the protein is reduced, ~20 nm, implying that the COM removal is fighting the field-induced motion.

  • In case (b), the protein covers ~ 300 nm in the same time interval. But the COM of the system experiences a drift, (i.e.) COM velocity of the system along the direction of the electric field is ~ 0.009 nm/ps.

  • However the temperature of the protein and the system are at 300 K in both the cases.

Question:
What is the recommended approach to perform NVT simulation, of such a system, in the presence of electric field while preventing the artifacts due to the COM motion?

I would greatly appreciate any insights or references to similar studies.

Thank you.

Best regards,
Afi

2

I don’t know what scientific question you are trying to answer. If you don’t, you need to figure that out.

Putting an infinite periodic array of charged proteins with very short separation in between in an electric field is a very artificial situation.

3

Dear Prof. @hess,

Thank you for your swift response.

To clarify, the ultimate objective is to study protein translocation through a solid-state nano channel under the influence of electric field.

The system mentioned, a protein in water and ions, is a simplified model to understand the effect of COM motion removal on the movement of protein.

Best regards,
Afi

4

When you have a nanopore, I assume you want to have that as a fixed reference, so you would remove only the COM of that or restrain it. Running without the pore is not relevant.

5

Dear Prof. @hess,

Thank you of your response and for pointing that out.

The nanopore/channel is indeed position restrained.

However even in the simulations involving the entire system - containing the restrained channel, protein and ions - I still observe discrepancy on the protein behaviour depending on whether comm-mode is turned on/off. For additional context on my setup, I am using the v-rescale thermostat with three separate temperature coupling groups: Protein, Ions+Sol, Nanochannel to maintain the temperature.

Therefore, I am still in a dilemma regarding which comm-mode setting would accurately capture the physical electrophoretic drift while simultaneously mitigating the numerical drift in the center of mass.

Any advice or recommendations on how to move forward with this setup would be highly appreciated.

Sincerely,
Afi

6

When you restrain part of the system, you do not want to remove COM motion. There is no numerical drift when part of the system is fixed. And removing COM motion will introduce strong artifacts.