normal AdResS of an atomistic protein in MARTINI

  • chenyuetian
  • chenyuetian's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
5 years 2 months ago #3601 by chenyuetian
AdResS of an atomistic protein in MARTINI was created by chenyuetian
Hi,
I am very impressed with the paper you recently published: Adaptive resolution simulation of an atomistic protein in MARTINI water. Like you said in the paper: "Your approach enables the study of biophysical phenomena that are beyond the scope of either atomistic or mesoscopic simulations."
I understand the philosophy behind the method, but I need to know more details to apply the Adaptive
Resolution Scheme (AdResS)method with MARTINI. Thus, I am asking:
(1) If there are appropriate tutorials introducing this method combining with MARTINI using the ESPResSo++ software package.
(2) Once get stuck, can I post my problem in this forum for help.
(3) If the method could be used for the study of interfacial property. i.e. a MD simulation of protein near an air/water or oil/water interface. I understand your study may not reach this area, but could you help to estimate the possibility here?
(4) You newest progress including extension of this methodology to polarizable CG water models and salt solutions.

Please Log in or Create an account to join the conversation.

More
5 years 2 months ago #3602 by mnmelo
Replied by mnmelo on topic AdResS of an atomistic protein in MARTINI
Hi, and thanks for interest in the work.

Currently there are two implementations that can achieve the AdResS multiscaling we describe in the paper: the ESPResSo one and a more recent gromacs one. I have worked more with the gromacs implementation and can guide you there. We are actually in the process of publishing a paper describing solvent properties, using either multiscaling implementation.

As far as I know there aren't many tutorials around. Do check the votca package and its documentation, which comes in handy for calculating the thermodynamic force. In any case, I'll be glad to help.
While there should be no problem simulating systems with interfaces, note that so far AdResS can only be used with solvent molecules that are mapped to a single bead; this might limit your applications

Good luck!
Manel

Please Log in or Create an account to join the conversation.

  • chenyuetian
  • chenyuetian's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
5 years 2 months ago - 5 years 2 months ago #3605 by chenyuetian
Replied by chenyuetian on topic AdResS of an atomistic protein in MARTINI
Thanks for the quick reply Manuel,
Now I am glad to know the AdResS has been implemented in gromacs, and I wish to know if you could provide more info. about this gromacs implementation. e.g. if the gromacs you are using has been modified, could you provide the modified version for me? I can try to compile and install it, but I don't know how to modify gramacs' source code. Meanwhile, could you provide me the top, mdp file etc. (in gromacs format) that can be used to reproduce the simulation in your recently published paper?
There is a concern about the interface simulation is what will be happen if the protein in high resolution region sphere approaches to the hydrophobic/hydrophilic interface. In my current simulation with either all-atom or martini force field, the protein is observed binding at the interface. Is this highly impossible to see this binding in the Adaptive resolution simulation? If it is possible, I think the shape of atomistic water surrounding the protein might not be sphere.
I got the supplement material.
You can reach me here or at This email address is being protected from spambots. You need JavaScript enabled to view it.
Best regards,

Yuwu
Last edit: 5 years 2 months ago by chenyuetian.

Please Log in or Create an account to join the conversation.

More
5 years 2 months ago #3606 by mnmelo
Replied by mnmelo on topic AdResS of an atomistic protein in MARTINI
Hi again,

I'm sorry I was not clearer: the work in the paper was done only in ESPResSo. I've further worked with gromacs and it should work there as well, provided you modify it a bit (I can provide you the modified code).

To do what you need you'll need first of all to determine the thermodynamic force to counteract density imbalances introduced by the resolution transition (read the AdResS papers). I have done it for water in gromacs using the votca package. The results will be published soon.

Also, as mentioned in the paper, you'll need some sort of bundling model to keep the waters together (assuming a multiple-to-1 mapping, like Martini), otherwise they drift apart and the CoM becomes meaningless. I've calculated the thermodynamic force for such a bundled water using gromacs and the votca package, but recommend you'd do it as well since I used a slab geometry instead of spherical and I expect it to be quite dependent of system geometry.

What would happen should the protein wander too close to the hybrid region is implementation specific. In gromacs, at least as far as I'm aware, it'll interact normally with the atomistic part of the molecules there. This sounds ok but it really isn't: you should expect the solvent structure in the hybrid region to be unphysical and interactions with it will probably lead to erroneous results.

Getting this all to work can be a bit tricky. I'll send you some relevant files in the next couple of days as well as some tips. Maybe after helping you I'll post a tutorial of sorts.

Please Log in or Create an account to join the conversation.

  • chenyuetian
  • chenyuetian's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
5 years 2 months ago #3607 by chenyuetian
Replied by chenyuetian on topic AdResS of an atomistic protein in MARTINI
Hi,
Yes, I noticed the work in your J Chem. Phy. was done only in ESPResSo. I want to reproduce the work by myself in gromacs not because of question your work in gromacs, but because if I can do it and make the results in gromacs comparable to your work, that means somehow I gained some knowledge about the basic ADResS method in gromacs. My idea is just that the learning process may involve some necessary practice. So really appreciate if you could send me any files and instructions.
Since the very first thing I need to know is the thermodynamic force calculation. I would like to spend more time on the Votca package since now, while reading other AdResS papers from your group.
Please keep in touch.
Best regards,

Please Log in or Create an account to join the conversation.

More
4 years 7 months ago #4092 by agoliaei
Replied by agoliaei on topic AdResS of an atomistic protein in MARTINI
Dear Manel,

I am very interested in this paper and was wondering if you could kindly provide the modified gromacs code to me as well. Any sample top or mdp file for a first test would be also greatly appreciated.

Thanks a lot for sharing this.
This email address is being protected from spambots. You need JavaScript enabled to view it.

Please Log in or Create an account to join the conversation.

More
4 years 7 months ago - 4 years 7 months ago #4093 by mnmelo
Replied by mnmelo on topic AdResS of an atomistic protein in MARTINI
Dear agoliaei,

The information you request is already online, at cgmartini.nl/cgmartini/index.php/example...ations2/multiscaling

EDIT: The files I pointed to pertain to the AdResS Martini solvents paper , but the patch in that link is the modification you're looking for.

Good luck,
Manuel
Last edit: 4 years 7 months ago by mnmelo.

Please Log in or Create an account to join the conversation.

Time to create page: 0.218 seconds