[gmx-users] Center-of-Mass motion removal setting for Membrane system with multiple peptides.

Wed Sep 28 16:17:46 CEST 2016

Periodic systems should be translation invariant, and diffusion across periodic boundaries normally is not a concern. There are occasional exceptions to this, mostly (only?) because of technical implementation details. Unless your system warrants special treatment (and from what you’ve told us so far I don’t think that’s the case) I’d say that separate coupling of COM removal would only add artefacts. Take the extreme example of separate COM-motion removal of a single Na+ and a single CL-. They would be locked in place throughout the simulation. The same effect would come into play also for a group of Na+ and a group of Cl- with separate COM-motion removal, albeit in a more subtle way. What you suggest is similar.

Kind regards,
Erik

> On 27 Sep 2016, at 11:33, Abhi Acharya <abhi117acharya at gmail.com> wrote:
> 
> Actually, I am really not sure whether diffusion out of the box will affect
> the results or not. As long as the system is able to correctly simulate the
> intended biological phenomenon i.e. interaction of peptides with membrane,
> it is fine. Intuitively, one may say that for a semi-isotropic membrane
> system diffusion along XY direction should be fine; but any diffusion along
> Z would not be acceptable. Not sure, how this can be correctly handled.
> 
> There have been studies like this with other peptides, but none of them
> have commented on the above aspect.
> 
> On Tue, Sep 27, 2016 at 2:36 PM, Erik Marklund <erik.marklund at kemi.uu.se>
> wrote:
> 
>> Hi Abhi,
>> 
>> No, restricting the COM motion of the entire system is perfectly fine in
>> most cases. From the conservation of momentum, the COM should not change
>> its velocity at all. One reason the COM motion needs to be kept at bay
>> explicitly is because the accumulation of numerical error during the
>> simulation. If you separately remove the COM motion of separate groups you
>> will however affect the diffusion of one group relative to the other, for
>> example. Is diffusion out of the simulation box really a concern in your
>> case?
>> 
>> Erik
>> 
>>> On 27 Sep 2016, at 10:40, Abhi Acharya <abhi117acharya at gmail.com> wrote:
>>> 
>>> Another thing which I don't understand is in case of the peptide group,
>>> which I expect to diffuse freely, would COM motion removal be non
>> physical?
>>> The starting system has 16 peptides added to one side of the membrane.
>> Now
>>> to allow the peptides to diffuse freely and interact with the membrane, I
>>> would assume that translational, rotational and conformational freedom
>>> would be required. Wouldn't restricting the COM restrict its dof?
>>> 
>>> Thanks.
>>> 
>>> On Tue, Sep 27, 2016 at 1:59 PM, Abhi Acharya <abhi117acharya at gmail.com>
>>> wrote:
>>> 
>>>> What I meant to ask was a way to ensure that the peptides and membrane
>> COM
>>>> don't drift out of the simulation box, but the peptides should be free
>> to
>>>> move ( relative to the membrane) within the box. Basically, the best
>> way to
>>>> simulate the diffusion and subsequent interaction of the peptides with
>> the
>>>> lipid membrane.
>>>> 
>>>> What I can surmise from previous similar studies is that creating
>> separate
>>>> comm groups for Membrane, solute and ion and the peptide should be the
>>>> correct way. Also, I wanted to know how different nstcomm values would
>>>> effect the result especially in the context of complex systems such as
>>>> this.
>>>> 
>>>> Just wanted to be sure, before I start the production runs.
>>>> 
>>>> 
>>>> 
>>>> On Tue, Sep 27, 2016 at 1:01 PM, Erik Marklund <
>> erik.marklund at kemi.uu.se>
>>>> wrote:
>>>> 
>>>>> 
>>>>>> On 27 Sep 2016, at 06:26, Abhi Acharya <abhi117acharya at gmail.com>
>>>>> wrote:
>>>>>> 
>>>>>> Dear Gromacs users,
>>>>>> 
>>>>>> I am trying to perform a simulations of different concentration of
>>>>> peptides
>>>>>> in a box with lipid bilayer. In this context, I had a query regarding
>>>>> the
>>>>>> correct Center-of-Mass removal settings; what would be the correct way
>>>>> to
>>>>>> ensure that the Membrane is stationary during long simulations while
>> the
>>>>>> peptides, solutes etc freely diffuse in the box. Based on my
>>>>> understanding,
>>>>>> I am using the following settings in the parameter file:
>>>>>> 
>>>>>> nstcomm                 = 100
>>>>>> comm_mode               = linear
>>>>>> comm_grps               = MEMB SOL_ION Peptide
>>>>>> 
>>>>>> Here the Peptide group includes a total of say 16 peptides. Would this
>>>>> be
>>>>>> the correct way to perform the simulation?
>>>>>> 
>>>>>> Is is possible to set individual nstcomm values for each group so as
>> to
>>>>>> ensure that the peptides diffuse freely while the membrane stays
>>>>> stationary?
>>>>> 
>>>>> Why do you want that in the first place? Membranes aren’t stationary.
>>>>> 
>>>>> Kind regards,
>>>>> Erik
>>>>> 
>>>>>> 
>>>>>> The full mdp settings I intend to use is provided in the following
>> link:
>>>>>> 
>>>>>> https://drive.google.com/file/d/0B9VrCGta6IxES3NHU3lRbGJ4d00
>>>>> /view?usp=sharing
>>>>>> 
>>>>>> Please advise of what would be the best settings to perform the
>>>>> simulation.
>>>>>> 
>>>>>> Best Regards,
>>>>>> Abhishek Acharya
>>>>>> C-CAMP
>>>>>> Bangalore.
>>>>>> --
>>>>>> Gromacs Users mailing list
>>>>>> 
>>>>>> * Please search the archive at http://www.gromacs.org/Support
>>>>> /Mailing_Lists/GMX-Users_List before posting!
>>>>>> 
>>>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>>> 
>>>>>> * For (un)subscribe requests visit
>>>>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>>>>> send a mail to gmx-users-request at gromacs.org.
>>>>> 
>>>>> --
>>>>> Gromacs Users mailing list
>>>>> 
>>>>> * Please search the archive at http://www.gromacs.org/Support
>>>>> /Mailing_Lists/GMX-Users_List before posting!
>>>>> 
>>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>> 
>>>>> * For (un)subscribe requests visit
>>>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>>>>> send a mail to gmx-users-request at gromacs.org.
>>>> 
>>>> 
>>>> 
>>> --
>>> Gromacs Users mailing list
>>> 
>>> * Please search the archive at http://www.gromacs.org/
>> Support/Mailing_Lists/GMX-Users_List before posting!
>>> 
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>> 
>>> * For (un)subscribe requests visit
>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to gmx-users-request at gromacs.org.
>> 
>> --
>> Gromacs Users mailing list
>> 
>> * Please search the archive at http://www.gromacs.org/
>> Support/Mailing_Lists/GMX-Users_List before posting!
>> 
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>> 
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to gmx-users-request at gromacs.org.
>> 
> -- 
> Gromacs Users mailing list
> 
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
> 
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.