[gmx-users] ligand moving out during umbrella sampling

[abhisek Mondal]

I did try the code successfully on a configuration generated after pulling.
The NVT approach with direction-periodic geometry worked nicely for the
particular configuration.

However, when I tried to reapply the same code (with modified COMs and thus
pull_vec) on a different configuration, something awkward happened. The
ligand got pulled through protein and got stuck inside it. I have put the
trajectory movie alongwith md_umbrella.mdp file here:
https://drive.google.com/drive/folders/0B6O-L5Y7BiGJQ1FIc2tIRFE2dE0

Would you please care to give some advise regarding this odd behavior.

Thank you.

On Fri, May 19, 2017 at 5:18 PM, Justin Lemkul <jalemkul at vt.edu> wrote:

>
>
> On 5/19/17 5:56 AM, abhisek Mondal wrote:
>
>> On Thu, May 18, 2017 at 6:48 PM, Justin Lemkul <jalemkul at vt.edu> wrote:
>>
>>
>>>
>>> On 5/17/17 8:55 AM, abhisek Mondal wrote:
>>>
>>> This time I think I got ligand restrained successfully during the
>>>> umbrella
>>>> sampling. I have removed the restrain from protein, as per your advice.
>>>> Defined the COM vector in md_umbrella.mdp, applied pull_k1=1000 and used
>>>> pull_rate1=0.0.
>>>> I have uploaded the trajectory movie (and other mdp files) in the
>>>> following
>>>> link:
>>>> https://drive.google.com/drive/folders/0B6O-L5Y7BiGJQ1FIc2tIRFE2dE0
>>>>
>>>> However, I'm facing a problem. Due to the withdrawal of the position
>>>> restrain of protein. The protein and ligand (together) is moving around
>>>> the
>>>> box and resulting in "Distance of pull group 1 (10.441990 nm) is larger
>>>> than 0.49 times the box size (10.646989)" error.
>>>>
>>>> As per the video I have uploaded, if I assume this approach worked, then
>>>> how can I avoid this error ? Is  there any way to make sure the
>>>> protein-ligand remains in the middle of the box (or nearby). I have
>>>> taken
>>>> pretty large box compared to the protein structure from the beginning.
>>>>
>>>> Please suggest me a way out.
>>>>
>>>>
>>>> Use a larger box or use direction-periodic geometry.
>>>
>>
>>
>>  For the sake of computational power I'm leaning towards
>> direction-periodic
>> geometry. However, from the mailing list entries I found out that pressure
>> coupling should not be used for this kind of geometry setup.
>> NVT coupling with no velocity generation is what I'm opting for. There are
>> a lot of doubt regarding the md_umbrella.mdp setup using NVT protocol.
>> Would you please suggest if the code (
>> https://drive.google.com/drive/folders/0B6O-L5Y7BiGJQ1FIc2tIRFE2dE0)
>> looks
>> sensible ?
>>
>> Eagerly waiting for your opinion.
>>
>>
> Try it and see what happens.
>
> -Justin
>
> --
> ==================================================
>
> Justin A. Lemkul, Ph.D.
> Ruth L. Kirschstein NRSA Postdoctoral Fellow
>
> Department of Pharmaceutical Sciences
> School of Pharmacy
> Health Sciences Facility II, Room 629
> University of Maryland, Baltimore
> 20 Penn St.
> Baltimore, MD 21201
>
> jalemkul at outerbanks.umaryland.edu | (410) 706-7441
> http://mackerell.umaryland.edu/~jalemkul
>
> ==================================================
> --
> 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.
>



-- 
Abhisek Mondal

*Senior Research Fellow*

*Structural Biology and Bioinformatics Division*
*CSIR-Indian Institute of Chemical Biology*

*Kolkata 700032*

*INDIA*