[gmx-users] pmf

[Justin Lemkul]

On 10/28/16 4:20 PM, Alex wrote:
> Thanks for your response.
> The version for all tasks is "gmx, VERSION 5.1.2".
>
> I hope extra windows would be required just in the gaps, not in the whole
> reaction path.

Yes, you just need more windows.

> I had 81 windows spacing mostly 0.01 nm and in the end 0.04 nm.
>

Your histograms indicate that you're not achieving this.  You're getting 
clusters of sampling with gaps.  A resolution of 0.01 nm is probably overkill 
for just about any sort of reaction coordinate; check what grompp tells you 
about each of these starting configurations because it appears you may not be 
getting what you think you are to start.  You have a ton of overlap (i.e., 
redundant sampling) in the histograms.

> The peptide surly has many rotations from top of the surface to the midd
> point of box where if free of solid surface, I was wondering if this PMF
> (which is COM N N Y) has already taken into account all that rotation, if
> not, how one can consider these changes?
>

You have to have converged sampling of all motions.  That requires sufficient 
time and an assessment of how well the PMF itself has converged over time.  You 
don't yet even have a reasonable PMF.

-Justin

> Best regards,
> Alex
>
>
> On Fri, Oct 28, 2016 at 10:09 PM, Justin Lemkul <jalemkul at vt.edu> wrote:
>
>>
>>
>> On 10/28/16 4:04 PM, Alex wrote:
>>
>>> Dear Gromacs user,
>>>
>>> Below are the result of PMF simulation of a peptide adsorbed into a solid
>>> surface.
>>> Would you please comment on them in order to improve the result.
>>>
>>> The force constant in the umberela sampling is 1100.
>>> For each windows:  1.2 ns NVT equilibration followed by 2 ns of NPT and 15
>>> ns of NVT produciton run.
>>> Should I change the force constant or increase the simulation time for
>>> each
>>> windows or just increasing the sampling in some region?
>>>
>>>
>>> https://drive.google.com/open?id=0B_CbyhnbKqQDMkpLc0didlMxTWs
>>>
>>> https://drive.google.com/open?id=0B_CbyhnbKqQDZ1RpT3JxMjdyazA
>>>
>>>
>> You need more windows.  You have obvious gaps in several parts of the
>> reaction coordinate.
>>
>> https://drive.google.com/open?id=0B_CbyhnbKqQDZlR4RGVIWk5KbVU
>>>
>>> It is just one dimension pulling by "N N Y" by pull-coord1-geometry =
>>> distance, then I was also wondeing why my pullx.xvg files has more than
>>> normal column so that gmx wham complain with error about it:
>>> Fatal error:
>>> Using 1 pull coodinates from prdd.0.tpr,
>>>  but found 4 data columns in pullx-prd.0.xvg (expected 3).
>>>
>>>
>> What version of GROMACS is this?  Is the version of gmx wham the same as
>> the version of grompp/mdrun used to create/run the simulations?  I would
>> expect 4 as well (COM x,y,z of reference and z-coordinate of pulled group),
>> but there have been a lot of changes to the pull code in recent versions.
>>
>> -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.
>>

-- 
==================================================

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

==================================================