[gmx-users] PMF

Justin Lemkul jalemkul at vt.edu
Thu Oct 13 00:02:38 CEST 2016



On 10/11/16 4:01 PM, Alex wrote:
> Dear gromacs user,
>
> I am trying to simulate the PMF a small peptide adsorbed to a solid surface
> but no success unfurtunatly.
> For some force constant the peptide would not desorbed from the surface and
> for higher force constants what you can see in enclosed file will happen.
> Some question is:
> If it is possible to somehow change the force constant for a coordinate
> while pulling step? for example the first 1 ns by by k1 = 5000 and then
> reduced to k1 = 1000 for the rest of pulling.
>

gmx wham won't account for changes in k over time.  But you can use different 
force constants in different windows; that's fine.

> Also, the force constant in the pulling step and umbrella sampling do not
> need to be the same necessarily?
>

Correct.

> How I can improve or increase the number of sampled frame?
>
>  I have enclosed some results as well as mdp file and force in pulling step
> in each step below, I would be so appreciated if you could guide me to
> improve the results?
>
> PMF result:
> https://drive.google.com/open?id=0B_CbyhnbKqQDMC1Ua3BWLUZ3XzA
>

If you have a non-zero pulling rate (which you have specified above), you're not 
doing umbrella sampling, you're doing non-equilibrium SMD.  So this PMF is junk. 
  You need to have multiple windows, each with a set reference restraint 
distance, and a zero pull rate so the conformations can oscillate about that 
reference length.

-Justin

> Force in pulling SMD
> https://drive.google.com/open?id=0B_CbyhnbKqQDVWxkd2VLUmozUjg
>
> position in pulling SMD
> https://drive.google.com/open?id=0B_CbyhnbKqQDa0VKX0E0Zlp1ODA
>
> Windows-com
> https://drive.google.com/open?id=0B_CbyhnbKqQDR0poVlM2RTVSNG8
>
> md-production.mdp
> https://drive.google.com/open?id=0B_CbyhnbKqQDdGFRbWQzb29Nbmc
>
> md-pulling.mdp
> https://drive.google.com/open?id=0B_CbyhnbKqQDc2NPdlVmSmwydFk
>
> nvt-umbrella.mdp
> https://drive.google.com/open?id=0B_CbyhnbKqQDSk9LS2d2Q1pTN2s
>
> npt-umbrella.mdp
> https://drive.google.com/open?id=0B_CbyhnbKqQDVnB4STFxeENfREU
>
> md-umbrella.mdp
> https://drive.google.com/open?id=0B_CbyhnbKqQDQmFTZWRkR1BkcUE
>
> Thank you very much in advance.
>
> Regards,
> Alex
>

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

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

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


More information about the gromacs.org_gmx-users mailing list