[gmx-users] Umbrella Sampling tutorial

[Justin Lemkul]

On 12/1/13 10:38 AM, shahab shariati wrote:
> Dear Justin
>
> When I placed drug within the lipid bilayer, after Generating Configurations
> step, I obtained 501 gro files. When I used distance.pl script for these
> gro files, my summary_distances.dat file is as follows:
>
> 0    0.2756352
> 1    0.2535359
> 2    0.2447009
> 3    0.3004653
> 4    0.2936225
> 5    0.1910549
> 6    0.2493761
> 7    0.2033079
> 8    0.1769952
> 9    0.2731782
> 10    0.2927040
> 11    0.3717106
> 12    0.3232497
> .
> .
> .
> .
> 485    0.7589852
> 486    0.6876000
> 487    0.6966222
> 488    0.7487700
> 489    0.7043532
> 490    0.6424745
> 491    0.7194880
> 492    0.6209456
> 493    0.6327312
> 494    0.6861565
> 495    0.5513341
> 496    0.6917742
> 497    0.7074009
> 498    0.7403648
> 499    0.6833771
> 500    0.7035147
>
> As can be seen the range of distance is 0.27-0.7, while the dimensions of
> box is  5.92819   6.43131   7.75764.
>
> My supposition was that the range of distance should be 0-3.9 (7.75764/2).
>
> You are right. The settings depend on where I place everything.
> This is my first experience in com pulling. Please guide me to use
> appropriate Pull code parameters in my case (drug molecules within the
> lipid bilayer).
>
> I used following Pull code parameters:
>
> ; Pull code
> pull            = umbrella
> pull_geometry   = position  ; simple distance increase
> pull_dim        = N N Y
> pull_start      = yes       ; define initial COM distance > 0
> pull_ngroups    = 1
> pull_group0     = CHOL_DOPC
> pull_group1     = drg
> pull_rate1      = 0.001      ; 0.01 nm per ps = 10 nm per ns
> pull_k1         = 100      ; kJ mol^-1 nm^-2
> pull_init1      = 0
> pull_vec1       = 0 0 1
>
> Except Pull code parameters, what things are important?
>

I have never attempted such a pulling procedure.  Hopefully someone else will 
comment on their experiences.  Do not try to hack the tutorial into working for 
your system; the approaches will be quite different.  The manual explains the 
logic behind the different geometries; you may have to experiment a bit to get 
things to work.

-Justin

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

Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441

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