AMBER Archive (2007)

Subject: Re: AMBER: umbrella sampling: distance restraint

From: Jerome.GOLEBIOWSKI_at_unice.fr
Date: Thu May 10 2007 - 07:19:35 CDT


Hi,
I don't understand why you need to restrain some protein residues. This
additionnal bias
won't help you to get a physically relevant free energy profile.
In your distance restaint input, although it is not absolutely
necessary, you should work with a harmonic restraint, it is not the case
here. I think that if you intend to use the WHAM code of A. Grossfield,
your potential have to be harmonic. If you want your distance to vary
around 13 A, put r2 and r3 to 13 and r1 and r4 far from this value (ie.
r1=0, r4=99.0), to ensure a harmonic potential.
Also, your starting US distance is 11.869. Don't pull to much this
distance, a increasing distance with windows separated by ~ 0.5A is
already a hard pulling. Here that's probably why your system explode
(vlimit exceeded...) I suggest a pulling increase of 0.4A or less, at
least at the begining of the unbinding process.
Additionnaly, in each window you have to perform an equilibration MD
prior to processing the ligand-protein distance for your PMF.
To give your an idea on a protein-ligand complex, we have performed a
PMF calculation with windows of 0.4 A lasting ~ 60 ps (eq. of 20 ps per
window). Smaller windows didn't gave converged PMF. Maybe you could try
this.
Good luck
Jerome

On Thu, 2007-05-10 at 18:25 +0800, Catein Catherine wrote:
> Dear Sir/Madam,
>
> I am new to PMF type calculations, and i could like to perform a umbrella
> sampling for a system like this. We are interest to monitor the process for
> the drug to move at a specific direction from the receptor binding pocket,
> like this:
>
> RECEPTOR ______________
> BINDING | Drug------------> remove from this direction only
> POCKET _______________
>
> I tried to set the distance restraint for a simiplied systems with RESIDUE
> 1to 4 as the receptor binding pocket, and residue 5 as the drug, and do the
> calculation as follows:
> (1) keeping the receptor binding pocket (RESIDUES 1,2,3,4) fixed
> (2) Increase the distance of the drug (RESIDUE 5, center of mass) from an
> atom in receptor
>
> However, I noted that the drug is in fact moving in more than one direction,
> and the drug catch with the receptor binding pocket. Here is my input
> files, could you please kindly help me to point out what did I did wrong?
>
> -------------------------------------MY INPUT
> FILE--------------------------------------------------
> Umbrella sampling MD
> &cntrl
> imin=0, irest=1, ntx=5, saltcon=0.4
> nstlim=3000, dt=0.002, ntc=1, ntr=1, nmropt=1
> ntpr=100, ntwx=100,
> cut=16, ntb=0, igb=1,
> ntt=3, gamma_ln=1.0,
> tempi=300.0, temp0=300.0,
> /
> &wt type='DUMPFREQ', istep1=10 /
> &wt type='END' /
> DISANG=residue.restraint
> DUMPAVE=residue.result
> Keep upper residues 1 2 3 4 fixed with weak restraints
> 10.0
> RES 1 4
> END
> END
>
> =========================MY residue.restraint
> FILE==================================
>
> # distance restrainst for drug
> &rst iat=4,5, r1=10.0, r2=13.0, r3=15.0, r4=20.0, rk2 =30.0, rk3 = 30.,
> iresid=1, ir6=0,
> ATNAM(1)=P,
> ATNAM(2)=C31,ATNAM(2)=C25,ATNAM(2)=C21,ATNAM(2)=C16,ATNAM(2)=C8,ATNAM(2)=O24,ATNAM(2)=C20,ATNAM(2)=O23,ATNAM(2)=O22,ATNAM(2)=C19,ATNAM(2)=C15,ATNAM(2)=C14,ATNAM(2)=O18,ATNAM(2)=N12,ATNAM(2)=C11,ATNAM(2)=C10,ATNAM(2)=C13,ATNAM(2)=C9,ATNAM(2)=N10,ATNAM(2)=C5,ATNAM(2)=C4,ATNAM(2)=C3,ATNAM(2)=C,ATNAM(2)=O,ATNAM(2)=C2,ATNAM(2)=C1,ATNAM(2)=C6,ATNAM(2)=C7,ATNAM(2)=C17,ATNAM(2)=C98
> /
>
> ===============================My OUTPUT FILE
> ==================================
>
> --------------------------------------------------------------------------------
> 4. RESULTS
> --------------------------------------------------------------------------------
>
>
> NSTEP = 100 TIME(PS) = 62.200 TEMP(K) = 1651.00 PRESS =
> 0.0
> Etot = 1193.3862 EKtot = 861.2264 EPtot =
> 332.1598
> BOND = 183.2061 ANGLE = 291.2733 DIHED =
> 131.1998
> 1-4 NB = 68.9142 1-4 EEL = -381.9480 VDWAALS =
> -22.4709
> EELEC = -139.7952 EGB = -302.3838 RESTRAINT =
> 504.1644
> EAMBER (non-restraint) = -172.0047
> ------------------------------------------------------------------------------
>
> NMR restraints: Bond = 0.140 Angle = 0.000 Torsion = 0.000
> ===============================================================================
> vlimit exceeded for step 130 ; vmax = 23.8542427527606
> vlimit exceeded for step 172 ; vmax = 49.3584438745089
>
> NSTEP = 200 TIME(PS) = 62.400 TEMP(K) = 2084.33 PRESS =
> 0.0
> Etot = 1538.8958 EKtot = 1087.2641 EPtot =
> 451.6318
> BOND = 307.2185 ANGLE = 388.1555 DIHED =
> 130.5633
> 1-4 NB = 65.1899 1-4 EEL = -399.4312 VDWAALS =
> -48.0505
> EELEC = -111.9660 EGB = -312.9043 RESTRAINT =
> 432.8567
> EAMBER (non-restraint) = 18.7751
> ------------------------------------------------------------------------------
>
> NMR restraints: Bond = 0.000 Angle = 0.000 Torsion = 0.000
> ===============================================================================
> vlimit exceeded for step 204 ; vmax = 27.5963598323955
> vlimit exceeded for step 209 ; vmax = 50.1837334599151
> vlimit exceeded for step 210 ; vmax = 27.0051722625023
> vlimit exceeded for step 213 ; vmax = 25.4720474595900
> ....
> ....
> ....
> ===============================================================================
>
> NSTEP = 1600 TIME(PS) = 65.200 TEMP(K) =********* PRESS =
> 0.0
> Etot = NaN EKtot = 955696.3416 EPtot = NaN
> BOND = 179061.9638 ANGLE = 13542.3571 DIHED =
> 778.8009
> 1-4 NB = 25462.5365 1-4 EEL = -395.0060 VDWAALS =
> 5476.7559
> EELEC = -101.9259 EGB = NaN RESTRAINT =
> **************
> EAMBER (non-restraint) = NaN
> ------------------------------------------------------------------------------
>
> NMR restraints: Bond = 0.022 Angle = 0.000 Torsion = 0.000
>
> =======================MY residue.result FILE ========================
> 0 11.869
> 10 12.228
> 20 12.772
> 30 12.905
> 40 12.841
> 50 12.837
> 60 13.088
> 70 13.198
> 80 13.075
> 90 12.799
> 100 12.946
> 110 13.168
> 120 13.350
> 130 13.455
> 140 13.397
> 150 13.396
> 160 13.242
> 170 13.005
> 180 12.694
> 190 12.779
> 200 13.053
> 210 13.073
> 220 13.142
> 230 13.191
> 240 13.181
> 250 13.453
> 260 13.560
> 270 13.554
> ....
> ....
> ....
> 470 15.027
> 480 15.027
> 490 15.027
> 500 15.027
> 510 15.027
> 520 15.027
> 530 15.027
> 540 15.027
> 550 15.027
> 560 15.027
> 570 15.027
> 580 15.027
> 590 15.027
> 600 15.027
> 610 15.027
> 620 15.027
> 630 15.027
> 640 15.027
> 650 15.027
> 660 15.027
> 670 15.027
> 680 15.027
> 690 15.027
> ... same till the end of the file
>
> =============================================================
>
> Cat
>
> _________________________________________________________________
> No masks required! Use MSN Messenger to chat with friends and family.
> http://go.msnserver.com/HK/25382.asp
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu

-- 
Jerome Golebiowski, PhD
Lab. Chemistry of Bioactive Molecules and Aromas, 
Team Chemometrics and Molecular Modeling
University of Nice, parc Valrose
06108 Nice Cedex2 France
tel : +33 (0)4 92 07 61 03
http://www.unice.fr/lcmba
http://www.unice.fr/lcmba/golebiowski
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu