Re: umbrella sampling of decalanine using the ABF module

From: Ilja Khavrutskii (ikhavru_at_mccammon.ucsd.edu)
Date: Mon Feb 25 2008 - 16:34:41 CST

Indeed there are alternatives to WHAM:

http://link.aip.org/link/?jcp/128/044106 (free access, just a click away)

This one is an umbrella sampling (restraints), but needs a little bit
stiffer springs.

Advantages:
- does not require the overlap between the windows (interpolates);
- can be applied to unlimited number of coordinates;
- works with both Cartesian and non-linear coordinates.
- PMF integration is iteration-free
- etc.

Storage: only needs to store averaged coordinates and not the timeseries.
etc.

-Ilja

On Mon, 25 Feb 2008, George Madalin Giambasu wrote:

> Hi!
>
> This is how I would do it - it might be the simpler way.
>
> As soon as you are able to constrain your desired "coordinate/coordinates"
> (and this can be done very easily in tclforce in namd) you are done. You will
> just have to output to a file (each for your each window along the
> pre-defined path) the instantaneous values of the "coordinate", constraining
> force constant and equilibrium value and feed those to a WHAM program (that
> you can find easily on the web).
>
> I am sure that there might other nicer ways . I would look in the free energy
> calculation chapter of the NAMD manual.
>
>
> George Giambasu
> Phd Student, UMN
>
> Xu, Jiancong wrote:
>> A follow-up to this question, is there a way to do umbrella sampling
>> calculations with NAMD? does NAMD offer such an option?
>> Thanks.
>>
>>
>> -----Original Message-----
>> From: owner-namd-l_at_ks.uiuc.edu on behalf of Jerome Henin
>> Sent: Mon 2/25/2008 11:44 AM
>> To: Philip Fowler
>> Cc: namd-l_at_ks.uiuc.edu
>> Subject: namd-l: Re: umbrella sampling of decalanine using the ABF module
>> Hi Phil,
>> usMode is but another abf-specific keyword, it's really a sub-command
>> of the abf keyword, like "coordinate" etc.
>> Jerome
>>
>> On Mon, Feb 25, 2008 at 7:21 AM, Philip Fowler
>> <philip.fowler_at_bioch.ox.ac.uk> wrote:
>>
>>> Hi,
>>>
>>> I am trying to reproduce the PMFs for the extension of decalanine from
>>> the NAMD/ABF paper [1]. I have successfully done this using adaptive
>>> biased forcing but now I want to try and use umbrella sampling.
>>>
>>> My problem is that NAMD does not appear to recognise the keyword
>>> "usMode" to force the ABF module to function in an umbrella sampling
>>> way. Even setting "usMode" to "no" results in NAMD dieing.
>>>
>>> I've tried moving the keyword around in the input file but NAMD always
>>> complains wherever it is. I am using NAMD2.6 and an example of my
>>> input file is appended. Any help would be gratefully appreciated.
>>>
>>> --Phil
>>>
>>> [1] J. Hénin and C. Chipot. Overcoming free energy barriers using
>>> unconstrained molecular dynamics simulations. J. Chem. Phys., 121(7):
>>> 2904-3004, 2004.
>>>
>>>
>>>
>>> #-----------------------------------------------------------------------------
>>> # BASIC parameters
>>> #-----------------------------------------------------------------------------
>>>
>>> temperature 301 # The initial
>>> temperature in K
>>> timestep 0.5 # Timestep in
>>> femtoseconds
>>> seed 19898473 # Allows repeatability of
>>> simulations.
>>> numsteps 4000000
>>>
>>> #-----------------------------------------------------------------------------
>>> # OUTPUT files.
>>> #-----------------------------------------------------------------------------
>>> outputname da-us-1.out
>>> restartname da-us-1.out.res
>>> velDCDfile da-us-1.out.vel.dcd
>>> DCDfile da-us-1.out.dcd
>>> XSTfile da-us-1.out.xst
>>>
>>> #-----------------------------------------------------------------------------
>>> # SETUP info
>>> #-----------------------------------------------------------------------------
>>>
>>> structure da.psf
>>> coordinates da.pdb
>>>
>>>
>>> paraTypeCharmm on
>>> parameters par_all27_prot_na.inp
>>>
>>> #-----------------------------------------------------------------------------
>>> # OUTPUT PARAMETERS
>>> #-----------------------------------------------------------------------------
>>>
>>> outputTiming 5000 # How often CPU
>>> times output
>>> outputEnergies 1000 # How often
>>> Energies are written
>>> to STDOUT
>>>
>>> binaryoutput yes # Use binary
>>> output files yes/no
>>> binaryrestart yes
>>> restartFreq 50000
>>> restartsave no
>>>
>>> DCDfreq 10000 # Interval
>>> between DCD trajector
>>>
>>> DCDUnitcell yes
>>> velDCDfreq 50000
>>>
>>> #-----------------------------------------------------------------------------
>>> # NON-BONDED PARAMETERS
>>> #-----------------------------------------------------------------------------
>>>
>>> dielectric 1 # Dielectric constant
>>>
>>> switching on # Smooth the transition
>>> switchdist 10 # Distance at which
>>> smoothing
>>> starts
>>> cutoff 12 # Distance at which all
>>> non-
>>> bonded interactions are zero
>>> pairlistdist 13.5 # Distance within which
>>> all
>>> atoms are considered for non-bonded
>>> #
>>> interactions. Must be greater than CUTOFF.
>>>
>>> stepspercycle 20
>>>
>>> PME off # Implement the
>>> Particle-Mesh-Ewald
>>> #
>>> electrostatics algorithm. Must be
>>> #
>>> used with periodic b.c.
>>> exclude scaled1-4 # Which bonded
>>> interactions
>>> to exclude from the non-bonded calculations
>>> 1-4scaling 1.0
>>>
>>>
>>> #-----------------------------------------------------------------------------
>>> # TEMPERATURE CONTROL
>>> #-----------------------------------------------------------------------------
>>>
>>> langevin on # use Langevin dynamics
>>> langevinTemp 300.0
>>> langevinDamping 1.0 # coupling coefficients to be
>>> applied to all atoms
>>> langevinHydrogen no # apply to hydrogens
>>>
>>>
>>> #-----------------------------------------------------------------------------
>>> # ADAPTIVE BIASED FORCE PARAMETERS
>>> #-----------------------------------------------------------------------------
>>>
>>> source
>>> /Users/fowler/packages/namd-2.6/lib/abf/abf.tcl
>>> abf coordinate distance
>>> usMode no
>>> abf abf1 3
>>> abf abf2 98
>>> abf dxi 0.1
>>> abf xiMin 12.0
>>> abf xiMax 32.0
>>> abf outFile da-1.abf.dat
>>> abf fullSamples 500
>>> abf inFiles {}
>>> abf distFile da-1.abf.dist
>>> abf dSmooth 0.0
>>>
>>> abf restraintList {
>>> dist1 {dist {BH 1 N} {BH 10 NT} 1.0 16.1}
>>> }
>>>
>>>
>>>
>>>
>>>
>>
>>
>>
>>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:20:55 CST