Re: ABF questions

From: Branko (bdrakuli_at_chem.bg.ac.yu)
Date: Wed Jan 06 2010 - 07:17:02 CST

Austin,

You mixed the ABF definition in 2.6 and 2.7: Just transfer ABF section
from namd configuration (1.emin_colvar.conf) to colvar configuration
(now colVarfile.txt) and uncomment (delete #) ABF lines, example is below:

abf {
   colvars d
   fullSamples 500
   hideJacobian no
   outputFreq 1000
   applyBias yes
}

Obviously that source - path is unnecessary
Also try to remove extension from colVarfile (without .txt), samie in
NAMD configuration

colvars section
colvars on
colvarsConfig colVarfile

It should work. When you use one colvar, 27b gives three outputs
<file>.count , <file>.grad , <file>.pmf.
For more details see 2.7b user guide:
http://www.ks.uiuc.edu/Research/namd/2.7b2/ug/

Hope this help.

Branko

Austin B. Yongye wrote:
> Dear all:
> A couple of questions:
> 1). I used my current version of namd (NAMD_2.7b2) with the initial input files I sent to you, and had this output. See attached file, atom_based_01.dat. Is column2 the file the free energy change at each xi value? Would that then represent the free energy change of the reaction coordinate?
> 2). I read the collective variables section of the NAMD_2.7b2 manual and prepared the .conf and collective variable files attached. When I run the simulation I get this error message:
> colvars: Warning: "colvar" keyword found without any configuration.
> colvars: Error: no collective variables defined.
> colvars: If this error message is unclear, try recompile with -DCOLVARS_DEBUG.
> FATAL ERROR: Error in the collective variables module: exiting.
>
> I thought of compiling with the -DCOLVARS_DEBUG option, but the colvars subdirectory of my NAMD_2.7b2 has only these two files:
> colvarproxy_standalone.C and colvarproxy_standalone.h, missing several of the files in the include statement missing. I downloaded the NAMD_2.7b2 source code, but the colvars subdirectory still had only these two files.
>
> Any help will be appreciated.
> Thanks,
> Austin-
>
>
> --- On Tue, 1/5/10, Jérôme Hénin <jhenin_at_ifr88.cnrs-mrs.fr> wrote:
>
>
>> From: Jérôme Hénin <jhenin_at_ifr88.cnrs-mrs.fr>
>> Subject: Re: namd-l: ABF questions
>> To: "Austin B. Yongye" <ybausty_at_yahoo.com>
>> Cc: "NAMD" <namd-l_at_ks.uiuc.edu>
>> Date: Tuesday, January 5, 2010, 9:44 AM
>> Dear Austin:
>>
>> 1) xiMin and xiMax are boundaries of the grid used to
>> collect data,
>> but the system is free to evolve between them.
>>
>> 2) you are using an older version of the code. The current
>> version is
>> part of the "collective variable calculations" module of
>> NAMD 2.7 and later.
>>
>> 3) any data analysis software should be able to read the
>> files (the
>> are plain text). I like xmgrace and gnuplot.
>>
>> Best,
>> Jerome
>>
>> 2010/1/5 Austin B. Yongye <ybausty_at_yahoo.com>:
>>
>>> Dear all,
>>> I am trying to apply ABF during a simulation. I have
>>>
>> read the ABF section in the NAMD27b1_Documentation file and
>> the ABF-Mar2008 tutorial. Please I have some questions:
>>
>>> 1). Is the distance between abf1 and abf2
>>>
>> decreased/increased progressively from xiMax to xiMin in the
>> course of the simulation; or are xiMax and xiMin simply used
>> as limits for creating bins, and then energies are computed
>> for structures in each bin?
>>
>>> 2). What keyword should I use to generate the .grad,
>>>
>> .count and .pmf output files? My current output does not
>> include these files.
>>
>>> 3). What tools do I need to generate the free energy
>>>
>> plot for the simulation (e.g. Figure 1 of the
>> ABF-Mar2008.pdf tutorial)?
>>
>>> Thanks,
>>> Austin-
>>>
>>> My configuration and abf.tcl files are included
>>>
>> below:
>>
>>> #protocol params
>>> numsteps 20000
>>>
>>> #initial config
>>> ambercoor ../3mem3mem.crd
>>> parmfile ../3mem3mem.top
>>>
>>> #output params
>>> outputname 3mem3mem_min
>>> binaryoutput on
>>>
>>> #force field params
>>> paraTypeCharmm off
>>> amber yes
>>> parameters parm94.dat
>>> exclude scaled1-4
>>> scnb 2.0
>>> 1-4scaling 0.83
>>> switching off
>>> switchdist 8.0
>>> cutoff 12.0
>>> pairlistdist 14.0
>>> pairlistsPerCycle 2
>>> margin 0.0
>>> stepspercycle 20
>>> rigidBonds all
>>> timestep 0.5
>>>
>>> #dcd file
>>> DCDfile 3mem3mem_atombased.dcd
>>> DCDfreq 10
>>>
>>> temperature 300
>>>
>>> #abf section
>>> source ./Tutorial-ABF/abf-1.8/abf.tcl
>>> abf coordinate distance
>>> abf abf1 90
>>> abf abf2 59
>>> abf xiMin 1.45
>>> abf xiMax 9.00
>>> abf dxi 0.1
>>>
>>>
>> ------------------------------------------------------
>>
>>>
>>>
>> #############################################
>>
>>> # Generic ABF code
>>>
>> #
>>
>>> # Jerome Henin <jerome.henin_at_uhp-nancy.fr>
>>>
>> #
>>
>>>
>>>
>> #############################################
>>
>>> ############
>>> # Startup #
>>> ############
>>>
>>> package provide abf 1.7.0
>>>
>>> #########################
>>> # Parameter definitions #
>>> #########################
>>>
>>> namespace eval ::ABF {
>>>
>>> set version "1.7"
>>>
>>> if {! [info exists ABFdir]} { set ABFdir [file dirname
>>>
>> [info script]] }
>>
>>> # If it fails, try the local directory
>>> if { $ABFdir == "" } { set ABFdir "." }
>>>
>>> TclForces on
>>> TclForcesScript $ABFdir/abf_script.tcl
>>>
>>> array set defaults {
>>> inFiles {}
>>> outFile atom_based_01.dat
>>> outputName 3mem3mem_abf
>>> historyFile none
>>> distFile distfile_abf.dat
>>> forceConst 500.0
>>> fullSamples 500
>>> outputFreq 100
>>> df 1.0
>>> fMax 60.0
>>> dSmooth 0.0
>>> writeXiFreq 100
>>> writeFxiFreq 100
>>> usMode no
>>> applyBias yes
>>> moveBoundary 0
>>> temp 300.0
>>> abf2 {}
>>> }
>>>
>>> set mandatory "coordinate xiMin xiMax dxi abf1"
>>>
>>> # these settings are not displayed at startup
>>> set silent "restraintList usMode SFM applyBias
>>>
>> direction abf3"
>>
>>> array set capitals {}
>>> foreach param [concat $silent $mandatory [array names
>>>
>> defaults]] {
>>
>>> set capitals([string tolower $param]) $param
>>> # not set yet
>>> set alreadySet($param) 0
>>> }
>>>
>>> } ;# namespace
>>>
>>> proc abf { keyword value } {
>>> set ::ABF::keyword $keyword
>>> set ::ABF::value $value
>>>
>>> namespace eval ::ABF {
>>>
>>> # Build list of all allowed parameter names
>>> set list [array names capitals]
>>>
>>> set lowercase [string tolower $keyword]
>>>
>>> # Process parameters
>>> if {[lsearch $list $lowercase] != -1} {
>>> set keyword $capitals($lowercase)
>>>
>>> if { $alreadySet($keyword) } {
>>> print "ABF> WARNING - multiple
>>>
>> definitions of parameter " $keyword
>>
>>> }
>>>
>>> set $keyword $value
>>> set alreadySet($keyword) 1
>>>
>>> return
>>> } else {
>>> error [format "Unknown ABF keyword: %s"
>>>
>> $keyword]
>>
>>> }
>>>
>>>
>>> } ;# namespace
>>> } ;# proc abf
>>>
>>> # define upper-case synonyms to proc abf
>>> proc ABF { keyword value } {
>>> abf $keyword $value
>>> }
>>> proc Abf { keyword value } {
>>> abf $keyword $value
>>> }
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>
>
>
> ------------------------------------------------------------------------
>
> #collective variables config file
>
> #globalOption value
> colvarsTrajFrequency 100
> colvarRestartFrequency 100
> analysis on
>
> colvar
> {
> name cv1 # needed to identify the variable
>
> # flags to define the behavior of 1st-colvar
> width 0.5
> lowerBoundary 1.0
> upperBoundary 9.0
>
>
> # all the following terms are summed together to provide the
> # value of 1st-colvar
> distance
> {
> # configuration for the 1st colvar component
> group1
> {
> atomNumbers {91 92 93}
> }
> group2
> {
> atomNumbers {57 60}
> }
> oneSiteSystemForce yes
> }
>
> }
>
> ------------------------------------------------------------------------
>
>
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 9.0.725 / Virus Database: 270.14.126/2601 - Release Date: 01/05/10 08:35:00
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:55:19 CST