From: Chris Harrison (charris5_at_gmail.com)
Date: Tue Mar 06 2012 - 10:11:24 CST
Oh, that's actually a really good point. I forgot that while namd cells
include bond distances, vmd frames do not (at least not as explicit
scalars), so in a typical aqueous system the measure minmax approach could
be off by approx. an angstrom or so, ie the O-H bond distance.
Hmm, now you got me thinking...there should be an algorithmic way to
determine the real box size from the wrapped cartesians if you know the
implicit bond distances (which VMD does) of atom sets along all sides and
edges .... the cost/benefit ratio of taking time to implement doesn't look
favorable, but maybe as a flag to measure minmax someday ....?
Guess I need to pull my head out of the namd code more often. ;)
Best,
Chris
On Tue, Mar 6, 2012 at 9:55 AM, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
> On Tue, Mar 6, 2012 at 10:44 AM, Chris Harrison <charris5_at_gmail.com>
> wrote:
> > There's no reason, if the user has the correct frame/structure loaded,
> with
> > coordinates properly wrapped, that the measure minmax approach will
> differ
> > from the molinfo command. I strongly suspect a misunderstanding of the
> > actual frame loaded or a similar inadvertent user error in Mingjun's
> case.
> >
> >
> > I'm advocating using *both* approaches, not just one. If they don't
> match,
> > find out why.
>
> sorry, chris,
>
> but measure minmax will be wrong, even if you
> wrap all coordinates, unless you have by chance
> always atoms *exactly* on the box boundaries
> in all directions. similarly, if you have (partially)
> unwrapped coordinates (even if only the water
> molecules) there will always be atoms "sticking out".
>
> thus measure minmax will be an approximation,
> and depending on wrapping settings either under
> or overestimate, but never give the exact box dimensions.
>
> cheers,
> axel.
>
>
> >
> > Best,
> > Chris
> >
> >
> > On Tue, Mar 6, 2012 at 9:32 AM, Ajasja Ljubetič <
> ajasja.ljubetic_at_gmail.com>
> > wrote:
> >>
> >> But this script is using measure minmax, so in principle this is the
> same
> >> as trying to reconstruct the periodic box from atomic coordinates
> (correct
> >> me if I'm wrong).
> >> I think a better version is:
> >>
> >> proc get_cell {{molid top}} {
> >> puts "cellBasisVector1 [molinfo $molid get a] 0 0"
> >> puts "cellBasisVector2 0 [molinfo $molid get b] 0"
> >> puts "cellBasisVector3 0 0 [molinfo $molid get c]"
> >>
> >> set all [atomselect $molid all]
> >> puts "cellOrigin [measure center $all] "
> >> $all delete
> >> }
> >>
> >> On Tue, Mar 6, 2012 at 16:23, Chris Harrison <charris5_at_gmail.com>
> wrote:
> >>>
> >>> And just in case JC's correct answer isn't enough for you ... this
> >>> question was previously answered on Feb 9
> >>> (http://www.ks.uiuc.edu/Research/namd/mailing_list/namd-l/15960.html),
> >>> providing the below tcl scripted solution which should always work no
> matter
> >>> what source file was used to load the frame. Please remember to
> search the
> >>> listserv (http://www.ks.uiuc.edu/Research/namd/mailing_list/) for
> possible
> >>> answers.
> >>>
> >>> Best,
> >>> Chris
> >>>
> >>> In the console enter the below function or
> >>> else save it in a text file and from the console type: source
> >>> directoryToYourTextFile/nameOfYourTextFile.
> >>>
> >>> proc get_cell {{molid top}} {
> >>> set all [atomselect $molid all]
> >>> set minmax [measure minmax $all]
> >>> set vec [vecsub [lindex $minmax 1] [lindex $minmax 0]]
> >>> puts "cellBasisVector1 [lindex $vec 0] 0 0"
> >>> puts "cellBasisVector2 0 [lindex $vec 1] 0"
> >>> puts "cellBasisVector3 0 0 [lindex $vec 2]"
> >>> set center [measure center $all]
> >>> puts "cellOrigin $center"
> >>> $all delete
> >>> }
> >>>
> >>> Now make sure your molecule is the top molecule in VMD (ie has the T
> >>> in it's line in the "VMD Main" window). Then in the console type:
> >>> get_cell. The basis vectors for your system should be output.
> >>>
> >>>
> >>> Best,
> >>> Chris
> >>>
> >>>
> >>> On Tue, Mar 6, 2012 at 9:00 AM, JC Gumbart <gumbart_at_ks.uiuc.edu>
> wrote:
> >>>>
> >>>> And to further clarify, VMD reads them from the dcd and can be
> accessed
> >>>> using
> >>>>
> >>>> molinfo top get a (b and c)
> >>>>
> >>>> -----Original Message-----
> >>>> From: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] On
> >>>> Behalf
> >>>> Of Axel Kohlmeyer
> >>>> Sent: Tuesday, March 06, 2012 8:53 AM
> >>>> To: Aron Broom
> >>>> Cc: mjyang; namd-l_at_ks.uiuc.edu
> >>>> Subject: Re: namd-l: how to calculate the box size
> >>>>
> >>>> On Tue, Mar 6, 2012 at 9:44 AM, Aron Broom <broomsday_at_gmail.com>
> wrote:
> >>>> > What if for the water molecules in your system you calculate from
> the
> >>>> > center of mass of the waters rather than just the atomic
> coordinates?
> >>>> > I don't know what the real answer is, but that would shrink your box
> >>>> > ever
> >>>> so slightly.
> >>>>
> >>>> that is nonsense. you cannot reconstruct the box size from the
> >>>> positions of
> >>>> the atoms. full stop. but the box dimensions are usually stored in the
> >>>> dcd
> >>>> trajectory file. those are the only ones you can use or those in a
> >>>> proper
> >>>> .xsc file.
> >>>>
> >>>> axel.
> >>>>
> >>>> >
> >>>> >
> >>>> > On Tue, Mar 6, 2012 at 2:48 AM, mjyang <mjyang_at_hku.hk> wrote:
> >>>> >>
> >>>> >> Dear NAMD users,
> >>>> >>
> >>>> >> I performed a simulation by NAMD for a protein system solvated
> by
> >>>> >> a cubic water box. Now I extracted the snapshots from NAMD
> trajectory
> >>>> >> for postpone analysis, in which the protein should be centered in
> the
> >>>> >> box. The box size was calculated by (max_x-min_x, max_y-min_y,
> >>>> >> max_z-min_z), where max_x,y,z and min_x, y, z are the maximum and
> >>>> >> minimum coordinate values of the extracted frame. However, the box
> >>>> >> size calculated in this way is a little larger than the real one,
> >>>> >> resulting in gaps in the recentered box. It seems that the box size
> >>>> >> written out by NAMD in the *.xsc file is better than what I
> >>>> >> calculated. But the NAMD output size (in *.xsc file) only
> corresponds
> >>>> >> to the last frame of a segment of trajectory file and I need the
> size
> >>>> >> for each frame in the traj. Could someone please tell me how NAMD
> >>>> calculate the box along MD simulations?
> >>>> >>
> >>>> >>
> >>>> >> Many thanks.
> >>>> >>
> >>>> >>
> >>>> >> Mingjun
> >>>> >>
> >>>> >
> >>>> >
> >>>> >
> >>>> > --
> >>>> > Aron Broom M.Sc
> >>>> > PhD Student
> >>>> > Department of Chemistry
> >>>> > University of Waterloo
> >>>> >
> >>>>
> >>>>
> >>>>
> >>>> --
> >>>> Dr. Axel Kohlmeyer
> >>>> akohlmey_at_gmail.com http://goo.gl/1wk0
> >>>>
> >>>> College of Science and Technology
> >>>> Temple University, Philadelphia PA, USA.
> >>>>
> >>>>
> >>>
> >>>
> >>>
> >>> --
> >>> Chris Harrison, Ph.D.
> >>> NIH Center for Macromolecular Modeling and Bioinformatics
> >>> Theoretical and Computational Biophysics Group
> >>> Beckman Institute for Advanced Science and Technology
> >>> University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801
> >>>
> >>> http://www.ks.uiuc.edu/Research/namd Voice: 773-570-6078
> >>> http://www.ks.uiuc.edu/~char Fax:
> 217-244-6078
> >>>
> >>>
> >>>
> >>
> >
> >
> >
> > --
> > Chris Harrison, Ph.D.
> > NIH Center for Macromolecular Modeling and Bioinformatics
> > Theoretical and Computational Biophysics Group
> > Beckman Institute for Advanced Science and Technology
> > University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801
> >
> > http://www.ks.uiuc.edu/Research/namd Voice: 773-570-6078
> > http://www.ks.uiuc.edu/~char Fax: 217-244-6078
> >
> >
> >
>
>
>
> --
> Dr. Axel Kohlmeyer
> akohlmey_at_gmail.com http://goo.gl/1wk0
>
> College of Science and Technology
> Temple University, Philadelphia PA, USA.
>
-- Chris Harrison, Ph.D. NIH Center for Macromolecular Modeling and Bioinformatics Theoretical and Computational Biophysics Group Beckman Institute for Advanced Science and Technology University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801 http://www.ks.uiuc.edu/Research/namd Voice: 773-570-6078 http://www.ks.uiuc.edu/~char Fax: 217-244-6078
This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:21:18 CST