VMD-L Mailing List

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Thu Jul 06 2006 - 13:12:39 CDT

Hi Rima,
the reason you're getting the error (which isn't really anything to
worry about) is that the atom names at the terminus are different in the
pdb file from what they should be in the patch. The patch fixes this,
but autopsf aliases the naming so that no error occurs. Either approach
should work; the autopsf one is probably marginally better. But in any
case, this is why you're seeing a differences in the output. You should
get a valid psf/pdb combo either way.

As far as your namd problems go, you're running with a 2 fs timestep but
without rigidbonds, which is a bad idea because it means that your
hydrogens will be oscillating on a timescale similar to that of your
timestep. You should either change your timestep or turn on rigidbonds,
as noted in the config file. I would also recommend giving yourself a
tiny bit of extra space in the initial periodic cell, on the order of
perhaps an angstrom on a side. This will avoid any initial clashes
(remember that the atomic radius is not counted in the measure minmax
commands in vmd) and if there is excess space the pressure control will
quickly take care of it.
Peter

Rima Chaudhuri wrote:
> Hi,
> The commands for generating the psf file is directly out of the
> tutorials which are:
>
> package require psfgen
> topology top_all27_prot_lipid.inp
> pdbalias residue HIS HSE
> pdbalias atom ILE CD1 CD
> segment U {pdb NL63.pdb}
>
> and this is where I am getting those errors.
> I was just concerned why the automated version is not reporting any
> errors at all while the manual is! I looked as the combo files and
> they looked ok to me too...but(?)!
> I thought it might be a minimization problem and increased the time to
> 10000 steps from 2000. As soon as the equilibration starts it gives me
> the global exclusion count error. I have attached the config file that
> I was originally using for your perusal..FYI, I am constraining a few
> residues for the first 10ps and then releasing them for the other 990ps.
>
> Thanks a ton
> -Rima
>
>
>
> */Peter Freddolino <petefred_at_ks.uiuc.edu>/* wrote:
>
> Also,
> the bad global exclusion count is not necessarily due to a bad psf
> (this
> is actually quite unlikely). I'd recommend having a visual look at
> the
> psf/pdb combo generated by autopsf to make sure all the bonds are
> in the
> right places (things look ok when I try it), and then check that your
> periodic cell is big enough, you've done enough minimization, and
> that
> you try increasing the margin a bit -- these are all far more likely
> culprits for the error you've described.
> Peter
>
> Rima Chaudhuri wrote:
> > Hi,
> > I am still encountering the problem of generating a psf file
> from the
> > attached pdb.
> > It gives me the following errors Although the atoms do exist in the
> > pdb. Secondly, it is trying to make a bond between the C(0) N(1) in
> > residue ASN:299 when it should'nt!
> >
> > building segment U
> > reading residues from pdb file NL63.pdb
> > extracted 300 residues from pdb file
> > Info: generating structure...
> > Info: skipping improper N-C-CA-HN at beginning of segment.
> > Info: skipping conformation C-N-CA-C at beginning of segment.
> > Info: skipping conformation C-CA-N-HN at beginning of segment.
> > ERROR: Missing atoms for bond C(0) N(1) in residue ASN:299
> > ERROR: Missing atoms for improper C(0) CA(0) N(1) O(0)
> > in residue ASN:299
> > Info: segment complete.
> >
> > If I use the autopsf functionality it does not give me any
> errors but
> > later in the simulation the system gives me 'bad exclusion count
> error
> > due to bad initial psf file.
> > Please help!
> >
> > Thanks
> > -Rima
> > University of Illinois at Chicago
> > Dept. of Bioengineering
> > Bioinfomatics
> > PhD candidate
> > rchaud5_at_uic.edu
> >
> >
> >
> >
> > "I can tell you fancy, I can tell you plain, you give something
> up for
> > everything you gain. Since every pleasure's got an edge of pain.
> Pay
> > for your ticket and don't complain" - Bob Dylan
> >
> >
> ------------------------------------------------------------------------
> > Yahoo! Music Unlimited - Access over 1 million songs. Try it free.
> >
>
>
>
>
> "I can tell you fancy, I can tell you plain, you give something up for
> everything you gain. Since every pleasure's got an edge of pain. Pay
> for your ticket and don't complain" - Bob Dylan
>
> ------------------------------------------------------------------------
> Do you Yahoo!?
> Get on board. You're invited
> <http://us.rd.yahoo.com/evt=40791/*http://advision.webevents.yahoo.com/handraisers>
> to try the new Yahoo! Mail Beta.