From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Sun Apr 05 2009 - 08:57:59 CDT

On Sun, 2009-04-05 at 04:06 -0400, JT wrote:
> Mike,
> That's pretty strange about the 4th and later ones being drawn ok. I
> think, that without a PSF, VMD draws bonds based on distance. You can

jeff, mike,

under normal circumstances the heuristic distance and covalent atomic
radius based bond guessing method works quite well. generating a psf
file only matters, if you have an unusual geometry or bonding pattern.

here it is definitely overkill and displaying the quoted .pdb file
works just fine with vmd 1.8.6 and the latest development cvs code.

the kind of misinterpreted .pdb file that mike describes usually
happens, if the pdb file is not properly formatted, i.e. columns
are not where they are supposed to be and thus parts of the coordinate
numbers get chopped off. since the file _does_ work for me, it may be
perhaps due to incorrect text file conversion, e.g. trying to load a
dos/windows formatted text file on a unix machine (CR/LF vs. LF).

mike, can you attach the file as .pdb.gz file, i.e. compress with gzip
fist, so that its text format is preserved and then send it to me for
a final check.

cheers,
   axel.

> fix it by creating a PSF and loading it in before or after the PDB.
> I'm sure there are other ways to generate the PSF, but you can use
> PSFGEN or CHARMM to do it. PSFGEN is part of VMD. You can access it
> through its GUI form (autopsf) in the extensions menu of VMD, or
> through its text command form in the text console of VMD. You'll need
> a topology file for the carbohydrate. With that you can easily
> generate the PSF by following one of the tutorials (e.g. http://www.ks.uiuc.edu/Training/Tutorials/namd/namd-tutorial-html/
> ).
> Jeff
>
>
> On Apr 5, 2009, at 2:27 AM, Crowley, Michael wrote:
>
> >
> > Dear VMD folks,
> >
> > I am having much the same trouble as is expressed in the thread:
> >
> > wrong bonds during loading pdb file(protein plus lipids)
> >
> > I am running 1.8.6 and 1.8.5 on mac os-x 10.4
> > Both versions make a mess of carbohydrate pdb's, but not NA or
> > protein.
> >
> > Below is a pdb that shows the trouble, that being bonds drawn
> > between all atoms.
> > The very odd thing is that if I load the same pdb 4 times as new
> > molecule, the fourth one and all subsequent ones are fine.
> >
> > ATOM 2027 C1 NDG A 401 -3.562 15.752 21.491 1.00 39.24
> > ATOM 2028 C2 NDG A 401 -2.121 16.154 21.042 1.00 37.51
> > ATOM 2029 C3 NDG A 401 -1.500 17.322 21.901 1.00 32.28
> > ATOM 2030 C4 NDG A 401 -2.515 18.464 22.046 1.00 34.07
> > ATOM 2031 C5 NDG A 401 -3.934 17.939 22.444 1.00 38.26
> > ATOM 2032 C6 NDG A 401 -5.062 18.974 22.487 1.00 41.57
> > ATOM 2033 C7 NDG A 401 -1.153 14.291 19.650 1.00 34.49
> > ATOM 2034 C8 NDG A 401 -0.090 13.211 19.579 1.00 30.83
> > ATOM 2035 O NDG A 401 -4.384 16.919 21.518 1.00 40.77
> > ATOM 2036 O3 NDG A 401 -0.263 17.836 21.378 1.00 21.01
> > ATOM 2037 O4 NDG A 401 -2.049 19.432 23.009 1.00 31.26
> > ATOM 2038 O6 NDG A 401 -5.072 19.956 21.438 1.00 49.57
> > ATOM 2039 O7 NDG A 401 -1.913 14.458 18.688 1.00 36.13
> > ATOM 2040 N2 NDG A 401 -1.202 15.034 20.775 1.00 35.82
> > ATOM 2041 O1L NDG A 401 -3.561 15.338 22.861 1.00 37.20
> > ATOM 2042 C1 GAL A 402 -2.158 20.795 22.635 1.00 27.02
> > ATOM 2043 C2 GAL A 402 -1.831 21.633 23.842 1.00 22.59
> > ATOM 2044 C3 GAL A 402 -1.728 23.122 23.528 1.00 22.91
> > ATOM 2045 C4 GAL A 402 -0.832 23.359 22.344 1.00 16.49
> > ATOM 2046 C5 GAL A 402 -1.159 22.398 21.229 1.00 20.72
> > ATOM 2047 C6 GAL A 402 -0.146 22.366 20.129 1.00 18.92
> > ATOM 2048 O2 GAL A 402 -2.732 21.366 24.891 1.00 25.35
> > ATOM 2049 O3 GAL A 402 -1.022 23.679 24.627 1.00 26.25
> > ATOM 2050 O4 GAL A 402 0.456 22.968 22.807 1.00 19.85
> > ATOM 2051 O5 GAL A 402 -1.067 21.091 21.782 1.00 26.14
> > ATOM 2052 O6 GAL A 402 -0.766 21.814 18.971 1.00 29.24
> >
> > Any help would be appreciated.
> > Mike
> >
> >
> >

-- 
=======================================================================
Axel Kohlmeyer   akohlmey_at_cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.