From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Mon Oct 13 2008 - 18:23:35 CDT

On Mon, 13 Oct 2008, Lan Hua wrote:

LH> Hi,
LH>

[...]

LH> However in the pdb file downloaded from pdb bank the coordinates for this
LH> residue with resid 31 look like:

[...]

LH> ATOM 97 CA ALYS A 31 17.941 -5.503 5.166 0.25 5.00
LH> C
LH> ANISOU 97 CA ALYS A 31 936 468 493 67 11 -34
LH> C
LH> ATOM 98 CA BLYS A 31 18.003 -5.680 5.325 0.25 5.14
LH> C
LH> ANISOU 98 CA BLYS A 31 876 425 650 -63 -22 -62
LH> C
LH> ATOM 99 CA CLYS A 31 17.758 -5.524 5.116 0.50 5.89
LH> C
LH> ANISOU 99 CA CLYS A 31 1129 534 573 -4 5 -84
LH> C
LH> ATOM 100 C ALYS A 31 18.911 -5.892 6.320 0.50 4.65
LH> C
LH> ANISOU 100 C ALYS A 31 802 436 528 88 126 -29
LH> C
LH> ATOM 101 C CLYS A 31 18.617 -5.972 6.302 0.50 6.75
LH> C
LH> ANISOU 101 C CLYS A 31 1074 723 764 64 126 45
LH> C
LH> ATOM 102 O ALYS A 31 18.659 -6.843 7.086 0.50 5.54
LH> O
LH> ANISOU 102 O ALYS A 31 1010 467 625 145 176 32
LH> O
LH> ATOM 103 O CLYS A 31 18.289 -6.912 7.046 0.50 7.74
LH> O
LH> ANISOU 103 O CLYS A 31 1256 816 867 56 181 118
LH> O

please note, that altloc B coordinates are missing here for C and O.

[...]

LH> So, the atomselection "protein" failed to choose the protein atoms with
LH> alternate location B for this residue. Of course, I can use the

if you look at the source code you can easily find that the
requirement of a residue (or its altloc) to be classified
as "protein" is the presense of an N, CA, C, and O atom for
the backbone. the altloc B does not have this and consequently
VMD does not classify this location as protein.

i don't see how this can be made differently. i'd rather say the
pdb file is incomplete/incorrect and VMD is right, unless somebody
can show a PDB rule for how to map missing altloc coordinates to
existing ones.

cheers,
   axel.

LH> atomselection like "protein or altloc B" instead. But I feel confused and it
LH> is not convenient if I need to check which protein atoms with alternate
LH> locations are not chosen. Is there a good day to choose all the protein
LH> atoms with different alternate locations?
LH>
LH>
LH> Thank you very much
LH>
LH> Lan
LH>

-- 
=======================================================================
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.