VMD-L Mailing List

From: Peter Schellenberg (peter.schellenberg_at_ipht-jena.de)
Date: Sun Aug 28 2005 - 05:28:29 CDT

Hi,

I tried to apply this proposal before, but it did not work to my satisfaction. It seems
that the selected atoms 1 to 4 did not match with the identical atoms 1 to 4 of the
second selection. For illustration I wanted to overlap two almost identical digoxin-
type steroids which are bound to an antibody-FAB fragment (1igj.pdb) and an
anticalin (1lnm.pdb) and just selected atoms appearing in both steroids, either by
index or by type:

set antibody [atomselect 3 "all"]
set steroid_dtx [atomselect 1 " type O14 O21 O23 C18 C5 C6 C12 "]
set steroid_dgx [atomselect 3 " type O14 O21 O23 C18 C5 C6 C12 and chain Y "]
set M [measure fit $steroid_dgx $steroid_dtx]
$antibody move $M

( O14, O21 etc refer to the same type of atom in both steroids). The steroids
overlapped along their long axis, but perpendicular to each other, and not in the way
that O14, O21 etc were close to each other. The same happened if I used the index
selection.
Is there any solution to this problem?

Kind regards Peter

Dr. Peter Schellenberg
Institute for Physical High Technology Jena
Dept. of Biotechnical Microsystems
Albert -Einstein -Str. 9
D-07745 Jena
+49 -3641 -206 308