VMD-L Mailing List

From: Luis Rosales (ludwig_at_biomedicas.unam.mx)
Date: Fri Nov 12 2004 - 00:44:56 CST

Hi...

Maybe you can do your alignment based on the neighboring residues alpha
carbons???

First you define a selection for each molecule (so that you can move ALL the
protein, cluster included)

set cluster1 [atomselect 0 "all"]
set cluster2 [atomselect 1 "all"]
set cluster3 [atomselect 2 "all"]
...
etc

Now you define your selections for the alignment:

set sel1 [atomselect 0 "name CA and resid w x y z"]
    => Note: This select CA atoms of residues number w, x, y or z from mol 0
set sel2 [atomselect 1 "name CA and resid a b c d"]
...
etc

Then you calculate the best fit for each pair

set aln1 [measure fit $sel1 $sel2]
set aln2 [measure fit $sel1 $sel3]
...
etc

Finally you apply the transformation matrix to each mol

$cluster2 move $aln1
$cluster3 move $aln2

If you need to do the alignment based on the Fe clusters you can define your
selections based on this...

(maybe like: set sel1 [atomselect 0 "type Fe and index x y z"] => this
selects all Fe atoms with index x, y or z
)

Hope this helps,

Luis