superimposing molecules

Michel PETITJEAN ptitjean at
Mon Dec 9 03:38:43 EST 1996

In article <58cn55$1cgu at>,
iropson at (Ira Ropson) wrote:

>I would like to superimpose three structures in the same family of
>proteins for comparison purposes for a paper I'm writing on the folding
>of proteins having considerable sequence heterogeneity but very similar
>X-ray structures. The proteins are intestinal fatty acid binding
>protein (131 aa), cellular retinol binding protein 2 (133 aa), and
>cellular retinoic acid binding protein (136 aa). I can superimpose them
>by eye useing Berkeley RASMAC, and they look pretty good, but I would
>like to do a better job by mathematically fitting the backbones of
>these proteins to each other. I have insightII available to me on an
>SGI, but it states that you have to have the same number of atoms to do
>the superposition. Does anyone know of an easy way to do this
>(preferably without spending any money)? Is there an easy way to do
>this in insightII?

If you can't produce a pairwise correspondence between the two subsets
to align, the rms methods cannot be used. In this situation, a more
general method is required, as that which appeared in J.Chem.Inf.Comput.Sci.
1996,36[5],1038-1049. This latter produces the pairwise correspondence
rather than reading it. The ternary correspondence for three subsets can
be deduced from the pairwise correspondences.

Michel Petitjean,                     Email: petitjean at
ITODYS (CNRS, URA34)                         ptitjean at

More information about the Molmodel mailing list