Geoff Barton gjb at
Thu May 30 04:10:39 EST 1996

Michael Bunce wrote:
> I am trying to get some idea of the structure of Calpastatin (the endogenous
> inhibitor of a group of proteases named the Calpains).  It is a protein with
> four domains and approx 700 amino acids.  Is it possible to predict the the
> structure of this? or even a 12 amino acid (repeated) inhibitory sequence in
> the structure?  I would be thankful of any advise in this area.
>      Mike Bunce

The straight answer is probably not.  However, if the protein shows
clear sequence similarity to another of known 3D structure then you 
may be able to build a low resolution model.  Techniques for fold
recognition might give you an even lower resolution model, and secondary
structure prediction methods may suggest the location of loops and
regular secondary structures.

I suggest you look at Rob Russell's nice on-line guide to protein 
structure prediction.

This gives advice on how to predict structure and provides numerous
references and links to most current methods.

With 700 residues you will need to subdivide the protein into 
reliably identified domains.  It is also a big help to work from a 
multiple alignment rather than a single sequence.

I hope this helps,


Geoffrey J. Barton, Laboratory of Molecular Biophysics, University of
Rex Richards Building, South Parks Road, Oxford OX1 3QU, U.K.
mailto:gjb at, Tel: +44 1865 275368, Fax: +44 1865 510454,,

More information about the Proteins mailing list