sequence alignment poorer than 3D alignment

Simon Brocklehurst simon.brocklehurst at
Thu Sep 17 06:37:21 EST 1998

Michel PETITJEAN wrote:

> Is there some pair of proteins having a sequence homology (% similarity)
> significantly lower than their % 3D-similarity ? (any definition of
> these percentages should work). References are welcome. Thanks.


There are many examples of what you're looking for.  For example, have a
look at the collection of 3-D structure-based alignments, and corresponding
pairwise sequence identity matrices on the Web server in Tom Blundell's
group in Cambridge:

Depending on your requirements, you may wish to focus on a lesser number of
cases where there is no detectable level of similarity at all at the
sequence level (e.g. no conservered hydrophobic residues). These can be
identified by 3-D structure superpositions, and predicted by methods which
take into account correlated amino acid residue substitutions (e.g.

Simon M. Brocklehurst, Ph.D.
Head of Bioinformatics
Cambridge Antibody Technology
The Science Park, Melbourn, Cambridgeshire, UK
mailto:simon.brocklehurst at

More information about the Molmodel mailing list