A Bauhaus for Biologists: 
An Introduction to Protein Architecture by A. M. Lesk

Paul Harrison & Mark Gerstein

Department of Molecular Biophysics and Biochemistry
Yale University
New Haven, CT 06250

Architectural styles of buildings come and go, whereas the
architecture of proteins has been with us for eons. Arthur Lesk gives
us an insightful survey into this most fundamental topic. It is
particularly timely for a new text on this subject as the current
availability of extensive web-based information resources for protein
sequences and structures modifies the traditional role for the
textbook.

Arthur Lesk, the author of An Introduction to Protein Architecture, is
one of the pioneers of the analysis of protein structure and its
evolution.  He was also one of the original developers of computerized
molecular graphics for proteins in the early 1980s.  So, needless to
say, this book is brilliantly illustrated with an extensive array of
large, clear stereo diagrams for the structures or motifs that are
referred to.  Moreover, we believe that the use of stereo figures is
essential for truly visualizing the complex 3D geometry of
proteins. Indeed, this is a substantially graphical textbook (about
half of the total page count, discounting the structural atlas
appendix, is made up of figures).  Set in an uncomplicated
easy-to-read style and logically structured, the text is seldom
wordy. There are brief sections on the basic building blocks of
protein structure followed by discussion about how different proteins
can be classified, with emphasis on web databases. The general
discussions of protein evolution and protein motions are particularly
good.  Detailed studies, which stand well on their own, are presented
of immune system proteins, serine proteases and NAD-binding domains in
dehydrogenases.  In particular, an up-to-date introduction to
structural immunology provides much information on the conformations
of antibodies (with a discussion of canonical loop structures in
antibodies), MHC proteins, T-cell receptors and their interactions.
There is some reference at appropriate points to methods of protein
structure prediction (although there could be more about topics such
as fold recognition or secondary structure prediction).  A nice
summary is provided of the biannual CASP competition for assessment of
prediction methods and the section on homology modeling is strong.

This textbook tackles well the problem of interfacing with internet
information resources. Extensive reference is made to web-based
catalogues of protein structure and search tools for protein structure
and sequence.  The author has designed a special sort of exercise
called the 'weblem'.  For this, the student is asked to use web tools
to solve simple tasks that help in understanding the links between
protein structure, sequence analysis and protein structure prediction.

There are (inevitably) some omissions in the subject matter in this
book.  Some detailed discussion of protein-nucleic acid binding and
the structural motifs involved would be welcome.  Also, a section on
membrane protein structure would be desirable (although there is a
very good opening chapter specifically describing the photosynthetic
reaction center). Finally, while genomics and structural genomics are
certainly mentioned, more emphasis could have been devoted to how
"global" genomic surveys of large numbers structures are changing our
view of what is typical and atypical in protein architecture.

These points aside, this text by a leading expert in the field is to
be highly recommended for anyone wishing to learn about protein
architecture in context of the current structure of bioinformatics
information resources.