[SE] Macromolecular Structure Database at the San Diego Supercomputer Center

bourne at sdsc.edu bourne at sdsc.edu
Tue Feb 27 02:33:10 EST 1996

A macromolecular structure database is available to Web users from multiple
servers located at the San Diego Supercomputer Center (SDSC). The database
can be accessed from the URL http://www.sdsc.edu/moose. 

The database, which contains a variety of native and derived data, is
incremented daily from the ftp archives at the Protein Data Bank (PDB). A
summary of the current contents of the database is available. Three classes
of query are supported: text, entity, and pattern. Each returns a subset
of structures which may be examined graphically (3-D rendering, property
profiles, contact maps, snapshot summaries, etc.) or used as input to a
more detailed query. 

Entity queries search for structures based on features of each polypeptide
chain, for example, arrangement of secondary structure elements, percent
alpha or beta, molecular weight. 

Pattern queries search for property patterns based on the amino acid
sequence which may be environmental (exposure, polarity, B values), static
(volume, isoelectric point, hydrophobicity) or a combination of both.
Patterns can be derived from distributions in known structures, in known
sequences (within some threshold), or in relation to absolute values. 

Hashing techniques are used to provide real time query and as an aside
provide current tables of property value distributions across the whole 

Comments are welcome and can be made via the Web page. The SDSC
macromolecular structure database was developed by Ilya Shindyalov, Rich
Toscano, and Phil Bourne.

More information about the Bio-www mailing list