3D modelling services 
Welcome to the BIOcomputing 3D modelling unit databases page. This page holds
the databases we designed and maintain as well as several other very useful
databases that were designed by our friends while they were still members
of the Sander group at the EMBL or EBI in the period that roughly spans the 90s...
Detailed acknowledgements are at the bottom of this page.
- DSSP: Definition of Secondary Structure of Proteins
- FSSP: Families of Structurally Similar Proteins
- HSSP: Homology-derived Secondary Structure of Proteins
- PDBSELECT: Unique subset of 3-D protein structures
- PDBFINDER: Your entry into the PDB database
- PDBREPORT: Structure verification reports for PDB structures
- GPCRDB: G protein-coupled receptor (GPCR) data
- Modules in extracellular proteins
DSSP is the de facto standard for the description of secondary structure of proteins.
DSSP was originally written by W. Kabsch and C. Sander, and was later 'optimised' by
M. Scharff. For every PDB file that contains
at least one protein there also exists a DSSP file. You can get access to the
database
of DSSP files and to the executable.
Please fill out the license form
if you want to use the DSSP software.
FSSP is the database produced by the program DALI
which was written and is still maintained by Liisa Holm. The
FSSP database is the most
extensive (and I personally think best) 3D structure comparison database. you can use
DALI via a WWW based server.
The HSSP database is a database of homology-derived secondary
structure of proteins. It was designed by Chris Sander and
Reinhard Schneider. For every PDB file that contains
at least one protein there also exists a HSSP file. An HSSP file holds a multiple sequence alignment of
all SwissProt files that are significantly homologous to a PDB file.
You can look at the abstract of the
paper describing HSSP
or get FTP access to the database itself
The PDBSELECT database is a subset of the structures in the
PDB that does not contain (highly)
homologous sequences. It was designed by Uwe Hobohm and Chris Sander.
You can read an explanation of the PDBselect database
and the method used to construct it or
get FTP access to the database itself.
The PDBFINDER database is a database that is constructed from the
PDB, DSSP and
HSSP
databases (using a
PERL script).
Many of the fields contained in the PDBFINDER
database are difficult to access from the original databases. Some
information is retrieved from the original literature.
The PDBFINDER database was constructed by
Rob W.W. Hooft,
Michael Scharf,
Chris Sander, and
Gert Vriend.
An overview of the PDBFINDER database
and FTP access to the database itself
are available.
Most PDB files contain some anomalies,
flaws, and sometimes even serious errors. The PDBREPORT
database was made by Rob Hooft and holds
for each PDB file a report which lists everything
we could find that is sufficiently different from what one would expect that it is
interesting to look at. These remarks indicate minor inconveniences, very interesting
aspects of the molecule, but often also errors.
The GPCRDB is an information
system for G protein coupled receptor data. This is an international
collaboration.
The curator/coordinator is Florence Horn.
The module pages contain information and
research tools on mobile protein domains, developed, collected and maintained by
the Bork group and various
collaborators.
Use
E-mail if you have questions.
(C) G.V., J.W 5-Mar-1999