What is DAPS?
DAPS stands for Database of Aligned Protein Structures. DAPS is
based on the FSSP, DSSP, PDB and CATH databases. There also exists a
subset of DAPS known as DDAPS (also pronounced DAPS) - Database of
Distant Aligned Protein Structures. It is a database of structures
that have low sequence similarity but share a similar fold. There are
a number of filters used to make the DDAPS list more useful. The
algorithm requires that an FSSP file exists for one of the members of
a pair and that the other member is listed in that FSSP file. It
requires that each member of the pair be within the CATH database and
share a common CAT classification. It also requires that the secondary
structure can be determined by DSSP.
The database is provided for people requiring such a database
for their research.
For citation purposes use:
Parag Mallick, Danny Rice and David Eisenberg "DAPS: Database of Distant Aligned
Protein Structures"
http://www.doe-mbi.ucla.edu/~parag/DAPS/, 2001.
How is DAPS constructed?
We begin with the set of all chains from the current release of the PDB.
An all on all search is done on the list to find pairs that have the same
fold acoording to both the FSSP and CATH databases and clustered into
groups by a representative structure (representative structures have less
than 25% sequence identity to each other). For each protein pair, regions
aligned by the DALI program are extracted from the corresponding FSSP
file, or recomputed using DALI-lite. In domain DAPS, only regions that
are called "domains" by CATH are included in the alignment. The amino
acid type, secondary structure type, and solvent accessibility are
extracted from the DSSP file and written pairwise into the database.
DAPS is updated with updates of CATH. An URL link to DAPS can be found at
www.doe-mbi.ucla.edu.