Homology recognition using environment-specific substitution tables and structure-dependent gap penalties against the TOCCATA database of annotated structural profiles based on SCOP and CATH.

The TOCCATA database is constructed from a mix of CATH superfamilies and SCOP families, where a consensus group is formed whenever all the members of a (super)family are classified under a single (super)family of the counterpart resource. A sample of a maximum of 25 representative sequences/structures for each grouping is aligned using our in-house program BATON and the resulting alignments are used as basis for FUGUE profiles, after enrichment from homologous sequences by PSI-BLAST. Query sequences are searched against the set of profiles with a modified version of FUGUE, giving the range of the sequence best matched by the significant hits. If the significant hit is part of a SCOP-CATH consensus profile, it is split into a result for each resource, leading to FUGUE (SCOP) and FUGUE (CATH).

Group Leader:Tom L Blundell