AceBrowser Enhancements - BLAST search
The BLAST Search query display uses WU-BLAST 2.0 (could be adapted to use other types of BLAST - the version on the UK CropNet server uses NCBI BLAST) to search a database of sequences (either nucleotide or amino acid) with a query sequence provided by the user.

Example: the BrassicaDB BLAST page on the John Innes Centre server.

Configuration
This display requires the BLAST executables (WU-BLAST) to be installed on your server. You also need to have built a BLAST database with the sequences from whatever acedb database you are constructing the BLAST search display for.

Looking at the source code you'll notice there is some configuration to be done. Generally this will mean changing the path to the BLAST distribution, the path to the BLAST databases, the location of the log file and modifing the database name translation function.

The display checks to see if the databases exist before generating the query form so that the correct types of BLAST can be offered. This is done by translating the database name into the root name of the BLAST database, appending the blast type extention to this root, and checking for that file in the BLAST database directory.

The BLAST database extensions are set in the %blast_exes hash. The distribution file uses "-DNA" for nucleotide sequence databases and "-Protein" for ammino acid sequence databases.

The BLAST databases are generated by:

• Nucleotide sequence
  • Generate a keyset of the DNA class.
  • Export this keyset as DNA sequence fasta file.
  • Ensure the name of the file is consistant with your BLAST database naming scheme.
  • Generate the BLAST database (pressdb).
• Amino acid sequence
  • Generate a keyset of the Peptide class.
  • Export this keyset as Protein sequence fasta file.
  • Ensure the name of the file is consistant with your BLAST database naming scheme.
  • Generate the BLAST database (setdb).