ECCB 2002 Poster sorted by: Author | Number

Next | Previous poster (in order of the view you have selected)

Title: Iterative Blast Machine
P35
Fleuren, Wilco; Bettler, Emmanuel; Vriend, Gert

e.bettler@cmbi.kun.nl
CMBI

During the development of a database, one major problem is to find as many data as possible about the class of protein of interest. By searching for keywords, it is possible to find only identified or putative proteins: new genomes, for example, are inaccessible. By basing our research on the sequence instead, we can overcome this problem. We have developed software using BLAST to retrieve known and unknown proteins of a same class. Starting with a sequence, and making a first call to BLAST called `iniBLAST', we then put in a database all the sequences from `iniBLAST' whose `evalue' is less than a previously defined cutoff.
Then we run BLAST for each sequence in the database. Each BLAST after the iniBLAST is called iterative BLAST. After each iterative BLAST, the programme checks if a sequence in the BLAST output is already in the database, and adds it if necessary. Thus, with the MCSIS technology, it will be possible to build new databases on any kind of class of proteins.