Recently some coworkers have been sequencing some cDNAs derived from
differential display analysis so they are likely to come up with some
unknown sequences. We have three ways to proceed in searching the
sequence databases:
1) Blast on the (US) National Library of Medicine computer
(blast at ncbi.nlm.nih.gov)
2) fasta on the GCG package on a local UNIX machine
3) MacVector search of a database on a local Macintosh
A preliminary Blast search with the first 30 bases of sequence of a
differential display product there was a nearly perfect match to a
cockroach protein. However, after 100 more bases was added to the
unknown cDNA, the Blast search came up with several other hits with
lesser matches, but the cockroach cDNA was not among them.
A fasta search came up with a different collection of hits.
Why did the Blast search lose the cockroach sequence when the length of
the query sequence was changed, even though none of the new hits gave as
many bases matched? What are the differences in output due to? What are
the differences between fast and BLAST and the Pustell matrix (used by
MacVector)?
Explanations I have read are either too technical or so simple that the
differences are unclear. Is there a FAQ for laymen that explains the
difference between a local and global homology search? Or explains the
advantages and disadvantages of BLAST vs. fasta searches?
Thanks in advance.
Jim Owens
____________________________________________________
The Secretary denies any knowledge of my activities.
