Section 11-3: Rigorous Searching in the Twilight Zone

[ Previous chapter ][ This chapter ][ Next chapter ]

Very sensitive search program implementations use the "Smith and Waterman" algorithms. In contrast to the heuristic methods mentioned above, rigorous searching will compute a complete alignment of each possible sequence pair of the query sequence versus the database sequence. Depending on the program or implementation, various matrices will be used that that time. Refer to the pairwise comparison section for details.

The framesearch program of the GCG package does this type of search for a protein sequence if a suitable DNA library is specified.

Programs running the Smith and Waterman type of rigorous searching might use quite a long time to achieve completion, or require special hardware in order to complete in shorter time. In particular, searches in DNA databases can take significant resources. Famous programs are 'swsearch' on the Bioccelerator and 'MPsrch' on the MasPar Computer.

Several mail servers offer to run these programs.

If you happen to have access to W.Pearson's sequence analysis software you could try the ssearch program. This requires that you should first convert the sequences of interest to STADEN format with the command tostaden . If you can get hold of an alignment of several sequences, and can produce a profile, use the program 'profilesearch' (see section on patterns .).