1. W. R. Pearson and D. J. Lipman (1988), "Improved Tools for Biological Sequence Analysis", PNAS 85:2444- 2448,
2. W. R. Pearson (1990) "Rapid and Sensitive Sequence Comparison with FASTP and FASTA" Methods in Enzymology 183:63- 98).

Contact:
For Support on this service: nik@genebee.msu.su

The Author:
William R. Pearson (email: wrp@virginia.edu)
Department of Biochemistry
Box 440, Jordan Hall
U. of Virginia
Charlottesville, VA

SHORT SEQUENCES
For very short nucleic acid sequences the strategy recommended is to decrease the word length (ktup) from 6 to 1. In this way a significant increase in sensitivity is achieved without loss of biological significans. Fasta may not return as many hits as blast does in this type of searches but the relevans of fasta results is much higher than with blast.

For Fastf3 and Fasts3 you must use the following format to enter the fragments:

>mgstm1
MGCEN,
MIDYP,
MLLAY,
MLLGY

When aligning two sequences together it is often required to insert gaps in them in order to optimise the alignment. This can be done on the basis of indentities alone, inserting gaps in the sequences as required where there are no matches. However, this is not recommended for biological sequence comparisions because similarities are then not taken into consideration. A scoring scheme, often referred to as a comparison matrix, is used which gives a high positive score when the identical residues or bases are properly aligned. Slightly less if a similarity or homology is posible (i.e. a conservative subsitution) and even negative scores for alignment pairs which are not biologicaly significant.

When two sequences are aligned together a diagonal is created which depicts the best alignment path for these. This diagonal may be broken in places due to mismatches. If there are too many of these the diagonal is subdivided into several smaller ones. In order to make the alignment better gap initiation and gap extension penalties are introduced which penalise the total alignment score.

In general,the lower the gapping penalties, the more gaps and more indentities are detected but this should be considered in relation to biological significance.

Fasta, blast, blitz and clustalw use slightly different terms to referr to gap initiation and gap extension penalties. In general, gapopen and opengap are the former while gapext and extendgap the later.

Some of the later improvement to these programs include the possibility to penalise gaps separately on the database sequences and the query sequence separately. Such is the case of blitz. In clustalw, a gap penalty exists which penalises separetly the length of a gap, closing a gap and the introducion of a pairwise gap in both sequences.