PRSS is a shuffle and alignment tool that is part of FASTA package. It allows an empirical approach to calculating sequence similarity significance by shuffling sequences and re-aligning them a number of times to generate an empirical distribution of scores. This allows the observed score for the best Smith-Waterman alignment to be evaluated in the context of these two sequences. If sequences are biased in sequence composition then the distribution will be skewed toward the higher end of the scores and the observed SW score will not be as significant.
prss34 query hit 1000
query and hit are files with a single sequence each in FASTA sequence format. 1000 indicates the number of shuffle alignments, although this is prompted. The user can also request a uniform shuffle or a window-based approach. The window-approach tends to be the most conservative although YMMV.