Suboptimal alignment

Suboptimal alignment is an extension of local alignment that finds multiple similar regions in two given genetic strings; for instance, if we want to find all occurrences of a certain repeat, perhaps one corresponding to a transposon. Suboptimal alignment works by finding an optimal local alignment, then discarding the two substrings attained in the alignment and starting over using shorter strings to find more regions of similarity.

As a result, we will obtain a collection of disjoint regions of similarity in the two sequences we are comparing; the "best" match will be provided first, and then matches of typically decreasing quality will be provided. It is the user's prerogative to determine when suboptimal alignment stops returning regions of similarity and starts detecting noise.