Motivation: Alignment of biological sequences is one of the most frequently performed computer tasks. The current state of the art involves the use of (multiple) spaced seeds for producing high quality alignments. A particular important class is that of neighbor seeds which combine high sensitivity with reduced space requirements. Current algorithms for computing good neighbor seeds are very slow (exponential).
Results: We give a polynomial-time heuristic algorithm that computes better neighbor seeds than previous ones while being several orders of magnitude faster.