The main difference between the Needleman-Wunsch and the Smith-Waterman algorithms is the fact that the Needleman-Wunsch algorithm tries to make a global alignment, over the whole length of the given proteins, whereas the Smith-Waterman algorithm finds a local alignment. A local alignment searches for regions of local similarity between two sequences and need not include the entire length of the sequences. This can be useful when searching for matches between small regions of sequences, for example between protein domains. If two sequences are known not to be homologous throughout their entire length, a local alignment should be the method of choice. Usually the two methods will give similar answers but if the homology is distant, a local alignment will be more likely to find the remaining regions of homology.

The web server parameters are filled out in exactly the same way as for the Needleman-Wunsch server.