This workshop will deal with string alignment problems in which one wishes to gain information about a newly sequenced piece of DNA by comparing, or aligning it, with a sequence of known function or structure. Detection of similarity between two different molecular sequences has led to the discovery of shared phenomena. (We have already referred to the discovery that the sequence for platelet derived factor, which causes growth in the body, is 87% identical to the sequence for v-sis, a cancer-causing gene, which led to the discovery that v-sis works by stimulating growth.) The quality of a match between two sequences can be determined by a scoring matrix and a charge for introducing gaps in one of the sequences to get a better match, and then a dynamic programming algorithm can be used to determine the largest number of places where two sequences (gaps added) agree. The theory of random graphs can be used to compare two random sequences and predict how good a match one can expect. Work on determining similarity between pairs of sequences can be expanded to work on detecting matches among a whole cluster of such sequences, and then algorithms or heuristics for determining clique-like structures in corresponding graphs can be useful for finding patterns. In this workshop, we will explore all of these ideas. We will investigate dot-matrix methods, global alignments, local alignments and hash coding methods, multiple alignments, measures of aminoacid similarity, and statistical significance of alignments. This workshop will also be closely coordinated with our Algorithm Implementation Challenge.