Computational Biology: A synopsis

The main strategy of algorithms in this class is to substitute “repeated strings”, i.e., exact replicas of a substring in a string, within the text with a convenient coding scheme. Those methods generally fall into the Macro Model for Data Compression introduced and studied by Storer and Szymanski (1982). The well known algorithms by Ziv and Lempel (1977, 1978), as well as many others, turn out to be instances of that paradigm, which is based on the principle of finding exact repeated substrings in a string and replacing them by pointers to a common copy. That is, redundancy in a string is translated into the notion of exact replicas of substrings. For instance, the Ziv and Lempel (1977) algorithm first parses the input string into distinct phrases and then each phrase is encoded by a triplet of numbers. An example of the parsing rule is provided in Fig. 1.