ADVANCE and ADAM: two algorithms for the analysis of global similarity between homologous informational sequences

Abstract

Two algorithms for the analysis of global similarity between sequences of informational polymeric molecules (nucleic acids and proteins) are proposed: one (ADVANCE) merely gives a quantification of the global similarity between two sequences, and is very fast; the other (ADAM) also provides an alignment of the sequences. Both are new algorithms, implement Sellers' theorem, do not require parameters, are able to analyze two sequences of 32,000 elements each, and are fast; nevertheless, they differ deeply in the algorithm, in the programming language type, and in their planning, and will thus have to be treated separately. In fact, the different nature of the required output in each program causes the common aim (to obtain power and speed) to be reached only through very different ways and methods.