A SYNTACTIC PATTERN RECOGNITION SYSTEM FOR DNA SEQUENCES

David B. Searls, Shan Dong · 1993

We review both theoretical and practical results of a linguistic approach to studying the structure of features of DNA sequences. Using generative grammars, complex assemblages can not only be described and analyzed abstractly, but also concretely, such that features can be searched for by a general-purpose parser. Our parser, called GenLang, uses an extended logic grammar formalism and has found features as complex as tRNA genes, group I introns, and protein-encoding genes, within input sequences on a genomic scale. 1. Introduction Recently a number of authors have adopted a "linguistic" view of DNA sequences. Most of their work has involved examinations of the occurrences of "words" in DNA in what is essentially an information-theoretic approach 1;2 , or the use of statistical analyses of vocabularies in the tradition of comparative linguistics 3;4 . These approaches, however, can be distinguished from another methodology in linguistics, pioneered by Noam Chomsky, which attempts...

Read the paper · More papers on PaperTik