identification of specific words), syntactic (the grouping of words into grammatically correct phrases), semantic (the assignment of meaning to words and phrases), and pragmatic (the role of a piece of text in the larger context). These match entirely well to genomic analysis: grouping bases into codons, genes, the function of the resulting protein, and the role of that protein in the larger molecular system.77

Linguistic analyses can reveal or explain relationships between bases that are far apart in a sequence. For example, an RNA structure called a stem-loop has a palindrome-like sequence, with Watson-Crick pairs at equal distances away from the center. Traditional probabilistic or pattern-searching approaches would have some difficulty recognizing this structure, but it is quite simple with a grammar that produces palindromes. Some sequences of nucleic acids result in ambiguous linguistic interpretations; while this is a difficulty for computer languages, it represents a strength of biological linguistic analysis, because these ambiguities correctly represent alternative secondary structures.78

This approach has been fruitful for analyzing genetic sequences and characterizing the complexity and structure of genes. GenLang, a software system that employs linguistic approaches, has successfully identified tRNA genes, group I introns, protein-encoding genes, and the specification of gene regulatory elements.79 Other important findings include placing RNA in the Chomsky hierarchy as at least beyond context-free languages. Finally, the approach provides a powerful tool for understanding the evolution of nucleic acid sequences; since the first sequences were most likely random (and thus regular languages), there must be a mechanism that somehow promoted sequence language into more powerful linguistic categories. This can be seen as an algebraic problem of operational closure, and the question is, For which string operations are regular languages and context-free languages not closed?80


D.B. Searls, “Reading the Book of Life,” Bioinformatics 17(7):579-580, 2001.


D.B. Searls, “The Language of Genes,” Nature 420(6912):211-217, 2002.


D.B. Searls, and S. Dong, “A Syntactic Pattern Recognition System for DNA Sequences” in Proceedings of the Second International Conference on Bioinformatics, Supercomputing, and Complex Genome Analysis, H.A. Lim, J. Fickett, C.R. Cantor, and R.J. Robbins, eds., World Scientific Publishing Co., pp. 89-101, 1993.


D.B. Searls, “Formal Language Theory and Biological Macromolecules,” Series in Discrete Mathematics and Theoretical Computer Science 47:117-140, 1999.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement