Genomic Grammars: Teaching Bioinformatics Using Language Theory

Language theory is an important part of engineering. It is usually taught in compiler theory, operating systems, and other courses. Therefore, the use of language theory, which is familiar to most engineering students, is a good tool to use to introduce bioinformatics. There are three elements to a language: alphabet, grammar, and semantics. The alphabet comprises the words, the grammar defines the rules, and the semantics give the meaning of the language. With respect to genomics, the genomic alphabet has been known for a long time: in DNA it’s the set {A, C, T, G} and in RNA the set {A, C, U, G}. The words in genomics are the genes, of which many have been identified. The semantics, or meaning, of the genes may not be known though. With the rush to patent genes during the near-past, shot-gun methods and others were used. Thus, genes were identified, without understanding the genes’ specific functions. This is akin in language theory as identifying the words (genes) of the language but not their meaning (specific functions). This paper describes grammars, not the semantics, of language theory. It also discusses the representation of genes using grammars. Furthermore, it gives examples of project assignments for engineering students. By teaching bioinformatics using familiar tools, such as language theory, the engineering student gains knowledge of biology, bioinformatics and the relationship of engineering principles to other disciplines.