It is common knowledge that the code of life is represented by strings consisting of four letters (A, T, C, and G), and the names of Watson and Crick are often referenced to represent the scientific efforts that made understanding the genetic code possible. Leaving aside the (very important and frequently discussed) controversy surrounding the 1962 Nobel Prize--that it, the omission of Rosalind Franklin’s contributions--this slim volume reminds readers of Erwin Chargaff’s important work in the development of the double-helix model. Chargaff provided widely accepted evidence for nucleotide base pairing, that is, the similarity of the frequencies on the one hand of A and T, and on the other of C and G. This is now explained by the double-helix structure, where pairs of these nucleotides bind with each other and therefore allow one strand of the helix to be able to specify the other strand. This simple structure can explain so much about the magic of life.

The pairing of A with T and C with G is just the tip of the iceberg. A more intriguing question is whether there is some relation between the frequencies of nucleotides independent of the pairings behind the double helix, such as the relations in single-stranded deoxyribonucleic acid (DNA). Chargaff studied these proportions in the data available to him in the 1940s and 50s, and proposed his second parity law. Yamagishi’s analysis starts here, using the extensive data available now, including thousands of sets of reference genomes.

The title of the monograph is slightly misleading: it does not set out a formal grammar in the sense usually attributed to the term, that is, as a set of structural rules to explain the formation of words from an alphabet, or alternatively phrases from words. Rather, it discusses the nonrandom distribution of frequencies of substrings on the alphabet, and posits reasons to explain this nonuniform distribution. The analysis is extended not only to individual nucleotide frequencies, but also to short sequences of two nucleotides (AT, AC, GC, and so on), three nucleotides, and so on. Sequences of three nucleotides have an important role, specifically in the regions of DNA that code for amino acids, which are the constituents of proteins that do the actual job of running the functions of cells and organisms. Not all nucleotide sequences code for proteins; some other parts are the promoter sequences that can switch on the transcription of coding regions, while there are many other regions whose functions are poorly understood.

The results described could be of great interest for understanding many intriguing features of heredity and biological robustness, and may be particularly relevant to synthetic biologists who posit new genetic devices coded in DNA or transfer genetic material from one species to another with mixed results. Could the fact that some designs work and others don’t be due to some unknown rules on the constitution of DNA? And crucially, could such rules be coded into formal phrase structure grammars and tools that can, finally, produce engineered biomatter that actually functions as well as biological forms honed through millions of years of natural selection?

The volume is appropriate for anyone interested in computational biology and synthetic biology. It is short and easy to read, and provides a refreshing view on many aspects of DNA structure at the level of abstraction of strings over a very short alphabet. The author shows very clearly and convincingly that the question of what is the actual code of life is far from answered.