Mutants in the engine room

The study of molecular evolution regularly attracts headlines, usually within the context of human origins. A recent example described the analysis of DNA taken from Neanderthal material, concluding that these ancient humanoids were probably not our direct ancestors. But serious work on human molecular evolution began more than ten years ago, with the first studies of extant human genetic variability. These suggested that all humanity descended from a group of females who carried a specific maternal DNA type and who lived in Africa 200,000 years ago. In other words, we all derive from one African female genotype, affectionately known as Eve. The descendants of females carrying other genotypes became extinct. Thus was born the out-of-Africa theory of recent human origins.

Clearly, we are most fascinated by studies that throw light on our own murky past and, while these remain controversial, the molecular revolution has generated an enormous impetus into evolutionary studies. One of my colleagues commented many years ago that the trouble with evolution was that there was no such thing as a bad idea, because evolutionary hypotheses were so difficult to test rigorously. In the early 1980s, he was correct, but times have changed and the spectacular progress in topics like human evolution owes an enormous debt, not only to developments in semi-automated laboratory techniques that allow accumulation of thousands of DNA sequences, but also to rigorous statistical treatments of these data. The molecular evolutionist therefore studies variation in DNA sequences between individuals, populations, and between species, both extant and extinct.

Comparing DNA sequence variation is not straightforward. There are many subtleties and there lurks lots of information within DNA that is not immediately obvious. Wen-Hsiung Li's book focuses on how to make sense of the data. The author's introductions are brief, covering in 60 pages the basics of gene and protein structure, and population genetics, the cornerstone of the subsequent mathematical analyses. This is obviously not a book for the first-year undergraduate. Initially I found it a little intimidating - my maths is poor, O-level standard - but I managed to trek my way through thanks to the examples.

Up to chapter 7, the narrative builds up to how phylogenetic trees, which chart the evolutionary relationships between a group of organisms, can be constructed, and then moves on to whether there are such things as molecular clocks, ie, do proteins, and DNA evolve at a uniform rate in all lineages? The "molecular clock" idea provided a powerful stimulus to the "neutral theory", for which evolution is little more than random events dictating whether new mutations are incorporated or lost. This challenged the Darwinian view that evolution occurred by natural selection, and, elegantly dressed up in mathematics, provided the null hypothesis against which to test whether natural selection had driven DNA sequence evolution. Li devotes an important section to the relevant statistical tests but an appendix with worked examples would have been useful.

Chapters 10 to 14 review the evolution of proteins by duplications and domain shuffling, how multigene families evolve together "in concert", how transposable elements jump between species and overall genome organisation. The final chapter concludes that while selection has contributed to the "wondrous diversity of plant life that is so apparent as we look around us", most evolutionary change is driven by mutations that have no visible effect, contrary to the neo-Darwinian view.

The study of molecular evolution will inevitably modify many of our long-held beliefs about genetic variation and the evolutionary process. The development of mathematical tools has moved the field from armchair theorising to a more pragmatic appreciation of how things were and how they came to be. Li's book carries this challenge to the next generation of advanced undergraduate and postgraduate students and I thoroughly recommend this text to them.

Charalambos P. Kyriacou is professor of genetics, University of Leicester.