In order to measure genetic distances, it is first necessary to find homologous sequences and align them. Then there are several ways of measuring distances, based on different models for base frequencies and substitution. And Page and Holmes touch on the complications of working with proteins rather than genes.
There are many approaches to constructing evolutionary trees from sequences. Distance methods first find the distances between sequences and then use those, while discrete methods consider each nucleotide or site directly. The two major discrete methods are maximum parsimony and maximum likelihood; their advantages and disadvantages include a tradeoff between power and computational cost. And of course we want to be able to test our inferred trees, and to put confidence limits on them.
Turning to models of molecular evolution, Page and Holmes focus on the neutral theory, describing its refinement under criticism. They look at the relationship between functional constraints and rates of substitution, patterns of base and codon distribution, the molecular clock, nearly neutral theory, variation within species, and evidence for natural selection at the molecular level.
A final chapter covers applications of molecular phylogenetics: organismal phylogeny and the choice between combining data sets or doing analyses separately, the relationship between gene and species trees, host-parasite cospeciation, finding the age of taxa and rates of diversification, and molecular epidemiology. (There is no discussion of software tools, with algorithms discussed in the abstract.)
Molecular Evolution is a textbook, complete with a bullet-list summary at the end of each chapter and "further reading" notes. But it's a nicely organised volume — it's the kind of textbook that deserves to be read widely, not just by students for whom it's a set text.
March 2004
- External links:
-
- buy from Amazon.com or Amazon.co.uk