This is a very comprehensive survey of the field of RNA viral evolution and it deserves to be a very highly cited work because it has been for some years and still remains the main text in the area. But as my title suggests it is just horribly full of contradictions and holes. One second Rabies is reported in 2300 B.C. and is 4300 years old and the next its tree says it is less than 2000. The book ends up asking more questions and revealing more about or complete ignorance of how RNA viruses evolve than it shows about what we know.
People use coalescence as it is the latest technique in phylogenetics but it is really inapplicable in viruses because of the infinite allele assumption which does not hold, unless you are looking at small time frames - so maybe a couple of years and certainly not decades. It will also be very sensitive to sampling as this will badly affect population inferences.
The author then tries to paint quasi-species into a corner by redefining it as a straw-man that he can then knock down, but this has more to do with rhetoric and politics than science. First evolution rates need to be fixed and evolution needs to be neutral (or coalescence does not work) then they vary widely and there is selection. There is just no consistency of argument but this reflects a field in flux, which is quite frankly a mess. There is no mathematical or theoretical consistency and most assumptions are not true in the case of viruses. My strongest criticism is the author's lack of understanding of proteins and structure. Evolution can mutate DNA forever in any way it likes by random changes but selection happens at the protein level and they have a very well defined landscape. Proteins cannot evolve without limit, they are constrained. Just look at the differences in evolutionary rates between protein families in the same organism to see the effects of constraints - cytochrome C is a good example compared to immunoglobulins. There is a clock for mutation but selection will have to go in fits and starts as different parts of the protein space become available. That is why dating beyond very short periods using coalescent methods is dangerous.
So why 4 stars - well because it is currently definitive, nobody else has come close and for all its faults it is essential reading even if it is just to know the current challenges in the field and how many opportunities there are for future research.
on 18 December 2009
A great book. Probably the only book i have seen out there which explains the process of viral evolution so thoroughly and clearly. The book provides a holistic view of the mechanisms of viral evolution, and their outcomes, which i have not found in other books. I particularly liked the in-depth examples e.g influenza, hiv, which are very useful for lecturers and students alike. These examples provide an interesting read to all biologists, not just the ones studying viruses.