Monday, February 12, 2007

Biology a la Freakonomics: freakology?   posted by p-ter @ 2/12/2007 04:36:00 AM

A review of Survival of the Sickest, by Sharon Moalem with Jonathan Prince

I've often remarked that the field of human genetics is a veritable gold mine for dinner party anecdotes. The material is so rich--superhuman street performers, ear wax, micropenis--you just can't go wrong. Of course, these stories, while sometimes amusing, are insprired by research into serious questions; in the proper hands, genetic data give revolutionary insight into why we are the way we are, how we got here, and where we're going.

It's in this spirit of both clever storytelling and serious inquiry that Sharon Moalem writes Survival of the Sickest, a book grounded in what has come to be called "evolutionary medicine". While the focus of medicine is traditionally on proximate causes of disease (altered metabolic pathways, non-functioning receptors, and the like), Moalem, like the original populizers of the subject Robert Neese and George Williams, wants to find their ultimate causes (the evolutionary forces allowing for the existence of disease). His treatment of the subject here is clearly targeted at the layman, though there is a references section for those who care to dig into the primary literature themselves.

Most of the chapters are centered around the population genetics of a disease. This is fascinating material, and Moalem (along with his co-author) does a wonderful job presenting it. Each chapter starts with an observation--a high prevalance of Type I diabetes in Europeans, the frequency of a genetic disease called hemochromatosis, or the geographic distribution of favism, for example--which sets the stage for a series of anecdotes that eventually leads the reader to his evolutionary hypothesis (for the examples given, these hypotheses are adaptation to cold and resistance to the plague and malaria, repectively). Some of these anecdotes are worth the price of the book alone (though, it must be noted, I didn't pay for my copy, so I suppose I can't judge)--there's an investigation of the biology of a toad that allows itself to freeze solid each winter that is particularly remarkable, and the section on host manipulation by parasites would make Carl Zimmer proud. A large number of human traits are touched on from this perspective--apart from the ones mentioned above, traits like skin color, alcoholism, taste, and even skull shape get mentions[1].

While these chapters are the highlight of the book, an alert reader may notice a couple hints that perhaps the science isn't definitive on some of these stories: first, the oft-added qualifier that a given theory (for example, that the high rate of hypertension in African-Americans is due to selection for salt retention on slave boats) is "controversial", and second, that there are a number of theories for some of the observations. The prevalence of diabetes, for instance, is attributed to metabolic systems unaccustomed to carbohydrate-rich diets (pg. 26), a selective sweep for better cold response during the Younger Dryas (pg. 46), and transgenerational epigentic effects, the so-called "thrifty phenotype" hypothesis (pg. 166). These possibilities are not mutually exclusive, but a reconciliation of all of them would certainly have been desirable.

But overall, Moalem gives a striking demonstration of the power of evolutionary hypotheses to discover new apects of our biology (as long as one keeps in mind that sometimes those hypotheses turn out to be dead wrong). On straying from the basic population genetic framework, he becomes a little hyperbolic[2], but as an advertisement for the use of evolutionary information in understanding disease, Survival of the Sickest is a resounding success. Altering Theodosius Dobzhansky's famous dictum has become almost cliche, but I'll do it anyways: this book is a convincing argument that nothing in medicine makes sense except in the light of evolution.

[1] Moalem gives a cursory look at the concept of "race" with regard to these traits, but essentially chooses not to discuss it, preferring to cite a Nature Genetics editorial from 2001 as saying that "population clusters identified by genotype analysis seem to more informative than those identified by skin color or self-declaration of race". Regular readers know that much has changed since 2001; in particular, there doesn't seem to be much of a distinction between genetic clusters and clusters based on self-declaration of race [Tang et al. 2005].

[2] The chapter on "jumping genes" is particularly filled with perplexing claims about the nature of evolution. On noting that some "knockout" mice--mice in which a gene has been disabled--show no observable defect (as an aside, the mouse geneticists I know tend to be skeptical of such reports--is it really possible to rule out any effect? How many possible phenotypes were tested?), Moalem writes, "If removing whole genes often has no effect on a creature, how could such minor changes be the only chance for the evolution of a new species, or even the successful adaptation of a new one? They probably can't". After several chapters on the evolution of humans by "minor" changes (much variation in skin color is controlled by a single base pair change, lactose tolerance as well), this statement is particular puzzling. His point is that retrotransposons are perhaps mobilized by stress and can get into the germ line, which is true enough; retrotransposon insertion can serve as a source of variation. But I wouldn't stop studying "minor changes" like SNPs or copy number variants any time soon. And the suggestion that retrotransposons are responsible for observations of "puncuated equilibria" in the fossil record is perplexing as well; it seems Moalem felt he needed to mention the theory of punctuated equilibrium, and hell, the section on jumping genes was as good as any.