When I introduce evolution in class, I tell my students that the strongest evidence that Darwin was right is the success of modern biology. Evolution is the theoretical foundation upon which life science is built. The fact that that life science works so well suggests that its unifying theory has a thing or two going for it.
The basic idea is that if organisms such as humans and yeast are descended with modification from common ancestors, then we should find some of the same genes in both creatures—which, of course, we do. We also ought to find some of the same gene teams in both creatures. That is, sets of genes encoding proteins that work together to carry out some biological function. We find these too.
Marcotte and colleagues reasoned that we should therefore be able to identify genes associated with, say, mammalian diseases by looking for mammalian gene teams whose members are also teammates on, for example, yeast gene teams.
As a demonstration, the researchers assembled a database of genes known to occur in both mice and yeast. They identified gene teams as sets of genes associated with a particular phenotype, such as a disease in mice or a drug sensitivity in yeast. Then they looked for human and yeast gene teams with overlapping membership.
One pair of overlapping teams consisted of a set of eight genes known to play a role in the development of blood vessels (angiogenesis) in mice and a set of 67 genes known to be associated with sensitivity to the drug lovastatin in yeast. These teams formed a pair because of the five genes that belonged to both.
The relationship between the yeast lovastatin team and the mouse angiogenesis team suggested that both teams might be bigger than previously recognized and that they might share more players in common. That is, the three angiogenesis genes from mice that are not already known to play for the yeast lovastatin team might, in fact, be members. And the 62 lovastatin genes from yeast not already known to belong to team angiogenesis in mice might play there as well.
Starting with this list of 62 candidates, the Marcotte and colleagues ran experiments (in frogs) confirming that at least five of the genes do, indeed, participate in the development of blood vessels. Three more of the candidates turned out to have already been identified as angiogenesis genes, but not included as such in the researchers’ database. Eight out of 62 is a much higher rate of hits than you’d get if you just picked genes at random and checked to see if they influence angiogenesis.
To recap: Scientists have used genetic data from yeast (an organsism with neither blood nor vessels) to identify genes that influence blood vessel growth in vertebrates.
I find this both astonishing and wonderful.
Descent with modification is a more powerful idea than Darwin himself could have possibly imagined.
The research report by Marcotte and colleagues is: McGary, K. L., T. J. Park et al. 2010. Systematic discovery of nonobvious human disease models through orthologous phenotypes. Proceedings of the National Academy of Scienes 107: 6544–6549.
The story from the Times is: Zimmer, Carl. 2010. The Search for Genes Leads to Unexpected Places. The New York Times 26 April (online) / 27 April (print).