Even the most elaborate pictures of the tree of life you can find online are gaunt shadows of life’s full diversity. In my new pieces for New York Times, I write about a team of scientists who are setting out to build a tree with every described species on Earth–and program it so that the entire scientific community can help tease out its branches and add more branches as they discover the six, sixty, or six hundred million more unnamed species on the planet.
27 plus ones
Shared publicly•View activity
View 6 previous comments
- I quotedin my story--he is absolutely right about the challenges. I think the tree-builders are well aware of the challenges ahead.
Thanks for your comment. DNA-based trees are no different in a conceptual way from traditional ones based on morphology. Scientists use statistical methods to work out the most likely arrangement of the branches. These models today are very sophisticated, including information about how DNA tends to mutate for example. And it's possible to use different lines of evidence to test the hypothesis generated from a particular set of data. (DNA, for example, confirms the tree of marsupials based on morphology.)
Inserting the traditional Linnean classification into the tree will be just a first step, until scientists get around to working out the precise relationships between the species.Jun 5, 2012
- "until scientists get around to working out the precise relationships between the species."
I've become persuaded that there will be no "precise relationships between species" just as there is no precise location of electrons. Because of issues like lateral gene transfer and others mentioned by I think there will be statistical relationships between species.Jun 5, 2012
- Actually, scientists can get pretty precise, even taking into account lateral gene transfer. For example, among living primates, humans are most closely related to chimpanzees, although certain alleles in the human genome are closer to gorillas thanks to incomplete lineage sorting. Humans also mated with Neanderthals and Denisovans, and we can see which parts of the human genome we inherited from them. And the human genome also contains endogenous retroviruses. But it's not as if it's all chaos--we don't have DNA from apples in our genome, and our ancestors never interbred with dolphins, in other words.Jun 5, 2012
- Perhaps not "chaos", but interesting things do happen, e.g. aphids have acquired carotenoid biosynthesis genes from fungi http://dx.doi.org/10.1126/science.1187113 (not quite apple DNA in the human genome, but not far off). The bigger question is to what extend lateral gene transfer renders the notion of a "tree of life" problematic. There is some elegant (if scary) mathematical work arguing that lateral gene transfer doesn't necessarily invalidate the notion of a tree, e.g. http://dx.doi.org/10.1016/j.jtbi.2010.05.031Jun 6, 2012
- Jun 6, 2012
- http://dx.doi.org/10.1073/pnas.1116871109 looks interesting. I especially like the notion that lateral gene transfers are helpful in rooting trees. Regardless of whether there is sufficient tree-like structure to talk of a "tree of life", I like being able to see individual trees, so I'm hoping the Open Tree of Life will let me see those. A visual metaphor I like is http://photosynth.net/ where you can see the individual photos as well as the 3D synthesis.At a quick glanceJun 6, 2012