Developing A Tree of Life Browser

Over the years researchers have sequenced DNA from thousands of species from jungles, tundras and museum drawers. They have used supercomputers to crunch the genetic data.

Dr. Michael Sanderson is a biologist at the University of Arizona and is part of an effort to figure out how all the estimated 500,000 species of plants are related to one another. Over the years researchers have sequenced DNA from thousands of species from jungles, tundras and museum drawers. In their efforts they have used supercomputers to crunch the genetic data and have gleaned clues to how today’s diversity of baobobs, dandelions, mosses and other plants evolved over the past 450 million years. Sanderson has hope that their progress will enable them draw the entire evolutionary tree of plants within the next few years. “It’s within striking distance,” he said. However, there is a bug in the program. “We have no way to visualize such a tree at the moment,” he said. If they tried, they would end up with a blurry, inscrutable thicket. “It would be ironic,” Sanderson said. “We’d be saying, ‘We’ve built it, but we can’t show it to you.’ ” In 1837, Charles Darwin drew a spindly sapling and since then biologists have relied on evolutionary trees to understand the history of life. Today biologists draw evolutionary trees to help them track the emergence of new diseases, identify species at risk of extinction, and trace the history of disease-related genes in the human genome. It is possible within a few decades biologists may figure out how the millions of species on Earth are related to one another. The bug is for people to actually see that tree of life; the tree itself will have to evolve. Biologists have turned to computer scientists and software designers and companies such as Google and Adobe to find a new way of looking at evolution. Why, to create a program that allows scientists and nonscientists alike to fly through evolutionary trees. “Just like Google Earth changed the way people look at geography, a sophisticated tree of life browser could really change the way we look at the life around us,” said Mark W. Westneat, the director of the Biodiversity Synthesis Center at the Field Museum in Chicago. When he was 28, Darwin drew the first evolutionary tree. He was back in England after his five-year voyage around the world aboard the Beagle, and his theory of evolution was still in an embryonic state. He felt evolution could explain the similarities and differences between species. The descendants of an ancestral species might have evolved into different forms, splitting into separate lineages “like the branching of a great tree from a single stem,” as he would later write in “On the Origin of Species.” Darwin’s first tree is well known and a familiar sight in books, museum exhibits and, of course, on Wikipedia. David Kohn, the director of the Darwin Digital Library at the American Museum of Natural History, has recently discovered 10 other trees that Darwin drew in later years. “It’s a long-term preoccupation,” Kohn said. “It feels like he’s using it to think.” While pondering how humans evolved, Darwin drew a cluster of branches to represent our common ancestry with apes and monkeys. However, during his life, Darwin published only a single tree. In the “Origin of Species” he included a set of branching lines, marked only by letters. “It was all at a very theoretical level,” said David Hillis, a biologist at the University of Texas. The work of figuring out what real evolutionary trees would look like was left for others, For example, in 1879, the German biologist Ernst Haeckel published a tree, complete with bark and leaves, showing humans and animals evolving from single-celled creatures. In the late 1900s, the science of tree-building took a significant step forward. Standard rules for comparing species and figuring out who was most closely related to who were set up Now that the biologists were all speaking the same scientific language, they were able to test each other’s hypotheses with new evidence. In addition, they could begin to get new kinds evidence for their trees. It was now possible to compare not just the skeleton or color patterns of species, but also their proteins and genes. In the beginning, the biologists could draw only small trees, typically with a dozen branches at most. What held them back was the fact that a group of species may possibly be related in many different ways. Add more species to a group and the possibilities explode. “For 25 species, there are more possible trees than there are stars in the known universe,” Westneat said. “For 80 species, there are more trees than there are atoms in the known universe.” It would be impossible to simply compare the trees, fortunately, mathematicians developed statistical methods for searching quickly through potential trees to find the ones that do the best job of explaining all the evidence. It was then possible for computers to do millions of calculations for biologists and store a growing database of information on Web sites. Trees began to grow hundreds of new branches, then thousands. “We’re overwhelmed with information,” Dr. Hillis said. Trees with thousands of branches are now sometimes called “supertrees” or “megatrees,” are starting to appear in scientific literature. The branches of these supertrees reveal patterns in evolution that were missed in smaller studies. An example, in 2007, Olaf Bininda-Edmonds, a biologist at Carl von Ossietzky University in Germany, and his colleagues published a tree of 4,500 mammals or just about every known mammal species. Researchers were, now, able to estimate the rate at which mammals have evolved into new lineages. Researchers have long argued most major groups of living mammals evolved after the dinosaurs became extinct 65 million years ago. Based on their mammal supertree, Bininda-Edmonds and his colleagues made the argument mammals were diversifying millions of years earlier. Now, the mammal supertree looks tiny. In a paper which will be published in the journal BMC Evolutionary Biology, Stephen Smith of the National Evolutionary Synthesis Center in North Carolina and his colleagues have created a tree containing 13,533 species of plants. They have show that ferns, which are sometimes considered as living fossils that have changed little for hundreds of millions of years, have actually been evolving faster than younger groups of plants, like conifers and flowering plants. Plants are not just related to more than each other; plants are related to us animals, fungi, bacteria and all other living things on Earth. The National Science Foundation is financing a project known as Assembling the Tree of Life. The goal of this project is “to reconstruct the evolutionary origins of all living things,” according to its Web site. Slices of the tree are being analyzed and mathematicians as well as computer scientists are working on methods to combine them into a single analysis. “You can just imagine how Darwin would have enjoyed it,” Kohn said. “Even when the mammal supertree is printed out at two meters by two meters, the species names remain virtually unreadable,” Bininda-Edmonds said. “It’s a Google Earth kind of problem. You can’t simultaneously see where Central Park is in New York, and where New York is in the United States.” Biologists are going to have to find new ways to draw evolutionary trees. “Our advances in understanding evolution are moving really fast now, but the tools for looking at these big trees are lagging behind,” Westneat said. A wall in Dr. Sanderson’s laboratory in Tucson may display the future of evolutionary trees. Sanderson and his colleagues have mounted a bank of flat-screen monitors that can show off a program they have designed called Paleoverde. Sanderson can transform an evolutionary tree into a three-dimensional structure, and then use his mouse to navigate through it, zooming in on particular branches he wants to inspect. “My program can handle 1,000 species fairly effectively. When you get to 5,000 species, it gets very slow and not very beautiful,” he said. Westneat has been organizing meetings over the past year to bring the computer scientists and biologists together. “It has the potential to move us beyond what biologists with a little bit of programming can do,” Westneat said. “It’s definitely not small potatoes — it’s cutting-edge research,” said Tamara Munzner, a computer scientist at the University of British Columbia. Munzner is working on methods to allow biologists to see details of the tree of life without losing sight of its overall shape. One of her programs acts like a fisheye lens, blowing up clusters of branches and she has also figured out how to make trees rubbery, so that a biologist can stretch some parts of it open and squeeze others down. Munzner’s programs can handle millions of branches. Dr. Hillis thinks drawing the tree of life will become a practical tool, in the same way online databases of DNA have become practical tools for geneticists. “What I’d really like is the entire tree of life on a small hand-held device,” Hillis said. Biologists could put a tissue sample from a plant, animal or other organism in the machine, which would then scan its DNA and find its place in the tree of life, even if it’s a new species. The data could then be uploaded to a database, so that every biologist’s machine would get an updated tree. “It would be a ‘tricorder’-like device, able to identify any species on Earth in the field,” he said. The tree of life, if biologists do succeed, will look very different from Darwin’s sketch. Lineages do branch as they evolve, but sometimes the branches join back together. Separate plant species sometimes produce hybrids that can no longer interbreed with their parent species; in other words, they become new species. When biologists draw the relationships of some groups of plant species, their pictures look more like webs than trees. In other cases, genes simply leap from one branch of life to another. Viruses can infect a new host species, and in the process they transfer genes from its previous host. Many species of bacteria can slurp up naked DNA or pass it to one another on tiny genetic ringlets. “Each gene has its own evolution. It’s not inherited from mother to daughter; it’s inherited from a neighbor,” said Peer Bork of the European Molecular Biology Laboratory. Biologists have just begun to understand how this different kind of heredity alters the tree of life. Although genes may move from one species to another fairly often, it may be rare that they become a permanent part of a new genome. Tal Dagan is a biologist at the University of Düsseldorf and has estimated their impact by analyzing hundreds of thousands of genes from microbes. She estimates that 80 percent of the genes in any microbe have been passed from one species to another at some point. “We had to have a new picture of evolution,” Dagan said. She still sees evolution shaped like a tree. “Most of the evolution is still going on in the branches,” she said. However, thousands of genes, over billions of years, have shuttled among the branches. To emphasize this, Dagan and her colleagues have drawn a dense filigree of lines between the branches of the tree of life. “You see the tree and you see the thousands of edges, and you know this is how it is,” she said. Will viewers become lost in the complexity? “It does make things more complicated,” Munzner said. “But it doesn’t mean it’s hopeless. My answer is, ‘Bring it on.’ ”