A new genetic map of nearly all living birds has shaken the avian evolutionary tree, scattering some species to new perches and adding odd flourishes to their tales.
Parrots roost closer to songbirds, a tiny wren shares a murderous lineage with the eagle, and the pencil-legged flamingo nests near the deep-diving grebe, according to a series of studies published Thursday in the journal Science.
The four-year, multimillion-dollar effort to map all the DNA of 48 bird species provides an enormous database for research that affects human health. Chickens, finches and pigeons are commonly used in laboratories, offering insight into brain and fetal development, toxin risks and speech impediments.
With more than 10,000 species, birds are the most diverse class of four-limbed vertebrates on Earth. Their lives are closely intertwined with those of humans, who have domesticated them for food, trained them as hunters, caged them for companionship, used them to detect deadly gas in coal mines and simply marveled at them in the wild. In 2011, an estimated 47 million Americans shelled out nearly $41 billion for transportation, lodging, equipment and other costs associated with birding, according to the U.S. Fish and Wildlife Service.
Plus, "birds are also the living descendants of dinosaurs," said Tom Gilbert, head of the evolutionary genomics section of the Natural History Museum of Denmark and an author of several of the studies. Yet despite centuries of close coexistence, he said, "it's been very, very hard for people to work out the simple relationship between different orders of birds."
By sequencing 45 new genomes and merging them with three others previously published, the Avian Phylogenomics Consortium reordered the bird tree of life — and opened multiple paths toward a better understanding of humans. For instance, one research group discovered remarkable analogies between genes involved in vocal learning in birds and the speech areas of the human brain.
"There are lots of disorders that affect speech, and we can't study nonhuman primates or mice for these," said Erich Jarvis, a neurobiologist at Duke University School of Medicine who helped lead the project.
The overall effort by more than 200 scientists at 80 institutions worldwide would have taken about 400 years for one computer to accomplish. Twenty-eight research papers were published Thursday in a variety of journals, and additional findings will emerge in the coming months.
For ornithologists and backyard birders, the studies have the potential to resolve — or extend — long-standing debates over the feathered dinosaur descendants.
It now appears that birds experienced several simultaneous "big bangs" that led to a wide array of species shortly after land-based dinosaurs became extinct some 66 million years ago.
Since then, birds gained the ability to learn sounds at least twice — once in the line leading to hummingbirds and again in the ancestor of parrots and songbirds. And at least twice, branches lost their ability for vocal learning — among the New Zealand wrens and a group that includes antbirds and manakins, according to the studies.
Other birds, such as the emu and the ostrich, jettisoned flight, and almost all lost their right ovary.
A big hunting bird that lived about 65 million years ago probably gave rise to all the major land birds, from the diminutive finch to the massive, flightless "terror birds" that became extinct about 2 million years ago, according to the studies.
"We are suggesting that the common ancestor of core land birds was an apex predator, and that this predatory trait has been lost several times," Jarvis said.
The research also appears to solve the long-standing mystery of why hens don't have teeth: The common ancestor of all birds lost the code for growing teeth about 116 million years ago, leaving its descendants to evolve horny beaks and a digestive tract that grinds food.
Physical characteristics were a primary means for sorting bird species until the advent of genetic sequencing. But many of these modern studies focused on small DNA segments, and they sprouted conflicting trees that sent species flitting from branch to branch.
That problem, known as incomplete lineage sorting, is particularly acute when so many species develop in a short time, said Tandy Warnow, the mathematician and computer scientist at the University of Illinois at Urbana-Champaign who headed the project's colossal data-crunching effort. Warnow's team resolved this in part by merging about 14,000 individual genetic trees.
The overall bird genome came out to be one-third the size of most mammals, with about 1.05 billion to 1.26 billion base pairs, said Guojie Zhang, head of comparative genomics at the National Genebank of the Beijing Genomics Institute-Shenzhen, another consortium leader.
"The bird genome has experienced a massive gene loss," Zhang said. "We found that 1,600 genes are completely missing from all bird genomes." Some of those are considered crucial to other animals, including genes regulating flexible lungs and those associated with sexual organs, he said.
Despite several parallel "big bang" speciations, birds have proved to be relatively slow evolvers, the studies found. Their rate of DNA shuffling varies among species, but it is lethargic compared with that of mammals.
Among those studied, however, no animal matches the molecular malingering of the crocodilians, which started slow and just got slower. Their genome was included in part because they are the closest living relative to birds, said University of Florida evolutionary biologist Edward Braun, a coauthor on several of the studies.
Which bird, then, is closest to the last common ancestor shared with saltwater crocodiles, American alligators and the slender-nosed Indian gharial? Either the chicken or the ostrich, depending on how you look at it. If a genome is a shelf of books, then the chicken has been a better librarian, but the ostrich has been a more faithful translator.
"It's actually the chicken that seems to have retained the most ancestral chromosomal organization compared to the other species," Jarvis said. But at the single base pair level, "the ostrich is evolving more slowly than the chicken. These are two different aspects of gene evolution."