An international effort to sequence the genomes of 45 avian species has yielded the most reliable tree of life for birds to date. This new avian family tree helps to clarify how modern birds--the most species-rich class of four-limbed vertebrates on the planet--emerged rapidly from a mass extinction event that wiped out all of the dinosaurs approximately 66 million years ago.

It reveals how some of the earliest bird species diverged, answering many long-standing questions about the common ancestor of birds, crocodilians, and dinosaurs--a group collectively known as archosaurs--and shedding new light on the evolution of avian sex chromosomes, vocal learning in both birds and humans, and the process that led to birds losing their teeth.

The project strengthens the theory of a "big bang" for bird evolution during the 10 to 15 million years that followed the dinosaurs' extinction at the Cretaceous-Paleogene Boundary. It also suggests that the earliest common ancestor of land birds, which include parrots and songbirds as well as hawks and eagles, was an apex predator.

This massive comparative genomics project, which needed several supercomputers to process all its data, took more than four years and involved hundreds of scientists from about 80 institutions in 20 different countries. The research was led by Guojie Zhang from BGI in Shenzhen, China, and the University of Copenhagen in Denmark; Erich Jarvis from the Howard Hughes Medical Institute and Duke University in Durham, North Carolina; and Thomas Gilbert from the Natural History Museum of Denmark and Curtin University in Australia.

The researchers produced about 28 studies--eight of which appear in the 12 December issue of Science. The others are published in journals such as Genome Biology and GigaScience.

"This study, which examines at least one genome from every major modern bird lineage, culminates in two main flagship papers," explained Laura Zahn, a senior editor at the journal Science. "They examine the genetic underpinnings of bird biology and evolution, solving the contentious relationships between the major groups of living birds."

In one of the flagships, Zhang and Cai Li from BGI and the Natural History Museum of Denmark, along with colleagues, describe their comparative analysis of 48 avian genomes, including the 45 new sequences they contributed (crow, duck, pigeon, falcon, woodpecker, eagle, ostrich, and many more) along with three genomes that were already available (chicken, turkey, and zebra finch).

Their findings help to explain why bird genomes, in general, are about 70% smaller than those of mammals. They also highlight specific regions of the birds' genomes that have been conserved for more than 100 million years--and the convergent evolution of certain avian traits along the way.

"In the past, people have been using one, two--up to 10 or 20 genes--to try to infer [bird] species relationships over the last 100 million years or so," said Jarvis. "Our theory has been: If you take the whole genome, you would have a more accurate species tree than just one or two genes [could provide] alone."

And it turns out that theory is supported: A report by Jarvis and Siavash Mirarab from the University of Illinois Urbana-Champaign in Champaign, Illinois, et al.--the other flagship paper--shows that protein-coding genes are not enough to get an accurate phylogenetic tree. Researchers must include non-coding sequences of DNA as well as regions between the genes to provide a more accurate picture, they say.

Next, a report by Qi Zhou from the University of California in Berkeley, California, and colleagues highlights the evolution of sex chromosomes in birds and reveals that, unlike the human Y chromosome, the avian W chromosome still has many active genes--and sex chromosomes of various bird species are currently at different stages of evolution.

Andreas Pfenning from Duke University with support from the Howard Hughes Medical Institute, along with his colleagues, reports shared molecular specializations between brain circuits that are important for singing in vocal-learning birds and speech in humans, while Osceola Whitney, also from Duke University, and his colleagues determine that 10% of a bird's genome is regulated by singing, with highly diverse patterns across singing brain regions mediated by epigenetic differences.

Another report by Ed Green from the University of California in Santa Cruz, California, and colleagues describes the sequencing of three crocodilian genomes--the American alligator's, the saltwater crocodile's, and the Indian gharial's--which represent birds' closest living relatives. The genomes of such crocodilians are evolving at an exceptionally slow pace, according to the researchers.

"The molecular evolution of birds is much faster than it is in crocs, turtles and other reptilian lineages," said Green. "So this avian lineage seems to be faster than other reptiles, but not faster than mammals."

A separate report by Robert Meredith from Montclair State University in Montclair, New Jersey, and colleagues suggests that the mutations that eliminated enamel and dentin from the teeth of modern birds--an event which eventually led to toothless beaks--began about 116 million years ago.

And a report by Mirarab and Tandy Warnow, also from the University of Illinois Urbana-Champaign, along with their colleagues, describes how their team of researchers was able to produce the most accurate phylogenetic trees from gene trees with a technique they call "statistical binning."

The group's findings are being reported nearly simultaneously in 23 papers - eight in a Dec. 12 special issue of Science and 15 more in Genome Biology, GigaScience and other journals.