A better family tree would help researchers better place turacos, such as this Knysna turaco from South Africa, on its branches.

Daniel J. Field

Bigger, better bird tree of life will soon fly into view

Bird lovers—and evolutionary biologists—can look forward to a new and improved avian family tree being finished in 4 years, thanks to The OpenWings Project officially launched this last week at the American Ornithological Meeting in Tucson, Arizona. The $1.42 million effort will be the first to include DNA data from the more than 10,500 known bird species to establish how they are all related. But it will not be the final word—another project seeking to sequence the full genomes of every avian species will follow if enough funding can be raised. OpenWings “will be a huge improvement over what we have now,” says Harvard University evolutionary biologist Scott Edwards. But, “Ultimately, OpenWings will be a stepping stone to the grand tree that the whole genomes [will generate].”  

In 2014, biologists published an avian tree based on the sequences of whole genomes of about 40 species. Another team published a different tree in 2015 after comparing a subset of the avian genome in hundreds of species. These phylogenies help researchers looking at the evolutionary histories of specific avian traits or the story of birds overall. But some researchers who specialize in building trees were not satisfied. “The current need for large phylogenies and the high priority placed on them by high impact journals can result in shortcuts, wherein large-scale phylogenetic trees are cobbled together from disparate existing sources, even taxonomy, but often without hard data behind the placement of many species,” Harvard evolutionary biologist Gustavo Bravo and his colleagues wrote on 30 January in PeerJ. “The question is how far do you compromise?” Edwards adds.

So even though some of the leaders of the 2014 avian tree effort launched the Bird 10,000 Genomes (B10K) Project, aiming to eventually sequence the whole genomes of all 10,560 bird species and from there build “the grand tree,” some bird researchers decided not to wait. Led by Brian Smith at the American Museum of Natural History in New York City and Brant Faircloth at Louisiana State University in Baton Rouge, they are taking a cheaper, faster route with the OpenWings Project. The U.S. National Science Foundation–funded effort will tap extensive museum collections as much as possible, instead of freshly caught bird samples, and will sequence about 5000 short pieces of the DNA, focusing on regions that are very highly conserved among all birds. The group plans to release data on an ongoing basis, rather than waiting for the project’s publication, so other researchers can make use of it.

The OpenWings tree will give “us a better understanding of the process of [bird] diversification and can potentially give us the information we need to move forward in understanding vertebrate diversification,” Faircloth says.

But how good a tree will it be? “I don’t think [OpenWings] is going to solve well enough their goal of generating an accurate tree, says B10K leader Erich Jarvis, a neurogenomicist at The Rockefeller University in New York City. In the 2014 work, he and his colleagues compared trees built from just the conserved DNA regions OpenWings will also examine to those arising from whole-genome comparisons and found inaccuracies in the former. He argues that whole genomes will not only provide more data for building trees, but will also enable researchers to study the evolution of regulatory sequences, transposable elements, and other aspects of the genome that are not covered with OpenWings sequencing. Once a genome is finished, no more sequencing of that species’s DNA will be ever needed, he argues.

Jarvis worries that OpenWings will dampen enthusiasm and slow support for B10K, which could cost up to $150 million for high quality, so-called reference genomes, and still isn’t fully funded. Furthermore, Jarvis is concerned about using up rare samples. Some birds, such as species endemic to Borneo, are very hard to find and collect, and there’s a chance the two projects may wind up competing for DNA from the few samples gathered. Edwards does not think that’s a big problem, however, and says B10K needs better DNA than can be retrieved from the museums specimens that OpenWings plans to use. Faircloth hopes his team will not use any of the samples used in the B10K project, as having DNA sequence from multiple individuals of the same species can be useful. 

With help from the Chinese sequencing giant BGI, B10K has already sequenced to some degree the genomes of more than 300 birds, with representatives for each family and subfamily. But additional support is required to go much further, Jarvis says. “I don’t want people to get the impression that once OpenWings is finished that we are finished [with bird DNA studies],” he adds.