Flock of avian genomes resets our view of bird evolution

A recent issue of Science magazine published a suite of amazing papers based upon full genomes sequenced from 48 bird species (and a few close relatives.)  They not only provide an amazingly complete and well-resolved bird tree of life, but they were used to answer pressing questions about brain evolution, bird's loss of teeth, and the evolution of song learning. As all of these genome sequences went public, eight papers were published in Science Magazine alone, and another 20 appeared in other journals.  This work is changing the way that we view birds.  And learning to deal with this much data is changing the way that we do science...

Why does the order of birds in your favorite field guide change from edition to edition?  Largely because ornithologists have been uncertain about the relationships among the major avian orders.  Only recently have genetic data clearly shown that ducks and chickens (Galloanserae) make up a separate lineage from the rest of birds (Neoaves).  But the biggest challenge in recovering the bird tree of life was that most bird orders seem to have appeared in a burst of evolution shortly after the other dinosaurs went extinct. It is almost as if they radiated into diverse forms to fill the void left by the departing dinosaurs.

So bird researchers have known that it would take a lot of data - no, I mean A LOT of data - to resolve the relationships among bird orders and accurately date the origins of each order.  But as several bird research groups were sequencing genomes of their favorite species to study things like neurological development, song learning,

After several years of work assembling complete genomes from all of the major bird orders, researchers from around the world have published a suite of paper summarizing their analyses of bird evolution. 

Some of the cool insights:
First let’s look at the phylogeny, which is a picture of the tree of life.

---

Figure 1 from Science Magazine.  Jarvis et al. 2014. Whole genome analyses resolve early branches in the tree of life of modern birds. Science 346(6215): 1320-1331.

This particular tree shows the branches as they occur, with time going from left (older) to right (younger), with a time-scale bar on the bottom of the tree. One obvious finding is that the major radiation of Neoaves bird orders occurred around the K-Pg boundary, 66 million years ago, around the time when the dinosaurs went extinct.  Presumably the large void left when the dinosaurs and other animals went extinct made room for the surviving birds to diversify into the many land, sea, and air birds that exist today.  The diversification happened rapidly, with most modern bird orders forming and diversifying by 50 million years ago. 

In the full genome tree, many of the traditional groups are retrieved in this phylogeny, but some aren’t. The falcons are split from the other hawks and eagles, and the falcons are related to parrots and songbirds.  The tree also highlights many higher groupings of bird orders, including

• Caprimulgimorphae that includes the nightjars, swifts, and hummingbirds,
• Otidimorphae that includes the cuckoos, turacos, and bustards,
• the “core waterbirds” (aka Aequornithia) that comprise pelicans, herons, ibises, cormorants, tube-nosers, penguins, and loons,
• the “core landbirds” (aka Telluraves) that include two major groups of landbirds whose early evolution was probably restricted geographically,

• the Afroaves that includes hawks, eagles, owls, mousebirds, bee-eaters, woodpeckers and other non-passerine groups, and
• the Australoaves that includes the Passeriformes and their nearest neighbor the parrots, falcons, and seriemas. 

But there are still a couple mysteries remaining.  For example, the owls are not well placed on this tree, and it is not clear whether they are sister to the hawks and eagles or whether they are sister to the Coraciimorphae that include the other Afroaves.  Although they are weakly placed as sister to the Coraciimorphae, the Barn Owl genome that they used was one of the poorest assembled genomes and the finding is weak and may change with additional analyses.

And finally, the order Passeriformes – which include nearly 50% of all living birds – started diversifying only about 40 million years ago, but has been one of the most rapidly evolving groups of birds.

And there are several other genome findings, summarized in a second paper.
Birds have smaller genomes than other non-fish vertebrates – and small genome size seems to be correlated with flight in general.  For example, bats have smaller genomes than their terrestrial relatives, and typical birds have smaller genomes than their cousins who have lost the ability to fly (like ostrich.)  Although this has been recognized and has sparked some interesting ideas, the genome data is suggesting that birds reduced their genomes by cutting out a lot of the junk – like transposable elements (aka jumping genes) and other repeat elements.  Looking across the avian tree of life, it looks like most of the genome reduction happened very early in bird evolution.  In fact, birds genomes are fragmented into many “microchromosomes”, and many of the missing pieces of DNA appear to be cut from the microchromosome break points.  In other words, the smaller chromosomes appear to be cut up in ways that eliminated the junk between important genes.

Most birds have a relatively similar arrangement of genes on their chromosomes, and across the avian tree of life, chromosomal rearrangements are relatively infrequent.  It is unknown whether this is by design or simply by constraint.  Most chromosome breaks and rearrangements are believed to occur at repeat regions – which are reduced in birds.  It may be that by reducing the overall genome size, they have reduced their ability to make huge chromosomal rearrangements.  This also means that there is a relatively slow rate of gene gain and gene loss in birds.

Having full genomes across the avian tree of life allows researchers to more closely examine the role many genes throughout evolutionary history.  For example, some traits, such as vocal learning evolved multiple times in birds, and the genes believed to be involved in vocal learning can be carefully studied to see how they change with respect to selection pressure.  Likewise, the bird visual system, the oxygen carrying proteins in red blood, feather production can all be studied for evidence of natural selection and changes that control each of these systems. Virtually any scientist studying any bird trait can examine the evolution of genes involved in her favorite trait across the entire bird tree of life. 

These 48 genomes are just the beginning, and many more genomes are released each month.  With this incredible amount of data, we can expect our knowledge of bird evolution to continue to take huge leaps forward.

And maybe we can hope that the order of birds in our favorite field guides will become more standard and stable.