Australasian Science: Australia's authority on science since 1938

The Big Bang of Bird Evolution

The evolutionary tree of modern birds

The evolutionary tree of modern birds estimated from genomic data. Art by Jon Fjeldså.

By Simon Ho

Genome studies have revealed whether the extinction of dinosaurs coincided with the rapid diversification of birds.

Studying the fossil record has allowed us to work out the timing of key events in evolution, such as when animals originated, when plants colonised land, and when we shared ancestors with our primate cousins. By dating fossils using geological methods, we can gain an estimate of the timescale of life.

But what if we are interested in the history of organisms whose fossil record is incomplete? If these organisms have left modern descendants, we can use molecular clocks to estimate their evolutionary timescales.

Molecular clocks are statistical models that describe evolutionary processes at the genomic level. They allow us to fill in some of the details of life’s evolutionary history that we cannot extract from the patchy fossil record.

I recently used molecular clocks as part of the largest-ever genetic study of bird evolution. Not only did this international study confirm that the major groups of modern birds diversified immediately after the demise of the dinosaurs, we also found that several bird lineages have been around for a lot longer and that the most diverse group, the passerines or perching birds, only rose to dominance in the past 40 million years.

To use molecular clocks, we need to collect information from the genomes of modern organisms. Genomes are essentially long strings of DNA made up of four building blocks called nucleotides. The human genome consists of about three billion nucleotides, but genomes vary dramatically in size across species.

As parents pass their genomes to their children, errors are made during the DNA copying process. These DNA mutations accumulate in the genome of a species over time, forming the basis of evolution at the molecular level.

Molecular evolution is influenced by a combination of natural selection and chance. In this way, each organism’s genome contains signals of its evolutionary past.

The simplest type of molecular clock was developed in the early 1960s. Although it was first applied to protein analysis, the molecular clock is now regularly used to study DNA sequences.

The model states that DNA mutates at a constant rate over time. If we determine the nucleotide sequences of the genomes of two organisms, the amount of difference in their DNA reflects the time since they last shared a common ancestor.

For example, consider our close primate relatives. Our genomes differ from those of chimpanzees by about 1.4%, whereas human and gorilla genomes are 1.8% different. These numbers actually depend on the part of the genome that is being examined, but they suggest that our evolutionary split from gorillas occurred about 30% earlier than our split from chimpanzees.

On its own, the measure of genetic difference is not enough to work out the evolutionary timescale. This would be like looking at a car’s odometer, seeing a reading of 100,000 km and trying to determine the total amount of time spent driving. To do this we would need to know the average speed. Working out the speed of the molecular clock is known as calibration.

Using fossils is the most popular method of calibrating molecular clocks. In the example above, fossil evidence indicates that humans and chimpanzees probably shared an ancestor at least 6 million years ago. Assuming that this date is correct, we can estimate that we split from gorillas 8 million years ago.

This example shows that while using molecular clocks does not completely remove our reliance on fossils, we can still estimate the timescales of poorly preserved organisms by using fossil evidence from those that are well preserved.

In practice, using molecular clocks is not so simple. A key problem is that DNA evolves at different rates among organisms. For example, flies evolve more rapidly than humans do. This is mainly because flies go through many generations in the space of a single human lifetime, meaning that their genomes are copied much more frequently.

Most DNA mutations are caused by errors during the copying process. This leads to a pattern known as the generation-time effect, whereby organisms with shorter generations tend to evolve more rapidly.

Another problem is that the fossil record rarely provides enough detail to allow us to pinpoint precise dates that we can use for calibrating molecular clocks. There is usually a large degree of uncertainty in fossil-based estimates of the timing of evolutionary events. Even for the familiar split between the ancestors of humans and chimpanzees, various palaeontologists have placed the date anywhere between 6–10 million years ago.

There are ongoing efforts to deal with these complex aspects of molecular clocks. Variation in rates of DNA mutation can be modelled using “relaxed” molecular clocks. Uncertainty in fossil-based calibrations can be taken into account using mathematical methods. These various developments over the past two decades have led to great improvements in the reliability of molecular clocks.

Molecular clocks are now used to study a variety of evolutionary questions across a wide span of time. These range from the evolutionary dynamics of viruses over the course of a few weeks, to the early evolution of microbial life billions of years ago. For example, molecular clocks have played an important role in revealing the evolutionary timescales of pathogens, including the emergence and spread of different viruses.

Some of the most fascinating questions in evolution have concerned the rapid appearances of major groups of organisms, such as birds, mammals and flowering plants. In all of these examples, the modern groups that we recognise today all appeared within a very short space of time (in evolutionary terms). Working out when these remarkable diversification events occurred can help us to understand what triggered them.

The Diversification of Modern Birds

Birds are a diverse, colourful, and captivating part of the modern fauna. They are found on all seven continents and occupy a broad range of ecological roles. For these reasons, the evolution of modern birds has been a constant source of interest – and debate – among biologists.

The traditional view has been that the major orders of modern birds only arose after the extinction of the dinosaurs, which were wiped out by an asteroid about 66 million years ago. When these reptiles met their demise, birds and mammals had the chance to occupy all of the ecological roles that had previously been filled by dinosaurs. This view is supported by the fossil record, with the modern orders of birds and mammals only appearing within the last 66 million years.

Over the past two decades, the traditional view was challenged by a number of studies based on molecular clocks. These studies found that the modern orders of birds diversified long before the dinosaurs became extinct.

Some of these studies even claimed that the diversification of modern birds took place 40 million years before their first fossils appeared – suggesting that a large chunk of the evolutionary history of birds is missing from the fossil record. However, these genetic studies were only based on small amounts of DNA data, representing a tiny fraction of the whole genome.

As part of a large team of Australian and international researchers, I was recently able to use whole genomes to revisit the genetic estimates of the evolutionary timescale of birds. We sequenced the genomes of 48 different species that represented the diversity of modern birds. About 10 of these species are found in Australia, including the zebra finch and budgerigar.

The 48 genomes formed the largest data set ever assembled to study the evolution of birds. The analyses of these genomes were carried out on nine different supercomputers, a mammoth task that would otherwise have taken more than 400 years on a fast desktop computer.

Using molecular clocks, we estimated that the ancestors of ratites went their own evolutionary way from other birds about 100 million years ago. The ratites are a group of mostly large, flightless birds, and include the ostrich, kiwi and rhea. In Australia, the ratites are represented by the emu and cassowary.

But the most important burst of diversification occurred when the dinosaurs became extinct. Within the space of less than 10 million years, almost all of the modern orders of birds appeared. They were all in existence by 40 million years ago. Some of the last orders to appear were the swifts, hummingbirds, bee-eaters and woodpeckers.

The evolution of the largest order of birds, the passerines, is particularly interesting. This group represents 60% of all living bird species, including iconic Australian species such as the magpie and lyrebird. The passerines split from their closest relatives, the parrots, about 55 million years ago. But the passerines reached their full diversity within only the last 40 million years. Today, the order is represented by about 6000 species.

The “Big Bang” of bird evolution took place in parallel with that of placental mammals. A recent analysis of mammalian genomes revealed that there was a massive burst of evolutionary diversification following the extinction of the dinosaurs.

The combined evidence supports the view that the mass extinction at the end of the Cretaceous opened up a wide range of ecological roles that were exploited by new types of birds and mammals.

Sequencing More Genomes

The molecular clock has proven to be a valuable tool in biology. It has vastly improved our understanding of the evolution of life, not just in the deep past but also in recent history. Behind the apparent simplicity of the molecular clock, however, lies a range of complex evolutionary processes. These are being dealt with by the combined work of biologists, mathematicians and computer scientists.

The greatest challenge to the molecular clock is the need to deal with the enormous genomic data sets that are now available. The recent study of 48 bird genomes was a huge undertaking, but plans to sequence the genomes of all 10,000 bird species are already underway. We will need new methods to handle such large amounts of data.

Efforts in this area will be rewarding because genome data will help us to understand the evolutionary processes that lie behind the face of the molecular clock.

Simon Ho is an Australian Research Council Queen Elizabeth II Research Fellow in the School of Biological Sciences at the University of Sydney.