The genomes of more than 600 plants, animals and fungi that went extinct tens or even hundreds of millions of years ago, leaving no physical trace on Earth, have been partially reconstructed by rewinding the evolutionary history of their living descendants, in the largest ever study of its kind. This has given us the best pictures so far of the genomes of various ancient human ancestors, from the 45-million-year-old ancestor of monkeys and apes all the way back to the ancestor of the first amphibians and of all vertebrates.
This approach will never allow us to completely reconstruct the genomes of extinct plants and animals, or bring them back to life, but it can help reveal the evolutionary history of life on Earth. It might one day even give us a better idea of what ancestral plants and animals were like physically.
“Going from the genome to what the animal looked like, that’s something we can hope for,” says at the École Normale Supérieure in Paris.
Advertisement
Almost all that we know about past life on Earth comes from fossils. As important as they are, most fossil species aren’t the direct ancestors of the organisms alive today. For instance, almost all dinosaurs died out, with just one branch surviving in the form of birds.
“It’s not easy to connect a fossil to modern species, because you don’t know where it is in the tree of life,” says Roest Crollius.
Unlike fossils, however, the genomes of living creatures give us a direct link to their ancestors, he says. Genomes are a record of evolutionary history going back to the dawn of life. And by comparing the genomes of two or more related species, it is possible to effectively rewind those historical changes to work out what the genome of a shared ancestor was like.
Biologists have been doing this ever since the genomes of different species started to become available. In 2017, for instance, at the University of London and his colleagues used a computer algorithm to partially reconstruct , including the ancestor of a group called Boreoeutheria that contains most placental mammals, and which lived around 95 million years ago.
Since then, the genomes of thousands more species have been sequenced. This means ancestral genomes can be reconstructed more accurately, but it also poses a challenge in terms of the amount of processing power required. So Roest Crollius and his colleagues have developed a streamlined algorithm that works out which genes an ancestor had and the order in which they were arranged along chromosomes.
Because more genomes are available and the algorithm is so fast, Roest Crollius’s team has been able to reconstruct hundreds of ancestral genomes in this way. They include the 45-million-year-old ancestor of monkeys and apes including us, the 50-million-year-old ancestor of grasses and the 102-million-year-old ancestor of placental mammals including elephants. The team even reconstructed the ancestor of all vertebrates and of all tetrapods (amphibians and their descendants), but these very ancient genomes are very incomplete and fragmentary.
Read more: A brief history of the human genome
The researchers say their reconstructions closely match previous ones where they exist, but are more complete. “Our approach does reconstructions more accurately and with greater resolution, and also it goes further back in time,” says Roest Crollius.
For instance, they were able to pinpoint the locations of an additional 2000 genes in the Boreoeutheria genome compared with the one reconstructed by Larkin’s team.
“It’s very fast,” says Larkin, of the new algorithm. “I think I will use it.”
Many of the other ancestral genomes have never been reconstructed before. This includes Strepsirrhini, the ancestor of lemurs, bushbabies and lorises, and Chiroptera, the ancestor of bats.
The team found that the rate at which chromosomes change varies a lot. For instance, in the lineage leading to gibbons, there have been 60 major rearrangements of chromosomes in the past 25 million years, far more than in other lineages.
With more genomes and more processing power, it may become possible to work out the genetic code of ancestral genes as well as their order. But reconstructing the sequence of the non-coding DNA between genes isn’t feasible, says Roest Crollius, because it changes too fast.
“Nobody will be able to go back very far in time reconstructing entire genomes,” he says.
In most species, most of this non-coding DNA is junk, consisting of thousands of mutated copies of genetic parasites called transposons. But some non-coding DNA helps regulate gene activities and plays a key role in the evolution – and without this regulatory DNA, it will be impossible to resurrect extinct species, says Roast Crollius.
Still, even if we can’t bring these species back to life, we might yet learn more about what they were like. “You have access to the molecular pathways that were present, the developmental pathways that were present and so you can start to make inferences about which cellular processes or developmental processes were present,” says Roest Crollius. “We will have enough genomes soon, given all the different projects that are ongoing, we should soon have a lot of data to start making those connections between genes and functions and morphology.”
How ancestral genomes could help bring back extinct species
It is impossible to reconstruct ancestral genomes accurately enough to bring those organisms back to life (see main story). But such genomes could still help efforts to resurrect species such as the woolly mammoth.
Several groups hope to create animals that resemble extinct species by sequencing the DNA in preserved samples, then editing the genome of a close living relative to make it a close match for the extinct species.
The problem with this approach, says at the University of Copenhagen in Denmark, is that ancient DNA breaks into lots of little pieces. The only way to reassemble them is to use the genome of a living relative as a guide. But both the extinct and living species will have changed since splitting from their common ancestor, so the reassembled ancient genome will have some key parts missing.
In a 2020 study, Gilbert and his colleagues showed that reconstructing the genome of a common ancestor and using that as a guide can .
“If you halve the evolutionary divergence by reconstructing the ancestor, it can improve the amount of data you can reconstruct from your crappy ancient DNA, as you are mapping to a reference that is closer,” says Gilbert.
Reference:
Sign up to Wild Wild Life, a free monthly newsletter celebrating the diversity and science of animals, plants and Earth’s other weird and wonderful inhabitants
