Original questions: “Is Jurassic Park feasible?” asked by Natt, and “If we had enough of its DNA, could we bring back an extinct species?” asked by Louise.

Jurassic Park suggests that we could extract dinosaur DNA from mosquitoes encased in amber, and use that DNA to resurrect the dinosaurs. This wouldn’t work, for reasons I will go into; but first, let us consider how we might bring back an animal from extinction if we had its DNA.

Bringing Animals Back from the Dead

This is just one potential strategy, coming from my perspective as a biologist. There might be other ways to approach this, particularly if you had intact cells, but I am assuming that we have somehow got hold of some mixed DNA from a preserved animal. There are a number of steps we would have to go through:

1) Obtain the DNA 2) Sequence the DNA and reconstruct the genome of the animal 3) Synthesise a copy of the DNA 4) Replace the DNA of an embryo with the synthetic DNA 5) Bring the embryo to full term

Let’s look at each step and consider current technology, and what is possible.

1) Obtain the DNA

Under the Premise of the question, we have already solved this problem somehow. If you have sufficient animal tissue, DNA extraction is quite easy even with household chemicals.

2) Sequence the DNA and reconstruct the genome of the animal

This step we can undoubtedly do today. DNA Sequencing is the process by which you can work out the order of the DNA ‘bases’ along the strands. Sequencing is a very advanced and reliable technology, and if we had enough DNA of any animal we could sequence its genome. The only restriction would be the cost, and if it’s mixed in with lots of other DNA, like from bacteria, it might be difficult, but doable. One of the most popular methods is that used by Illumina machines, but there are a number of approaches that could work.

3) Synthesise a copy of the DNA

This is harder. We can reliably synthesise fairly substantial pieces of DNA, but current technology cannot simply ‘print out’ a chromosome. We are at the point where we can quite rapidly synthesise a bacterial genome, but chromosomes are larger and far more complex. This is something we will likely be able to do in future, but the technology isn’t quite there yet. Also, this is a slight oversimplification, as chromosomes need a lot of proteins associated with them in order for them to function. We may have to synthesise these proteins and coerce them into forming proper chromosome complexes or even a complete cell nucleus before continuing.

4) Replace the DNA of an embryo with the synthetic DNA

This is probably quite doable. Assuming the DNA isn’t wholly incompatible, we have the technology to put the nucleus of one animal into the embryo of another. We can perform this with Human nuclei in mouse cells, for example.

5) Bring the embryo to full term

This is potentially quite difficult. You might to find an animal that can support the embryo through its normal development, by either replacing the embryo of an egg with your hybrid embryo, or by performing in-vitro fertilisation of a mammal. This may require modification of the host mother in order to suppress tissue rejection mechanisms. This is a challenge that could only be overcome by experimentation.

Alternatively if ‘ex-vitro’ technology advances to the point where we can raise an embryo entirely in the lab, you could in effect create ‘test tube animals’ with no requirement for a host womb or egg. We do not currently have this technology, though progress is being made into supporting undeveloped embryos outside of their normal environments.

So in conclusion, we could probably bring back an extinct species if we had enough DNA, but we do not have all the technology to do so just yet.

Why this couldn’t work for Dinosaurs

Attempting to do this with dinosaurs will fail at the first hurdle. We simply cannot have the DNA to begin with. DNA degrades over time, even if left in a glass jar all alone. The bonds between the bases of DNA have a small chance of breaking at any moment. The rate of this decay is such that half of the bonds in a piece of DNA will break over a period of 521 years. DNA sequencing requires us to break apart the DNA into small fragments for analysis, but beyond a certain point the fragments will be too small to convey any useful information. This ‘half life’ of DNA means we cannot possibly sequence the genome of any animal sample older than about 1.5 million years, which is a lot more recent than the 66 million years since the extinction of the dinosaurs. All we really have are the mineralised impressions of what the shapes of these animals were. We could really only use this process to revive animals that existed relatively recently, like Mammoths.

However, there is another option. Animals evolve through mutations to the DNA, which are passed on to their offspring. As a species splits, both new resulting species will carry the same inherited sequences as each other, but begin to develop new and unique ones. If you knew the genome sequences of every animal, you could conceivably work out the ‘consensus’ sequence of the last common ancestor of any two groups: the set of gene sequences which are shared by all of the animals, subtracting the new ones which each group has generated since their division.

Birds are the descendants of dinosaurs that survived the extinction, and are most closely related to dinosaurs such as Velociraptor. You could in theory use this approach to work out the genome sequence of the last common ancestor of ‘Ratite’ birds like Ostriches and ‘Neoavian’ birds like Finches. This bird is predicted to have lived before the extinction of the dinosaurs, and would thus represent a species that lived alongside animals like the Tyrannosaurus rex. You could also work out the last common ancestor of birds and crocodiles, or birds and mammals, both of which would be lizards that predated the dinosaurs.

Alternatively, you could just genetically engineer a bird to be more like the dinosaurs we imagine. All animals bear the baggage of our evolutionary history, and birds are no exception. Recently scientists at Harvard suppressed genes within a chicken that led to the embryo developing a crocodile-like snout. Chickens can also be induced to grow dinosaur-like teeth. This sort of experimental ‘atavism’ could allow us to engineer a ‘dinosaur’ from a bird. You could also pair this with the above ‘consensus’ approach to explore the effect of certain mutations along the course of evolution between the different known genomes, but not when it comes to branches of the tree of life which have no surviving members, such as the families of Stegosaurus and Triceratops.

There’s actually a fan theory that this is the real methodology behind Jurassic Park, and that John Hammond has invited the palaeontologists to the island to see if he can ‘fool’ them into believing his ruse. Proponents of the theory say this is why the dinosaurs look exactly like artists impressions of dinosaurs, when the probability of them looking exactly like our drawings of the animals is almost zero. For example, it’s now widely believed that dinosaurs would have had feathers. Also the island is full of extinct plants, despite the fact that mosquitoes do not feed from plants. Finally, mosquitos would be a terrible way to get DNA, as any would be digested by the insects digestive systems long before preservation could set in.

So, is Jurassic Park feasible? Perhaps: but only if you are comfortable with taking significant liberties with the concept, and with lying to Sam Neil.

References and further reading

How to extract DNA using household chemicals:


Illumina four-color sequencing by synthesis:


First Self-Replicating Synthetic Bacterial Cell:


Debate over human-mouse mix:


DNA has a 521-year half-life:


‘Dino-chickens’ reveal how the beak was born:


Mutant Chicken Grows Alligator-like Teeth: