Planarian regeneration model discovered by artificial intelligence

June 4, 2015
Tufts University biologists developed an artificial intelligence system that for the first time reverse-engineered the regeneration mechanism of planaria--the small worms whose extraordinary power to regrow body parts makes them a research model in human regenerative medicine. The discovery presents the first model of regeneration discovered by a non-human intelligence and the first comprehensive model of planarian regeneration, which has eluded human scientists for a century. Credit: Daniel Lobo/Michael Levin-Tufts University

An artificial intelligence system has for the first time reverse-engineered the regeneration mechanism of planaria—the small worms whose extraordinary power to regrow body parts has made them a research model in human regenerative medicine.

The discovery by Tufts University biologists presents the first model of regeneration discovered by a non-human intelligence and the first comprehensive model of planarian regeneration, which had eluded human scientists for over 100 years. The work, published in the June 4, 2015, issue of PLOS Computational Biology, demonstrates how "robot science" can help human scientists in the future.

In order to bioengineer complex organs, scientists need to understand the mechanisms by which those shapes are normally produced by the living organism. However, a significant knowledge gap persists between molecular genetic components identified as being necessary to produce a particular organism shape and understanding how and why that particular complex shape is generated in the correct size, shape and orientation, said the paper's senior author, Michael Levin, Ph.D., Vannevar Bush professor of biology and director of the Tufts Center for Regenerative and Developmental Biology.

"Most regenerative models today derived from genetic experiments are arrow diagrams, showing which gene regulates which other gene. That's fine, but it doesn't tell you what the ultimate shape will be. You cannot tell if the outcome of many genetic pathway models will look like a tree, an octopus or a human," said Levin. "Most models show some necessary components for the process to happen, but not what dynamics are sufficient to produce the shape, step by step. What we need are algorithmic or constructive models, which you could follow precisely and there would be no mystery or uncertainty. You follow the recipe and out comes the shape."

Such models are required in order to know what triggers could be applied to such a system to cause regeneration of particular components, or other desired changes in shape. However, no such tools yet exist for mining the fast-growing mountain of published experimental data in regeneration and , said the paper's first author, Daniel Lobo, Ph.D., post-doctoral fellow in the Levin lab.

To address this challenge, Lobo and Levin developed an algorithm that would use evolutionary computation to produce regulatory networks able to "evolve" to accurately predict the results of published laboratory experiments that the researchers entered into a database.

"Our goal was to identify a that could be executed in every cell in a virtual worm so that the head-tail patterning outcomes of simulated experiments would match the published data," Lobo said.

As expected, the initial random regulatory networks usually could not produce any of the experimental results. New candidate networks were generated by randomly combining previous networks and performing random changes, additions and deletions. Each candidate network was tested in a virtual worm, under simulated experiments. The algorithm compared the resulting shape from the simulation with real published data in the database. As evolution proceeded, gradually the new networks could explain more experiments in the database comprising most of the known planarian experimental literature regarding head vs. tail regeneration.

Planaria algorithmic. Credit: Levin et al.

First Regenerative Model Discovered by Artificial Intelligence

The researchers ultimately applied the algorithm to a combined experimental dataset of 16 key planarian regeneration experiments to determine if the approach could identify a comprehensive regulatory network of planarian generation. After 42 hours, the algorithm returned the discovered regulatory network, which correctly predicted all 16 experiments in the dataset. The network comprised seven known regulatory molecules as well as two proteins that had not yet been identified in existing papers on planarian regeneration.

"This represents the most comprehensive model of planarian regeneration found to date. It is the only known model that mechanistically explains head-tail polarity determination in planaria under many different functional experiments and is the first regenerative model discovered by artificial intelligence," said Levin.

Lobo and Levin are both trained in computer science and bring an unusual perspective to the field of developmental biology. Levin majored in computer science and biology at Tufts before earning his Ph.D. in genetics. Lobo earned a Ph.D. in the field before joining the Levin lab.

The paper represents a successful application of the growing field of "robot science" - which Levin says can help human researchers by doing much more than crunch enormous datasets quickly.

"While the artificial intelligence in this project did have to do a whole lot of computations, the outcome is a theory of what the worm is doing, and coming up with theories of what's going on in nature is pretty much the most creative, intuitive aspect of the scientist's job," Levin said. "One of the most remarkable aspects of the project was that the model it found was not a hopelessly-tangled network that no human could actually understand, but a reasonably simple model that people can readily comprehend. All this suggests to me that can help with every aspect of science, not only data mining but also inference of meaning of the data."

Explore further: Flatworms lose their heads but not their memories: Study finds memories stored outside the brain

More information: Lobo D, Levin M (2015) Inferring Regulatory Networks from Experimental Morphological Phenotypes: A Computational Method Reverse-Engineers Planarian Regeneration. PLOS Comput Biol, 11(6): e1004295. DOI: 10.1371/journal.pcbi.1004295

Related Stories

PCP genetic pathway acts as stop sign for cell growth

March 15, 2012

The genetic pathway that regulates the way cells align themselves relative to each other has been found to act as a "stop sign" that signals organisms when to halt cell growth, according to new research published by biologists ...

Recommended for you

Making energy-harvesting computers reliable

October 28, 2016

A revolutionary and emerging class of energy-harvesting computer systems require neither a battery nor a power outlet to operate, instead operating by harvesting energy from their environment. While radio waves, solar energy, ...

Microsoft aims at Apple with high-end PCs, 3D software

October 26, 2016

Microsoft launched a new consumer offensive Wednesday, unveiling a high-end computer that challenges the Apple iMac along with an updated Windows operating system that showcases three-dimensional content and "mixed reality."


Adjust slider to filter visible comments by rank

Display comments: newest first

4.5 / 5 (2) Jun 04, 2015
This is amazing,The power of software programs to assist engineers and humans is becoming the only way forward. Engineers would be crippled without FEA\CFD\nSPICE etc software. This new future AI infrencing shown above will place another level of abstraction in the hands of designers which will allow goal oriented approaches to design.
1.5 / 5 (10) Jun 04, 2015
So biologists are complete garbage and can be replaced by a calculator.
1.8 / 5 (4) Jun 05, 2015
While the fact that AI discovered this genetic pattern, what is much more significant is that theoretically such augmentative genetic patterns can be used to improve the genome of a human adult now:


I realize this is a scientific article, but at the risk of straining my credibility further, ever hear of Wolverine and the X-Men? Let me add that with the digitizing of genes coming way down in price, and gigantic data-bases of that information being data-mined by AI, we can expect many more ideas on how to improve the human adult genome. This is the age of Transhumanism.
5 / 5 (5) Jun 05, 2015
Your god is a complete party pooper and that's why people have to redesign the world. I.e. if we are looking at the world from your point of view.
For sensible people it's just people exploring what they find interesting.
Dhiresh Yadav
not rated yet Jun 05, 2015
Yes, this kind of feats and leaps are possible in Artificial Intelligence but with the help of Human or any kind of advanced Natural Intelligence like Human or other Intelligence, of course. But, as I said in the discussion of "Google's genius futurist has one theory that he says will rule the future — and it's a little terrifying" by Drake Baer on Ray Kurzweil and his thesis-"the Law of Accelerating Returns." , AI>NI or HI, unless NI or HI is involved i.e. combined with AI, isn't possible.
not rated yet Jun 06, 2015
I'm sure they immediately changed the parameters to disallow the first solution and ran the 3 day experiment again to determine if there are other solutions?
5 / 5 (1) Jun 08, 2015
AI to humans: I'll be baaaaack!
3 / 5 (4) Jun 12, 2015
Ren82 claims
God says that the wages of sin is death
Eh ?
Which god ?
When did it last communicate & with who ?
What is its best method of communication ?
How reliable is this claimed communication, how do we check with this unknown god ?

Ren82 claims
And these scientists explore God's creations to discover the mechanisms that can make man immortal
Isnt it already known, via DNA & its enzymatic processes of replication ?

Ren82 asks
And they believe they will succeed?
Who knows, what has your god recently said about it & when exactly ?
How do I check any god's comment so I can confirm you haven't misheard ?

Ren82 claims
I think that they will be disappointed of this approach
Why ?
Over what period ?
What measurement method is suitable to your god ?

Ren82 claims
This is too high a ball human intelligence
Prove it ?

What has been the best means as to HOW your claimed god/Devil ever communicated ?

continued, why your god is really a Devil !
3 / 5 (4) Jun 12, 2015
Ren82 claimed
Who would tolerate eternal sin in this world?
What sin ?
Do you mean your god setting up Eve with a tree it put in the garden AND pointing it out AND knowing in advance Eve would be lied to by Satan which god specifically created to make sure she would fail, isn't this called "Entrapment" & viewed as the lowest form of ugly manipulation ?

So what sort of a loving god can yours possibly be Ren82, as it forces ALL of creation to suffer for EVER because of entrapment of a young woman of age of puberty ie 12

What sort of insane god/Devil is this ?

Is this god a GREAT example of how parents should treat their children AND grand-children ?

If not then WHY not, as it should suggest you'd IGNORE the gods example if you did not punish all your descendants for the rest of time as well ?

Since your god created immense suffering for EVERYONE, wouldn't it be logical you be against ANYONE that alleviates god's suffering ie Doctors & kill them Ren82 ?

3 / 5 (4) Jun 12, 2015
Ren82 observed
Who would tolerate eternal sin in this world?
We have that already, why haven't you noticed ?
So even if you accept all of jesus' writings & follow him then you STILL suffer as do your children & grand-children etc So NOTHING improves - the ONLY thing your belief in jesus does is make you feel better about your suffering - all due to a young woman.

This proves your god is really a Devil in disguise & has suspended critical thinking of followers ?

Ren82 claims
There is only one solution to this problem and every person on this planet has free access to it
Its called Science & its up to date, unlike your god who is ONLY the object of claim !

Ren82 claims
Where is artifical inteligence when software recreates and imitates the human way of thinking in embedded by programmers algorithm?
Your assumptions about modern IT re AI show you are ignorant of pattern recognition, Sparse Data Representation & other advanced AI far beyond you !

3 / 5 (4) Jun 12, 2015
Ren82 claims
There is no independent generation of solutions from the system created by man
What solutions do you imagine ?

Do you imagine god will make correct eg AGW & Pollution according to what communication ?

Nature, as claimed produced by a (personal human oriented) god, shows that everywhere it is "Eat & be Eaten", the predator/prey is all through nature as the chemical basis for all life interactions - which is evident in all higher life forms as:- Food, Shelter, Sex.

So what particular solutions has your claimed god ever communicated directly & when ?

Why can't your claimed god - which was only claimed in the middle east & the idea of it ONLY passed by lazy politically driven humans - ever communicate directly to humans without a priest ?

Ren82 claims
This system have no the ability to accumulate knowledge and to understand observational facts
Prove it !

Ren82 ONLY claims, like men who wrote about a claimed god but, god STILL completely silent !

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.