The first map of a brain, of a fly larva, brings closer the dream of understanding the human mind

Humanity had only been able to map cell by cell three tiny nervous systems with a few hundred neurons: that of the laboratory worm

, that of the larva of the marine invertebrate

, and that of a tiny animal that lives attached to the rocks. ocean rocks,

.

A team headed by the Spanish biologist Albert Cardona and his Croatian colleague Marta Zlatic have now signed off on a scientific feat: the map of the complete brain of the fruit fly larva, a structure with 3,016 neurons and 548,000 connections between them.

"We have multiplied by 10 what had been achieved so far," celebrates Cardona, from the legendary Laboratory of Molecular Biology in Cambridge (United Kingdom), whose scientists have won a dozen Nobel prizes.

The biologist, born in Tarragona 44 years ago, explains the magnitude of the progress.

“Let's imagine that a city metro had 3,000 stations and each one of them was connected to another 200 ″, he illustrates.

The complexity of the brain of the larva of the fruit fly, however, pales in comparison with the most sophisticated structure on the face of the Earth: the human brain, an organ of a kilo and a half with 86,000 million neurons.

“Three thousand neurons seems very few, but this larva is capable of navigating gradients of light or odors, it can find food by itself, it has short- and long-term memory.

It is a very self-sufficient animal”, points out Cardona.

magazine

, a showcase for the best world science.

The person who began to map the human brain was the Spaniard Santiago Ramón y Cajal, in 1888. With a rudimentary microscope in his laboratory in Barcelona, ​​the researcher demonstrated that the organ of the mind was not a diffuse mass, as was thought. until then, but was organized in individual cells: neurons.

Cajal then began a titanic task, masterfully drawing by hand each brain structure, cell by cell, with its connections, which he poetically called “kisses”.

The Spanish biologist Albert Cardona, in his office at the Cambridge Molecular Biology Laboratory.Albert Cardona

Albert Cardona's team has used more sophisticated methods.

A dozen years ago, scientists removed the nervous system of a fruit fly larva with tweezers.

They cut it into about 5,000 ultra-thin slices and observed them under an electron microscope.

The biologist devised

that allows these images to be precisely joined —just like a mobile phone joins several photos into a single panorama— and to navigate through that three-dimensional volume, as if it were Google Maps.

The applications of a brain map are unimaginable.

Cardona cites the work of a colleague from his laboratory, neurobiologist Pedro Gómez Gálvez, one of the Spanish scientists who in 2018 announced the discovery of new geometric shapes: scutoids, a kind of twisted prisms observed for the first time in salivary glands. of fruit flies.

Gómez Gálvez is comparing the whole brains of normal larvae with those of other larvae that have been genetically modified to mimic the symptoms of Parkinson's.

Other little-understood disorders—such as autism, schizophrenia, and epilepsy—arise from a deviation from typical brain development.

The American physicist Emerson Pugh left a sentence for history before he died in 1981: "If the human brain were so simple that we could understand it, we would be so simple that we would not understand it."

It is the paradox of the brain, a structure so sophisticated that it is incapable of imagining itself.

Cardona, however, is optimistic.

She believes that getting a map of the human brain with its neuron-to-neuron connections—the so-called connectome—is only a matter of time.

“The mouse brain is going to be made in the next 10 or 15 years.

The question is how much it will cost.

There are several projects proposing it, but we are talking about between 500 and 1,000 million dollars just to do the preliminary work”, calculates the biologist.

"And the human brain will require an absurd amount of resources," she predicts.

A fruit fly adult, a pupa and a larva of the same insect. Weigmann et al.

Cardona explains that his colleague Gregory Jefferis is already mapping the brain of the adult fly in Cambridge.

The results are expected from next year.

Another obvious target would be the honey bee, with a million neurons.

“We have to get into the bee's brain, because it has the ability to speak and to remember specific places across miles of landscapes.

As it does?

How do you explain to another bee how to go to a place?

All of this can be studied if its neural wiring is known”, says Cardona.

The neuroscientist Rafael Yuste, professor at Columbia University (USA), considers the map of the larva's brain "spectacular".

This researcher was in the laboratory of the South African biologist Sydney Brenner, also in Cambridge, in 1985, when this team made a first attempt to map the 302 neurons of

.

That study had a provocative title: The Mind of the Worm.

Yuste remembers that that pioneering work was very traditional, "almost heroic", whereas now it is practically an industrial process.

In his opinion, the progress towards more complex brains is “inexorable”.

Yuste is one of the promoters of the future Spain Neurotech National Center for Neurotechnology, in Madrid.

“It is very difficult to carry out these studies, huge teams are needed, with a lot of investment of time and work.

That is why it is important to coordinate national and international funding and efforts in neurotechnology.

Mapping the mouse connectome is being studied in a worldwide collaboration”, argues the professor.

The brain of a mouse is a million times larger than that of a fly larva.

Albert Cardona says that they have found a "surprise" in the brain of the larva.

Its architecture closely resembles modern artificial neural networks, such as ResNet, DenseNet, and U-Net, used in sophisticated machine learning software.

“In traditional neural networks, each layer of neurons only connects to the next.

The crux of the matter is connections that skip layers.

That is the root of his exceptional abilities”, explains Cardona.

The Spanish biologist considers "amazing" what these beings are capable of doing with only 3,000 neurons.

Cardona points out that the larvae of the fruit fly, like many other insects, usually have a parasitic wasp inside, like the monster in the movie

.

"The larva of the fly detects it and will eat food enriched in alcohol, fermented fruit, to medicate, because that alcohol kills the parasite inside it," says the researcher.

Cardona underlines the organizational complexity of those 3,000 neurons.

In addition to layer hopping, there are looping connections, similar to LSTM artificial neural networks, which are used on billions of computers daily.

The biologist is confident that the brain of the fly larva will give rise to new artificial intelligence systems, with more powerful automatic learning than the current ones.

"There are already computer scientists taking inspiration from the brain circuits of our larva," he applauded.

In the long term, the goal is much more ambitious, as another of the map's co-authors, biomedical engineer Joshua Vogelstein, from Johns Hopkins University (USA), has proclaimed: "Understanding who we are and how we think."

You can write to us at