New_Scientist_3_08_2019

12 | New Scientist | 3 August 2019

IS THE placenta home to

communities of bacteria? Recent

evidence that the organ has its

own microbiome promised

to rewrite our understanding

of it, but now research suggests

that the earlier experiments may

have been contaminated.

Although the gut is known to

host a rich, diverse community

of bacteria, as well as viruses

and fungi, the fetus and

placenta had long been

assumed to be sterile.

That picture started to change

seven years ago, when evidence

emerged that babies are born

with a gut microbiome, which

seems to start taking shape

before birth. Then in 2014,

a team led by Kjersti Aagaard

at Baylor College of Medicine

in Texas found low levels of

bacteria in the placenta.

Julian Parkhill at the Wellcome

Sanger Institute in Cambridge,

UK, and his colleagues wanted

to find out if the make-up

of the placenta’s microbiome

might influence the outcome

of a pregnancy. But as their

research progressed, they came

to a different conclusion: that

the placenta doesn’t seem to

have a microbiome after all.

Parkhill and his team

performed tests on samples

of placenta taken from 578

women. In one experiment, they

found that the same strain of the

Escherichia coli bacterium was

present in all of the placenta

samples (Nature, DOI: 10.1038/

s41586-019-1451-5).

“That’s impossible from a

cohort of hundreds of pregnant

women,” he says. The bacteria

must have come from the

experiment kit, he says.

What Parkhill sees as

contamination, Aagaard views

as evidence of a microbiome.

While the microbiome of the

placenta might not be as rich

and diverse as that of the gut,

she says, it could still play an

important role in birth outcomes

and a baby’s metabolic and

brain development.

The debate continues.

“Believe you me, I am as

exhausted by the seemingly

endless back and forth as

anyone is,” says Aagaard. ❚

A COMPUTER chip inspired

by the brain could pave the way

for artificial intelligence with

a broader range of abilities.

The chip has been used to control

an autonomous bicycle, but one

day it might power self-driving

cars and smart robots.

Shi Luping at Tsinghua

University in China and his

colleagues made the chip with

hardware based on the structure

of the human brain that can run

two types of algorithm.

To date, most approaches to

developing AIs fall into two camps.

The first, and more common, type

of algorithm is artificial neural

networks, which are simplistic

models of neurons designed for

specific computer processing

tasks such as recognising objects.

Then there are biologically

inspired circuits, known as

spiking neural networks, which

are rarer. These aim to more

closely replicate the human brain,

with its billions of neurons that

communicate via electric pulses.

Whereas neurons in artificial

neural networks can “fire” at

different intensities, neurons

in spiking neural networks are

either on or off.

Shi’s chip is built to deal with

the operations required by both

algorithms, so it is potentially

more efficient.

His team has shown how the

hybrid chip could be used in an

autonomous bicycle. The bike

combines several different

algorithms for different tasks,

allowing it to respond to voice

commands, to self-balance,

avoid obstacles, and detect

and follow a person.

Brain-inspired chip

could transform AI

Machine learning Pregnancy

Donna Lu Jessica Hamzelou

STEVE ALLEN PHOTO/GETTY

News

Its processor, known as the

Tianjic chip, combined five

different neural networks,

including a spiking neural

network that recognises voice

commands such as “straight”

and “speed up” (Nature, DOI:

10.1038/s41586-019-1424-8).

The algorithms were all trained

before being loaded onto the chip,

where they were run in parallel in

the bicycle. Shi says the computer

chip could be used for other

applications including intelligent

robots and self-driving cars.

Alessandro Oltramari at

engineering firm Bosch says

the hybrid chip may be suitable

for a variety of tasks. “You really

need to have an architecture that

is complex enough, that can deal

with different problems using

different mechanisms,” he says.

But more general AIs are still

some way off, he adds. Although

the Tianjic chip was effective for

autonomous navigation in a

bicycle, to prove its general ability,

it would need to succeed at a range

of totally different tasks, such as

manipulating objects, playing

games or holding a conversation.

Oltramari says the chip

overlooks a third approach that

may be required for artificial

general intelligence: cognitive-

inspired neural architectures.

Instead of trying to replicate the

anatomical structures of the brain,

this aims to mimic the cognitive

abilities of the human mind,

such as its ability to pay attention,

remember and make predictions. ❚

The placenta

supplies food and

oxygen to a fetus

“One day, this flexible
computer chip might
power self-driving cars
and smart robots”

Debate reignited

over the placenta’s

microbiome

PEI ET AL., NATURE.

A new

computer

chip has been

used to control

a self-riding

bicycle