13
CONNECTOMICS
micrometre scale. On the other hand, volume electron
microscopy (EM) can resolve neurons and their
connection sites, but is painfully slow, given its need
to image samples one two-dimensional slice at a time.
The biggest result based on EM, published in October
this year by dozens of international labs making up the
“FlyWire” consortium, was the first connectome of the
entire brain of the common fruit fly – a feat that took
five years of data collection, and a further five years of
data processing.
X-rays from the ESRF–EBS have the potential to
greatly accelerate the generation of larger connectomes
– and, in doing so, transform our view of the brain. At
the ID16A beamline, X-ray holographic nano-
tomography (XNH) has much higher resolution than
MRI – around 20 to 30 nanometres, which is enough
to resolve neurons and their connections. But the
technique also offers important speed benefits, because
the nature of X-rays allows them to deeply penetrate
samples and directly produce three-dimensional
images, without having to laboriously slice up or
“ablate” samples and image them, again and again.
When ESRFnews first highlighted the work of
ESRF scientist Alexandra Pacureanu in 2020, she
had completed a proof-of-principle study, having just
received a grant from the European Research Council.
Four years later, several results are in, and they are
already having far-reaching impacts – whether
revealing how creatures coordinate movement, or
inspiring robotics, or providing insights into autism.
All in the detail
According to Pacureanu, it’s all possible because of
the Extremely Brilliant Source (EBS). “ID16A was
built with the EBS in mind,” she says. “For us that’s a
big advantage, as our technique relies on the EBS X-ray
coherence – it’s what provides that quality of data.”
Despite the EBS, finessing XNH has not been
straightforward Pacureanu and her colleagues have
had to work hard refining their approaches to data
reconstruction installing new detectors adjusting
mechanical stabilising systems and developing
machinelearning protocols to obtain the detail they
want Clearly it has been worth it as the reproduction of
a entire pygmy squids connectome on this page testifies
Taken by a team led by Albert Cardona at the MRC
Laboratory of Molecular Biology in Cambridge UK in
collaboration with Pacureanu the multiterabyte sized
dataset depicts the first whole complex organism imaged
at subcellular level The work is not yet published still
an image of this nature has potential to help answer
4
“Our technique relies on the EBS
X-ray coherence – it’s what provides
that quality of data”
December 2024 ESRFnews