Solved, riddle that could cure the incurable

AI unlocks proteins that cause disease

SCIENTISTS have used artificial intelligence to achieve a ‘stunning’ historic breakthrough which may lead to successful treatment for currently incurable diseases.

World-leading AI company DeepMind – owned by Google and based in London – yesterday announced that it had cracked what is known as the ‘protein folding problem’.

The challenge had puzzled researchers for 50 years and some experts had feared it would never be solved.

Proteins are large and complex molecules, made up of chains of amino acids, and each has a unique 3D structure which they ‘fold’ into. It is such a complicated puzzle because there are so many proteins in the body and their 3D shapes are difficult to map out.

Almost all diseases are related to the way these proteins function. Protein ‘misfolding’ is believed to be the primary cause of Alzheimer’s, Parkinson’s, Huntington’s, CJD and cystic fibrosis.

There are 200million known proteins at present but only a fraction have been ‘unfolded’ to fully understand what they do and how they work – a process which usually takes years. But DeepMind’s AlphaFold program has been found to determine highly accurate structures in a matter of days.

Professor Venki Ramakrishnan, Nobel Laureate and president of the Royal Society, said: ‘This computational work represents a stunning advance on the protein-folding problem, a 50-year- old grand challenge in biology.

‘It has occurred decades before many people in the field would have predicted.’

He added: ‘It will be exciting to see the many ways in which it will fundamentally change biological research. It is hoped that protein structure predictions could contribute to understanding specific diseases, for example helping to identify proteins that have malfunctioned.’

DeepMind – co-founded by Cambridge-educated Demis Hassabis, a former child chess prodigy – has worked on the project since 1994, along with a group of scientists called the 14th Community Wide Experiment on the Critical Assessment of Techniques for Protein Structure Prediction (CASP14).

The group’s chair Dr John Moult said: ‘Proteins are extremely complicated molecules, and their precise three-dimensional structure is key to the many roles they perform. For example the insulin that regulates sugar levels in our blood and the antibodies that help us fight infections.

‘Even tiny rearrangements of these vital molecules can have catastrophic effects on our health, so one of the most efficient ways to understand disease and find new treatments is to study the proteins involved.’ During the lat

‘Catastrophic effects on health’

est test, DeepMind said their program was able to determine the shape of around two thirds of the proteins with accuracy comparable to lab experiments.

But researchers behind the project say there is still more work to be done, including figuring out how multiple proteins form complexes – a group of two or more chains – and how they interact with DNA.

DeepMind is planning to submit a paper detailing its system to a peer-reviewed journal to be scrutinised by the wider scientific community.