IT Pro is supported by its audience. When you purchase through links on our site, we may earn an affiliate commission. Learn more

Google DeepMind uses AI tech to predict protein shapes

It's one of the most accurate and fastest ways of determining very complex 3D shapes

Brain, network, network intelligence, AI, computing, connected, artificial intelligence, machine learning

The latest scientific application of Google's DeepMind AI project is identifying and predicting the complexity of proteins via its AlphaFold discovery programme.

In doing so, the reasons some of the most widespread genetic illnesses exist could potentially be uncovered.

DeepMind has brought together some of the most advanced thinkers in structural biology, physics, and machine learning to predict 3D shapes in proteins, based on its genetic sequence.

AlphaFold has been running for more than two years and the team behind the project thinks it's the most accurate way of predicting the 3D shape of a protein yet. This means it could potentially uncover faults in DNA that causes conditions such as Alzheimer's, Parkinson's, Huntington's and cystic fibrosis.

"The challenge is that DNA only contains information about the sequence of a protein's building blocks called amino acid residues, which form long chains," the DeepMind Project team said in a blog post.

"Predicting how those chains will fold into the intricate 3D structure of a protein is what's known as the "protein folding problem. The bigger the protein, the more complicated and difficult it is to model because there are more interactions between amino acids to take into account."

According to researchers, it would take longer than the age of the universe to work out every possible configuration of proteins before working out the 3D structure. But AlphaMind's advanced intelligence and modelling system could get to the bottom of the complexities.

Using neural networks, the AlphaFold project has managed to determine the distances between pairs of amino acids and the angles between chemical bonds that connect those amino acids. These are then matched with structures that are already known. The next stage was training a neural network to invent new fragments, which could improve the accuracy of the model.

"The success of our first foray into protein folding is indicative of how machine learning systems can integrate diverse sources of information to help scientists come up with creative solutions to complex problems at speed,"

"Just as we've seen how AI can help people master complex games through systems like AlphaGo and AlphaZero, we similarly hope that one day, AI breakthroughs will help us master fundamental scientific problems, too."

Featured Resources

Accelerating AI modernisation with data infrastructure

Generate business value from your AI initiatives

Free Download

Recommendations for managing AI risks

Integrate your external AI tool findings into your broader security programs

Free Download

Modernise your legacy databases in the cloud

An introduction to cloud databases

Free Download

Powering through to innovation

IT agility drive digital transformation

Free Download

Most Popular

Actively exploited server backdoor remains undetected in most organisations' networks
cyber attacks

Actively exploited server backdoor remains undetected in most organisations' networks

1 Jul 2022
Macmillan Publishers hit by apparent cyber attack as systems are forced offline
Security

Macmillan Publishers hit by apparent cyber attack as systems are forced offline

30 Jun 2022
Former Uber security chief to face fraud charges over hack coverup
data breaches

Former Uber security chief to face fraud charges over hack coverup

29 Jun 2022