AI and chemistry: Why a Nobel Prize is just the beginning

BY Steve Ranger | Editor in Chief

The award of the 2024 Nobel Prize for Chemistry has put the use of artificial intelligence in science, and computational chemistry in particular, in the spotlight again.

In October, biochemist David Baker won the award for his work on building new kinds of proteins, alongside Demis Hassabis and John Jumper who shared the chemistry Nobel for their work on developing AI models, which have allowed researchers to predict the complex structures of proteins.

What made this year’s award unusual is that Hassabis and Jumper work for tech company Google DeepMind – as CEO and senior research scientist, respectively – and these Nobel prizes are more usually won by academic researchers.

In 2020, Hassabis and Jumper presented an AI model called AlphaFold 2. Using it, they have been able to predict the structure of virtually all the 200 million known proteins. The tool has now been used by more than 2m people from 190 countries and it has helped researchers to do everything from better understand antibiotic resistance to create images of enzymes that can decompose plastic.

Earlier in 2024, the latest version – AlphaFold 3 – was launched which goes beyond proteins to a broad spectrum of biomolecules, something that Google said could help unlock more transformative science, from developing biorenewable materials and more resilient crops, to accelerating drug design and genomics research.

‘Life could not exist without proteins. That we can now predict protein structures and design our own proteins confers the greatest benefit to humankind,’ said the Royal Swedish Academy of Sciences, when announcing the award.

AI was also part of the award for the Nobel Prize for physics, which went to John Hopfield and Geoffrey Hinton for their foundational discoveries and inventions that enable machine learning with artificial neural networks.

The two awards reflect how AI is having an increasing impact on science, creating new options and requirements for research.

Charlotte Deane, Professor of structural bioinformatics at the University of Oxford, UK, said it is an exciting time to be working in science, particularly in these interdisciplinary areas, as AI not only starts solving really hard problems but also changing the way we do science.

‘Proteins are the functional components in every living organism so being able to predict their shape and design them with a specific shape and function has ramifications across all of medicine, biology and many other areas,’ she said.

The use of AI to predict protein structure is a huge advance with ‘a myriad’ of uses in biology, medicine, and beyond, said Andy Cooper, Director of the Materials Innovation Factory at the University of Liverpool, UK.

‘A challenge for AI in chemistry more broadly is to deal with areas where data are sparse, less systematised, and where compositional diversity is much greater – energy materials is one such example,’ he said.

Ewan Birney, Deputy Director General of the European Molecular Biology Laboratory (EMBL), headquartered in Heidelberg, Germany, noted that tools like AlphaFold are built on decades of experimental work and made possible thanks to a culture inside molecular biology of openly sharing data worldwide.

‘There is a vast treasure trove of public data available in databases such as the ones managed by EMBL. We hope to see these data informing yet more discoveries. The potential of big data alongside AI and technology developments is limitless – and this is the start,’ Birney said.

Jeremy Frey, Professor of physical chemistry at the University of Southampton. UK, said the award of the Nobel Prize in chemistry for the prediction and design of protein structures, illustrates the need for data. It would not have been possible without the experimentally obtained structures in the Protein Data Bank, and also shows the importance of chemical knowledge built into the model pipelines.

‘AlphaFold and Rosetta have made enormously useful contributions to the analysis of protein structure and have accelerated many investigations,’ Frey noted.

‘There is still so much to be done; unstructured proteins, the structure of DNA, RNA protein interactions, and many other aspects of the dynamics and function of proteins. As with many Nobel prizes, the prize highlights the beginning not the end of the area.’