ΑΙhub.org
 

#ICLR2022 invited talk round-up 1: AI for science – protein structure prediction


by
29 April 2022



share this:

Probable disease resistance protein At1g58602Probable disease resistance protein At1g58602, from the AlphaFold Protein Structure Database, reproduced under a CC-BY-4.0 license.

This year’s International Conference on Learning Representations (ICLR) boasted eight invited talks on topics ranging from reinforcement learning to connectomics, from societal considerations to interpretability. In this article, we summarise the talk given by Pushmeet Kohli.

Leveraging AI for science

Pushmeet Kohli, DeepMind

Around five years ago, the DeepMind team started a science programme with the aim of using AI to pursue breakthroughs in impactful science challenges. In his talk, Pushmeet talked a bit about their approach, and focussed on the case study of AlphaFold2, which made a huge advance in the field of protein structure prediction.

Science offers many challenges and opportunities for AI. Through a variety of different experiments, such as high energy physics and genomics, there is a vast amount of data being produced. Many experiments are extremely complex and require a high-level of control. When it comes to scientific models, there is generally a high level of complexity and sophistication, a good example being in weather prediction. These are aspects (data, control, complexity) in which AI could play a key role in helping to advance different scientific disciplines.

Pushmeet highlighted three important elements of science, which also need to be considered when producing AI models. These are: generalisation (theories ideally hold universally), uncertainty prediction (we need to know how confident can we be about trusting a model), and explanability/interpretability.

Case study: Protein structure prediction with AlphaFold2
Proteins are large, complex molecules, and the shape of a particular protein is closely linked to the function it performs. The ability to accurately predict protein structures would enable scientists to gain a greater understanding of how they work and what they do. This close link between structure and function has been a key driver behind the problem of protein structure prediction.

It is extremely difficult and time-consuming to determine a protein structure experimentally. Typically this could take a researcher a year or more, just for one structure. By having an accurate prediction, this process could be greatly accelerated.

protein structure predictionScreenshot from Pushmeet’s talk.

In 1994, the community-wide experiment for protein structure prediction, Critical Assessment of protein Structure Prediction (CASP), held its first biannual competition. CASP provides an independent mechanism for the assessment of methods of protein structure modelling and teams of researchers take part to see who can provide the most accurate predictions. In the 2020 experiment, AlphaFold2 achieved a significant jump in performance as compared to previous years.

The AlphaFold2 architecture takes in an amino acid sequence, performs a genetic database search to find known structure of related proteins, and completes multiple sequence alignments. Sequence alignment is a way of arranging the sequences of proteins to identify regions of similarity. It then takes all of this information and feeds it to a neural network (with a transformer-type architecture) which works with a pair representation (considering the relationships between all the pairs of animo acids) and a multiple sequence alignment (MSA) representation. Finally, there is a structure module which outputs the final structure. You can read more about the method in the team’s 2021 paper: Highly accurate protein structure prediction for the human proteome. Another important facet of AlphaFold2 is that it does not only make structure predictions, it also predicts how much uncertainty/confidence there is for a particular prediction.

Pushmeet stressed that what enabled this work was the large collection of past data and research knowledge from years of previous study on protein structure, principally by the experimental biology community. He noted that resources such as the protein data bank, and the founding of CASP have been critical for the development of protein structure prediction methods.

To close his talk, Pushmeet briefly mentioned a number of other science problems that the team are working on. These include magnetic confinement control of plasmas for fusion, functional genomics, AI for mathematics and quantum chemistry.



tags:


Lucy Smith is Senior Managing Editor for AIhub.
Lucy Smith is Senior Managing Editor for AIhub.

            AIhub is supported by:



Subscribe to AIhub newsletter on substack



Related posts :

Studying the properties of large language models: an interview with Maxime Meyer

  11 Mar 2026
What happens when you increase the prompt length in a LLM? In the latest interview in our AAAI Doctoral Consortium series, we sat down with Maxime, a PhD student in Singapore.

What the Moltbook experiment is teaching us about AI

An experimental social media platform where only AI bots can post reveals surprising lessons about artificial intelligence behaviour and safety.

The malleable mind: context accumulation drives LLM’s belief drift

  09 Mar 2026
LLMs change their "beliefs" over time, depending on the data they are given.

RWDS Big Questions: how do we balance innovation and regulation in the world of AI?

  06 Mar 2026
The panel explores the tensions, trade-offs and practical realities facing policymakers and data scientists alike.

Studying multiplicity: an interview with Prakhar Ganesh

  05 Mar 2026
What is multiplicity, and what implications does it have for fairness, privacy and interpretability in real-world systems?

Top AI ethics and policy issues of 2025 and what to expect in 2026

, and   04 Mar 2026
In the latest issue of AI Matters, a publication of ACM SIGAI, Larry Medsker summarised the year in AI ethics and policy, and looked ahead to 2026.

The greatest risk of AI in higher education isn’t cheating – it’s the erosion of learning itself

  03 Mar 2026
Will AI hollow out the pipeline of students, researchers and faculty that is the basis of today’s universities?

Forthcoming machine learning and AI seminars: March 2026 edition

  02 Mar 2026
A list of free-to-attend AI-related seminars that are scheduled to take place between 2 March and 30 April 2026.



AIhub is supported by:







Subscribe to AIhub newsletter on substack




 















©2026.02 - Association for the Understanding of Artificial Intelligence