Celebrating the 2024 Nobel Prize in Physics and Chemistry

Friday 6 Dec 2024, 15:00 → 19:05 Europe/Stockholm

FD5 (AlbaNova Main Building)

The 2024 Nobel Prizes in Physics and Chemistry highlight significant advancements in artificial intelligence (AI) and machine learning (ML) that are reshaping scientific research. Physics laureates John Hopfield and Geoffrey Hinton pioneered methods in associative memory networks and Boltzmann machines, enabling AI to recognize patterns and autonomously extract properties from data. In Chemistry, David Baker, Demis Hassabis, and John Jumper are recognized for their innovative work in computational protein design and structure prediction. Baker's groundbreaking protein designs and the success of AlphaFold2 have revolutionised our understanding of biological processes and the development of new therapies.

The EDUCATE initiative within the OKC invites everyone to celebrate this year's Nobel Prizes and explore the impact of AI and ML across diverse scientific disciplines. Experts from mathematics, biology, climatology, and geology will share insights into current applications, challenges, ethical considerations, and future directions of these technologies.

We encourage students and researchers from the Faculty of Science to join us for an engaging afternoon that fosters interdisciplinary collaboration and inspires future generations of scientists to harness the power of AI in their work.

 

1 Opening Remarks

2 Patrick Bryant: Structure prediction and design of protein interactions

The advent of AlphaFold revolutionized our understanding of protein structure prediction, delivering atomic-level accuracy for individual proteins. In this talk, I will present our advances in leveraging and extending the capabilities of predictive models like AlphaFold to predict and design functional protein complexes. Our work builds on the structural insights provided by AlphaFold, developing methodologies to predict and design binders directly from sequence data. This approach has led to the creation of EvoBind, a platform capable of designing cyclic peptide binders with sub-nanomolar affinities in a single shot.

3 Anneli Kruve: Discovering toxic chemicals with machine learning

Development of chemical industry over the last century enabled our comfortable life; however, the technology is imperfect and many unknown chemicals with unknown effects end up in the consumer products directly impacting humans or are emitted to the environment thorough waste streams. Due to the fact that these chemicals are unknown, novel data driven methods are needed detecting and understanding the impact of these chemicals. In this talk I will give an overview of the machine learning approaches that have been recently developed for discovering toxic chemicals and how combining chemical methods with machine learning can improve over understanding of the chemical exposure.

16:10 Break and refreshments

4 Golnaz Taheri: From Genes to Algorithms: How AI is Reshaping Life sciences

In this talk, I explore the transformative role of AI and ML in tackling some of the most critical challenges in computational biology. I will discuss how AI/ML, as powerful tools, offer solutions to decode complex biological systems, predict drug interactions, and advance cancer research. Highlighting the synergy between innovative methods and practical applications, this presentation focuses on two key research areas: identifying critical genes and pathways in cancer using deep learning and developing a novel framework for drug-drug interaction prediction using biological networks. By integrating cutting-edge AI/ML techniques with real-world challenges, we aim to contribute to shaping the future of life sciences.

5 Chun-Biu Li: If you can't explain machine learning simply, you don't understand it well enough

Despite their high popularity in both academic and industrial applications, machine learning (ML) and AI usually give the impression of blackbox models, existence of artifacts, counterintuitive results, diversity of methods, complicated training, etc. In this talk, I will share my research and teaching experiences in ML/AI in order to promote the importance of building up intuitive and comprehensive understanding of ML/AI methods for students and researchers who are interested in applying ML/AI to their studies.

17:30 Break and refreshments

6 Panel discussion and closing remarkds

7 Networking session