Track / Overview

The scientific community is currently witnessing rapid technological advancements in artificial intelligence (AI) applied to the natural sciences. Some notable breakthroughs have involved the domains of proteins and small molecules, where innovation in AI has allowed to accurately predict the protein structure from sequence [1, 2], design novel bioactive molecules in a ‘rule-free’ manner [3–6], protein design and optimization [7, 8], and organic synthesis prediction [9–12]. Such breakthroughs have been driven by unique synergies between deep learning developments and biological/chemical knowledge.

The domains of small molecules and proteins are highly intertwined, and synergies between these domains bear tremendous potential to tackle multiple scientific challenges, e.g., to accelerate drug discovery, design safer molecules and for receptor ‘de-orphaning’.

Despite this untapped potential, the communities hardly interact and are progressively diverging in the methods and applications. Moreover, experiments are often ne- glected in AI-focused research, but are crucial to validate the predictions and generate value in adjacent scientific domains. Future breakthroughs will be possible only thanks to interdisciplinary research, at the interface between domain knowledge, method development, and experimental application.

How will AI help us to design the small molecules and proteins of tomorrow? In this track, we will bring together researchers with multidisciplinary backgrounds (e.g., chem- and bioinformatics, deep and machine learning, medicinal chemistry and chemical biology, and protein engineering) to discuss grand challenges in biology and chemistry, which only become ac- cessible when experts from diverse fields meet and interact.

Track / Schedule

Introduction

Optimal Transport Modeling of Population Dynamics: Applications in Single-Cell Biology

With Charlotte Bunne

Interpretable Embeddings from Molecular Simulations using Gaussian Mixture Variational Autoencoders

With Yasemin Bozkurt Varolgünes

Graph Neural Networks beyond Weisfeiler-Lehman and vanilla Message Passing

With Michael Bronstein

Tailored de Novo Protein Design with Deep Neural Networks

With Zander Harteveld

Goal-directed Generation of Molecules with Conditional Generative Models

With Amina Mollaysa

Regression Transformer: Concurrent Conditional Molecular Generation and Regression by Blending Numerical and Textual Tokens

With Jannis Born

Break & Posters

Protein Language Models: Generative Properties and Data Representation

With Anne-Florence Bitbol

Learning from Literature-Extracted Synthesis Actions for Organic Synthesis

With Alain Vaucher

Optimization Enriched with Knowledge Learned from the Past

With Loïc Roch

Machine Learning Aided Discovery of Membrane Disruptive, Non-Hemolytic Anticancer Peptides

With Markus Orsi

Wrap-up

Track / Speakers

Amina Mollaysa

PhD Student, University of Geneva

Jannis Born

PhD student, IBM Research Zurich

Michael Bronstein

Oxford / Twitter / IDSIA

Alain Vaucher

Research Scientist, IBM Research Europe

Loïc Roch

CTO and co-Founder, Atinary Technologies

Markus Orsi

PhD Student, Universität Bern

Yasemin Bozkurt Varolgünes

Postdoc, Max Planck Institute for Polymer Research

Zander Harteveld

PhD Student, EPFL

Anne-Florence Bitbol

Assistant Professor, EPFL

Charlotte Bunne

PhD Student, ETH Zurich

Track / Co-organizers

Philippe Schwaller

Assistant Professor, EPFL

Ilia Igashov

PhD Student, EPFL

Francesca Grisoni

Assistant Professor, TU/e

Théophile Gaudin

PhD Student, University of Toronto

Bruno Correia

Tenure Track Assistant Professor, EPFL

AMLD EPFL 2022 / Tracks & talks

AMLD Keynote Session – Monday morning

Marcel Salathé, Lenka Zdeborová, Carmela Troncoso, Chiara Enderle, Patrick Barbey, Thomas Wolf, Gunther Jansen, Laure Willemin, Simon Hefti, Arthur Gassner

10:00-12:00 March 28Auditorium A

AI & Physics

Francesca Mignacco, Gert-Jan Both, Michael Unser, Thomas Asikis, Dalila Salamani, Pietro Rotondo, Tom Beucler, Giulio Biroli

12:30-18:00 March 285BC

AI & Pharma

Asif Jan, Jonas Richiardi, Patrick Schwab, Naghmeh Ghazaleh, Alexander Büsser, Carlos Ciller, Caibin Sheng, Silvia Zaoli, Félix Balazard, Giulia Capestro, Marianna Rapsomaniki, Martijn van Attekum

13:30-17:30 March 281BC

AMLD / Global partners