Summer School Schedule
Registering for the summer school also grants you access to MoML 2024. You can buy tickets for MoML separately here.
Time & Location
June 12th - 21st
Summer school: June 12th - 18th
-
Location: (Centre PHI - 407 Rue Saint-Pierre, Montréal, QC H2Y 2M3, Canada)
MoML 2024: June 19th
-
Location: (Mila Agora- 6650 Rue Saint-Urbain, Montréal, QC H2S 3G9, Canada)
Hackathon: June 20th - 21st
-
Location: (Centre PHI - 407 Rue Saint-Pierre, Montréal, QC H2Y 2M3, Canada)
Program Details
Day 1 (Wed, June 12th) - Foundations and ML in ligand-based modeling:
In this module, we start off with an introduction to the exciting field of applied machine learning in drug discovery. Then, we will explore how machine learning techniques are used for 2D and 3D ligand modeling. From understanding molecular representations to model explicability, the participant will learn how ligand-based molecular models are trained and deployed to accelerate drug discovery.
​
Schedule:
-
8:00 am - 9:00 am: Registration
-
9:00 am - 9:30 am: Opening Remarks
-
9:30 am - 10:30 am: Intro to ML in Drug Discovery: Principles & Applications - Bharath Ramsundar
-
10:30 am - 11:30 am: Molecular Representation & Scoring - Emmanuel Noutahi
-
11:30 am - 12:30 pm: Graph Neural Networks for Chemistry - Dominique Beaini
-
12:30 pm - 2:00 pm: Lunch
-
2:00 pm - 3:00 pm: Learning Geometry & 3D Symmetries - Mario Geiger
-
3:00 pm - 3:30 pm: Lab 1 - Virtual Screening Explanation
-
3:30 pm - 4:00 pm: Break
-
4:00 pm - 5:00 pm: Lab 1 - Virtual Screening
-
5:00 pm - 5:30 pm: Lab 1 Recap
Day 2 (Thurs, June 13th) - ML in Structure-Based Drug Discovery
In this module, we explore how machine learning is applied to structure-based drug discovery. From foundational concepts to advanced applications, we will cover how ML methods can help with binding affinity prediction, intermolecular interactions modeling, free energy computations, QM and MD simulations, and predictive modeling.
​
Schedule:
-
8:30 am - 9:30 am: ML in Structure-Based Drug Discovery - Gabriele Corso
-
9:30 am - 10:30 am: Learning ML Interatomic Potentials - Gianni De Fabritiis
-
10:30 am - 11:00 am: Break
-
11:00 am - 12:00 pm: Accelerate Atomistic Simulations, Sampling, and Dynamics - Pratyush Tiwary
-
12:00 pm - 1:30 pm: Lunch
-
1:30 pm - 2:30 pm: Coarse-Grained Biological Systems - Jacopo Venturin
-
2:30 pm - 3:00 pm: Lab 2 - Binding Affinity Prediction with ML-Based Docking Explanation
-
3:00 pm - 3:30 pm: Break
-
3:30 pm - 4:30 pm: Lab 2 - Binding Affinity Prediction with ML-Based Docking
-
4:30 pm - 5:00 pm: Lab 2 Recap
Day 3 (Fri, June 14th) - Generative Models and Molecular Design:
This module uncovers the potential of machine learning in reshaping drug discovery through generative models of molecular structures. From understanding the exploration in molecular space and active learning to exploring the intricacies of molecular synthesis, this module provides the tools to navigate the unknown chemical space to find molecules that satisfy properties of interest.
​
Schedule:
-
8:30 am - 9:30 am: Generative Models of Molecular Structures - Camille Bilodeau
-
9:30 am - 10:30 am: Exploring Molecular Space & Active Learning - Yoshua Bengio
-
10:30 am - 11:00 am: Break
-
11:00 am - 12:00 pm: Synthesizability & Molecular Synthesis - Connor Coley
-
12:00 pm - 1:30 pm: Lunch
-
1:30 pm - 2:30 pm: Sampling Physical & 3D Systems - Michael Bronstein
-
2:30 pm - 3:00 pm: Lab 3 - De Novo Generation Explanation
-
3:00 pm - 3:30 pm: Break
-
3:30 pm - 4:30 pm: Lab 3 - De Novo Generation
-
4:30 pm - 5:00 pm: Lab 3 Recap
Day 4 (Mon, June 17th) - Target Discovery and deconvolution
This module explores the synergy between omics data, molecular data, unveiling novel insights in target identification, and mechanism discovery. From causal discovery to representation learning and single-cell population dynamics, this module equips participants to navigate the intricate landscape of biological mechanisms using cutting-edge omics and AI techniques.
​
Schedule:
-
8:30 am - 9:30 am: Phenomics in Drug Discovery - Anne E. Carpenter
-
9:30 am - 10:30 am: Multi-Modal Omics & AI - Sébastien Lemieux
-
10:30 am - 11:00 am: Break
-
11:00 am - 12:00 pm: Causal Discovery & Representation Learning - Jason Hartford
-
12:00 pm - 1:30 pm: Lunch
-
1:30 pm - 2:30 pm: Modeling Population Dynamics - Charlotte Bunne
-
2:30 pm - 3:00 pm: Lab 4 - Target Deconvolution Explanation
-
3:00 pm - 3:30 pm: Break
-
3:30 pm - 4:30 pm: Lab 4 - Target Deconvolution
-
4:30 pm - 5:00 pm: Lab 4 Recap
Day 5 (Tues, June 18th) - Frontiers in AI for Drug Discovery:
In this module, we embark on an insightful exploration of artificial intelligence's expanding role in drug discovery, with a keen focus on ethical considerations and the potential for bias. We will dive into topics such as the application of Large Language Models (LLMs) to chemistry and the impact of foundation models in drug discovery. We will also engage in discussions around open-source initiatives, benchmarking efforts, and ethical concerns to deepen your understanding and empower you as responsible researchers. The module concludes with a panel discussion and Q&A session, providing a platform for collaborative reflection and forward-thinking dialogue.
​
Schedule:
-
8:30 am - 9:30 am: LLMs in Drug Discovery - Andres Camilo Marulanda Bran
-
9:30 am - 10:30 am: Protein Folding & Design - Alexander Tong
-
10:30 am - 11:00 am: Break
-
11:00 am - 12:00 pm: Open-Source Initiatives & Benchmarking Efforts - Karmen Condic-Jurkic
-
12:00 pm - 1:30 pm: Lunch
-
1:30 pm - 2:30 pm: Ethical & Bias Concerns - Afaf Taik
-
2:30 pm - 3:00 pm: Closing remarks
June 20th - 21st
Hackathon
Put to test the newfound knowledge and skills in the grand finale of our summer school journey! Apply machine learning techniques to real-world drug discovery problems.
​
June 20th
​
-
9:00 AM - 9:45 AM: Task Introduction + Q&A
-
10:00 AM - 4:00 PM: Hacking
-
4:30 PM - 5:00 PM: Winning team showcases methods
​
June 21st
-
9:00 AM - 9:45 AM: Task Introduction + Q&A
-
10:00 AM - 4:00 PM: Hacking
-
4:30 PM - 5:00 PM: Winning team showcases methods
