Welcome and Registration
Our team will be ready to assist with your registration and answer any queries. Coffee and light breakfast will be provided. Don't miss our swag buffet where you can pick up a few useful giveaways!
* All times are based on Canada/Eastern EDT.
Canada/Eastern
Canada/Eastern
Canada/Eastern
With AI transforming drug discovery, it can be difficult to separate breakthroughs from hype. This keynote offers a clear-eyed assessment of what AI is truly achieving today, what remains aspirational, and where the field is headed next. Derek Lowe - Author of In The Pipeline and Director in Exploratory Chemical Biology at Novartis
Canada/Eastern
2 parallel sessionsThis session features experts creating the benchmarks needed to measure progress, tackle key challenges, and build a strong and collaborative computational drug discovery community. Rethinking Rigor: Towards Community Standards for the Validation of AI Models in Biomedicine Gustavo Stolovitzky - Dir. Biomedical Data Sci. Hub at NYU Langone Health & Founding Chair, DREAM Challenges The explosive growth of data-driven artificial intelligence in biomedicine is reshaping the scientific landscape, powering innovative healthcare solutions and transformative insights in life sciences. AI is proving to be a new paradigm for predictive modeling, which in turn demands a paradigm shift in our approach to model evaluation. Traditional algorithmic performance metrics alone are insufficient for comprehensive validation. Concurrent with the rise of AI modeling opportunities has been a proliferation of community-led benchmarking initiatives, fragmenting efforts and limiting their collective synergy and impact. This talk proposes fostering a coalition of Critical Assessment efforts. Such a coalition can enhance collaboration, establish rigorous validation standards, and build sustained trust in evolving AI methodologies, ensuring their robust and reliable integration into biomedical research. Popularizing best practices through the Polaris benchmarking platform Cas Wognum - Senior Scientist at Valance Labs Machine learning (ML) is driving innovations in drug discovery, but we need to be mindful of the circumstances that set this application apart. The absence of standardized, domain-appropriate datasets, guidelines and tools for the evaluation and comparison of methods has led to a growing gap between perceived progress and real-world impact, delaying the adoption of ML in drug discovery. To bridge this gap, we formed Polaris: An open-science, cross-industry and interdisciplinary collaboration to develop benchmarking protocols tailored to drug discovery. With Polaris, we go beyond publishing papers and develop a multimodal benchmarking platform for machine learning in drug discovery to simplify the adoption of these protocols. This talk will share some of the guiding design principles we followed when building Polaris and discuss the role of open-source software in popularizing appropriate benchmarking protocols.
The Developing Medicines through Open Science (DMOS) program is designed to foster collaborative drug discovery and development in areas often overlooked by the pharmaceutical industry, such as rare and orphan diseases and antimicrobial resistance. The inaugural round of this program is offering a total of $5M in funding and partnership opportunities to support significant advancements toward drug candidates. In this session, the first three DMOS grantees will present their projects. Small Molecule PRMT6 Inhibitors for the Treatment of Spinal Bulbar Muscular Atrophy (SBMA) Peter Sampson - VP of Drug Discovery and Development, Agora Open Science Open Drug Discovery to Treat Primary Sclerosing Cholangitis Sonya MacParland - Senior Scientist, University Health Network Pre-Clinical Development of an ALK2 Inhibitor for the Treatment of Diffuse Intrinsic Pontine Glioma (DIPG) Peter Sampson - VP of Drug Discovery and Development, M4K Pharma
Canada/Eastern
2 parallel sessionsChallenges have long been a catalyst for AI breakthroughs, with CASP’s role in AlphaFold’s success a prime example. This session highlights ongoing Challenges in AI-driven drug discovery, explores key lessons learned, and brings together experts to discuss how these efforts can complement each other to push the field forward. The CACHE Computational Hit Finding Challenges: Lessons Learnt so Far Matthieu Schapira - Professor in the Department of Pharmacology and Toxicology and PI, Research Informatics, Structural Genomics Consortium, University of Toronto The Critical Assessment of Computational Hit-finding Experiments (CACHE) are prospective benchmarking challenges to outline the state-of-the-art in computational hit finding, accelerate progress and set the stage for future breakthroughs. A review of the first three CACHE challenges shows that active learning to accelerate virtual screening and fragment-based design can be successful strategies. The field is clearly in an exploratory mode, where physics-based and machine learning approaches are combined in a multitude of possible ways, but a breakthrough remains to be seen ASAP Discovery: A Community Blind Challenge for Open Source Structure-enabled Drug Discovery for Pandemic Preparedness Jenke Scheen - Senior Scientist at ASAP Discovery & Open Molecular Software Foundation During the SARS-CoV-2 pandemic the COVID Moonshot consortium demonstrated the feasibility of developing a straight-to-generics Mpro inhibitor in record time. Building on this success, the ASAP Discovery Consortium was launched as an AViDD center that leverages structure-enabled and artificial-intelligence approaches together with a global team to tackle structural biology, medicinal chemistry, biochemistry and (pre-)clinical trials for antiviral drug discovery. With over three years of drug discovery under its belt, ASAP Discovery hosted a community blind challenge using the Polaris platform in collaboration with Valence Labs and OpenADMET. The challenge consisted of several sub-challenges (ligand potency, pose and ADMET predictions) designed to be closely analogous to the challenges that ASAP Discovery faced during these campaigns. During this talk, the background to the data will be outlined as well as the challenge outcomes and the practical processes considered during the design of this challenge
A central challenge in drug development programs aimed at solving unmet health needs is to figure out how to incentivize development while accelerating research. In this session, we will explore the interplay of open science and intellectual property, identifying strategies to move drug development forward. Richard Gold - Chief Policy and Partnerships Officer at Conscience, CIGI senior fellow, and James McGill Professor with McGill University's School of Law
Canada/Eastern
As drug discovery embraces openness, research is shifting from isolated labs and private collaborations to dynamic, large-scale communities and open platforms. This session explores some of the most innovative initiatives driving this transformation and shaping the future of drug discovery. MAINFRAME, an international, Open Science, ML network to advance pre-competitive drug discovery Albert A. Antolin - Principal Investigator at the Bellvitge Biomedical Research Institute (IDIBELL) and the Catalan Institute of Oncology (ICO) We recently launched MAINFRAME—an international, open-science network of machine learning scientists and computational chemists. MAINFRAME seeks to harness the data produced by Target2035 and AI to develop more robust, open-source hit identification methods for the benefit of all. The release of binding data orders of magnitude greater than what is currently available offers an unprecedented opportunity to enhance AI-driven hit identification and develop more generalizable, foundational models. At MAINFRAME, we are committed to keeping these models open-source and accessible to the whole scientific community. We invite all interested scientists to join us in this mission. Can We Build Roller Coasters? Karmen Čondić-Jurkić - Executive Director at Open Molecular Software Foundation The Open Molecular Software Foundation (OMSF) is a nonprofit organization dedicated to advancing the development, accessibility, and sustainability of open-source software in the molecular sciences. By fostering collaboration among researchers from academia and industry, OMSF aims to accelerate scientific progress and innovation in the world of molecules. This talk will highlight OMSF's mission to create a sustainable ecosystem for open molecular software, its key initiatives, and how the community can contribute to and benefit from its efforts. It will also cover some of the challenges and learnings we made on the way.
Canada/Eastern
Each flash talk presenter will give a 3-minute presentation. This session will be immediately followed by our cocktail and networking reception. Each flash talk presenter will be assigned a numbered table at the reception, during which attendees can find and engage with presenters, who may choose to bring handouts or a laptop to aid discussions (similar to a poster session, but without the posters!)