The focus of the proposed workshop was the mathematical analysis of models used to learn low-dimensional em-beddings and dynamical projections of molecular dynamical systems. The workshop convened mathematicians, statisticians, theoretical chemists, and biophysicists to discuss theoretical and numerical techniques, to define open challenges and pressing needs in the field, and to establish new disciplinary bridges and collaboration opportunities. Although primarily mathematical in nature, the history of this field has taught us that the most challenging questions can only be tackled by a synthesis of interdisciplinary methods and tools. The workshop promoted participation of early-career and female researchers. Successful outcomes of the workshop will include a freely available online workshop report and the establishment of new collaborations, and the definition of set of open questions to guide future work in the field.
Workshop
Dimensionality reduction techniques for molecular dynamics

02 Jun 2025 - 06 Jun 2025

ICMS, Bayes Centre, Edinburgh
Scientific Organisers
About Workshop
Programme
09.00-09.45
Registration and refreshments
09.45-10.00
Welcome and housekeeping
10.00-12.00
Marina Meila, University of Washington
Tutorial - Manifold learning from the user's perspective
12.00-13.30
Lunch
13.30-15.30
Andrew Ferguson, University of Chicago
Tutorial - Molecular Latent Space Simulators (LSSs): Spatially and Temporally Continuous Data-Driven Surrogate Dynamical Models of Molecular Systems
15.30-16.00
Lightning talks
16.00-17.30
Poster session and welcome reception hosted at ICMS
09.15-10.00
Jutta Rogal, Flatiron Institute
10.00-10.45
Hao Wu, Shanghai Jiao Tong University
Flow based dimension reduction for molecular kinetics
10.45-11.15
Refreshments
11.15-12.00
Pilar Cossio, Flatiron Institute
Cryo-electron microscopy images are low-dimensional
12.00-12.45
Paraskevi Gkeka, Sanofi R&D
Enhancing Sampling in Molecular Dynamics: Integrating Autoencoders and Linear Discriminant Analysis for the identification of collective variables
12.45-14.00
Lunch
14.00-14.45
Mauro Maggionni, Johns Hopkins University
Nonlinear Model Reduction for Slow-Fast Stochastic Systems near Unknown Invariant Manifolds
14.45-15.30
Feliks Nüske, Max-Planck-Institute DCTS Magdeburg
Kinetically Consistent Coarse Graining using Kernel-based Extended Dynamic Mode Decomposition
15.30-16.00
Refreshments
18.00-19.00
Public Lecture hosted in G.03 (ground floor)
ChatGPT for proteins: Designing molecular machines using text prompts by Andrew Ferguson, University of Chicago
09.15-10.00
Carsten Hartmann, BTU Cottbus-Senftenberg
Coarse graining of diffusion processes in the absence of time scale separation
10.00-10.45
Maria Cameron, University of Maryland
Learning collective variables for accurate transition rate estimation
10.45-11.15
Refreshments
11.15-12.00
Bernd Ensing, University of Amsterdam
12.00-12.45
Jérôme Hénin, CNRS
12.45-14.00
Lunch
14.00-14.45
Gerhard Hummer, Max Planck Institute of Biophysics
Learning from molecular simulations
15.30-16.00
Refreshments
19.00
Workshop dinner hosted at Playfair Library
10.00-10.45
Thomas Pigeon, IFPEN
Approximating committor functions: Objective functions and training data sampling
10.45-11.15
Refreshments
11.15-12.00
Wei Zhang, Zuse Institute Berlin
Mathematical aspects of deep-learning techniques for identifying collective variables of molecular dynamics
12.00-12.45
Neelanjana Sengupta, Indian Institute of Science Education and Research (IISER) Kolkata
Effectiveness of Machine Learned Collective Variable Projection in Bio-molecular Energy Landscapes
12.45-14.00
Lunch
14.00-14.45
Antonia Mey, University of Edinburgh
14.45-15.30
Grigorios Pavliotis, Imperial College London
Clustering for interacting particle systems with short range attractive potentials
15.30-16.00
Refreshments
09.15-10.00
Sapna Sarupria, University of Minnesota
Seeing the invisible: Learning Pathways to Polymorphs through machine learning analysis of atomic trajectories
10.00-10.45
Hong Duong, University of Birmingham
Ergodicity and asymptotic limits for (relativistic) Langevin dynamics
10.45-11.00
Refreshments
11.00-11.45
Gareth Tribello, Queen's University Belfast
Reconnaissance metadynamics Rides Again
11.45
Lunch and end of workshop
Sponsors and Funders