Publications
Selected Papers
All
Selected Papers
Cell, 187:25 P7045-7063, (2024)
Cells are essential to understanding health and disease, yet traditional models fall short of modeling and simulating their function and behavior. Advances in AI and omics offer groundbreaking opportunities to create an AI virtual cell (AIVC), a multi-scale, multi-modal large-neural-network-based model that can represent and simulate the behavior of molecules, cells, and tissues across diverse states. This Perspective provides a vision on their design and how collaborative efforts to build AIVCs will transform biological research by allowing high-fidelity simulations, accelerating discoveries, and guiding experimental studies, offering new opportunities for understanding cellular functions and fostering interdisciplinary collaborations in open science.
Nature Reviews Methods Primers, 4:58 (2024)
High-throughput single-cell profiling provides an unprecedented ability to uncover the molecular states of millions of cells. These technologies are, however, destructive to cells and tissues, raising practical challenges when aiming to track dynamic biological processes. As the same cell cannot be observed at multiple time points, as it changes in time and space in response to a stimulus or perturbation, these large-scale measurements only produce unaligned data sets. In this Primer, we show how such challenges can be effectively addressed using the unifying framework of optimal transport theory and tackled using the many algorithms that have been proposed for the range of scenarios of key interest in computational biology. We further review recent advances integrating optimal transport and deep learning that allow forecasting heterogeneous cellular dynamics and behaviour, crucial in particular for pressing problems in personalized medicine.
Nature Methods volume 20, pages 1759–1768 (2023)
Understanding and predicting molecular responses in single cells upon chemical, genetic or mechanical perturbations is a core question in biology. Obtaining single-cell measurements typically requires the cells to be destroyed. This makes learning heterogeneous perturbation responses challenging as we only observe unpaired distributions of perturbed or non-perturbed cells. Here we leverage the theory of optimal transport and the recent advent of input convex neural architectures to present CellOT, a framework for learning the response of individual cells to a given perturbation by mapping these unpaired distributions. CellOT outperforms current methods at predicting single-cell drug responses, as profiled by scRNA-seq and a multiplexed protein-imaging technology. Further, we illustrate that CellOT generalizes well on unseen settings by (1) predicting the scRNA-seq responses of holdout patients with lupus exposed to interferon-β and patients with glioblastoma to panobinostat; (2) inferring lipopolysaccharide responses across different species; and (3) modeling the hematopoietic developmental trajectories of different subpopulations.
All
2025
NeurIPS
MTBBench: A Multimodal Sequential Clinical Decision-Making Benchmark in Oncology
Advances in Neural Information Processing Systems (NeurIPS) Datasets and Benchmarks Track, 2025
Test-Time View Selection for Multi-Modal Decision Making
International Conference on Learning Representations (ICLR) Workshop on Machine Learning for Genomics Explorations, 2025
Contributed Talk at ICLR MLGenX Workshop, 2025
Best Paper
AI-powered virtual tissues from spatial proteomics for clinical diagnostics and biomedical discovery
arXiv Preprint arXiv:2501.06039, 2025
Best Paper Award at the ICLR MLGenX Workshop, 2025
ICLR
Modeling Complex System Dynamics with Flow Matching Across Time and Conditions
International Conference on Learning Representations (ICLR), 2025
Spotlight Talk at ICLR (Top 5.2 percent). Also presented at NeurIPS Workshop on AI for New Drug Modalities, 2024
Best Paper
AISTATS
Cross-modality Matching and Prediction of Perturbation Responses with Labeled Gromov-Wasserstein Optimal Transport
International Conference on Artificial Intelligence and Statistics (AISTATS), 2025
Best Paper Award and Contributed Talk at ICML AI for Science Workshop, 2024
Nature Methods
Reproducible image-based profiling with Pycytominer
Nature Methods, 2025
2024
Cell
How to build the virtual cell with artificial intelligence: Priorities and opportunities
Cell, 2024
Nature Reviews
Optimal transport for single-cell and spatial omics
Nature Reviews Methods Primers, 2024
3DReact: Geometric Deep Learning for Chemical Reactions
Journal of Chemical Information and Modeling, 2024
2023
Transition Path Sampling with Boltzmann Generator-based MCMC Moves
arXiv preprint arXiv:2312.05340, 2023
Unbalanced Diffusion Schrödinger Bridge
arXiv Preprint arXiv:2306.09099, 2023
ICML Workshop on New Frontiers for Learning, Control, and Dynamical Systems, 2023
Nature Methods
Learning Single-Cell Perturbation Responses using Neural Optimal Transport
Nature Methods, 2023
Highlighted as Research Briefing in Nature Methods. Also presented at the NeurIPS Workshop on Optimal Transport and Machine Learning (OTML), 2021
UAI
Aligned Diffusion Schrödinger Bridges
Conference on Uncertainty in Artificial Intelligence (UAI), 2023
Also presented at the ICML Workshop on New Frontiers for Learning, Control, and Dynamical Systems, 2023
Neural Unbalanced Optimal Transport via Cycle-Consistent Semi-Couplings
arXiv Preprint arXiv:2209.15621, 2023
Spotlight Talk at NeurIPS Meaningful Representations of Life Workshop, 2022
AISTATS
The Schrödinger Bridge between Gaussian Measures has a Closed Form
International Conference on Artificial Intelligence and Statistics (AISTATS), 2023
Oral Presentation at AISTATS, Top 1.9 percent of Submitted Papers. Also presented at the ICML Workshop on Continuous Time Methods for Machine Learning, 2022
2022
NeurIPS
Supervised Training of Conditional Monge Maps
Advances in Neural Information Processing Systems (NeurIPS), 2022
Also presented at the ICML Workshop on Interpretable Machine Learning in Healthcare (IMLH), 2022
Best Paper
AISTATS
Proximal Optimal Transport Modeling of Population Dynamics
International Conference on Artificial Intelligence and Statistics (AISTATS), 2022
Best Paper Award and Contributed Talk at ICML Time Series Workshop, 2021
ICLR
Independent SE (3)-Equivariant Models for End-to-End Rigid Protein Docking
International Conference on Learning Representations (ICLR), 2022
Spotlight Talk at ICLR and Ranked Top 15 among 3326 Submissions (Top 0.4 percent) and Contributed Talk at ELLIS Machine Learning for Molecule Discovery Workshop, 2021
Machine intelligence for chemical reaction space
Wiley Interdisciplinary Reviews: Computational Molecular Science, 2022
Selected as Featured Cover of Volume 12, Issue 5
Optimal Transport Tools (OTT): A JAX Toolbox for all things Wasserstein
arXiv Preprint arXiv:2201.12324, 2022
Invariant Causal Mechanisms through Distribution Matching
arXiv Preprint arXiv:2206.11646, 2022
2021
NeurIPS
Multi-Scale Representation Learning on Proteins
Advances in Neural Information Processing Systems (NeurIPS), 2021
Spotlight Talk at ICML Computational Biology Workshop, 2021
Best Paper
NeurIPS
Learning Graph Models for Retrosynthesis Prediction
Advances in Neural Information Processing Systems (NeurIPS), 2021
Best Paper Award and Contributed Talk at ICML Workshop on Graph Representation Learning and Beyond, 2019
2020
COSIFER: a Python package for the consensus inference of molecular interaction networks
Bioinformatics, 2020
2019
Best Paper
ICML
Learning Generative Models across Incomparable Spaces
International Conference on Machine Learning (ICML), 2019
Best Paper Award and Contributed Talk at NeurIPS Workshop on Relational Representation Learning, 2018
Exponential Growth of Glioblastoma In Vivo Driven by Rapidly Dividing and Outwardly Migrating Cancer Stem Cells
bioRxiv Preprint bioRxiv:10.1101/723601v1, 2019
2015
Backbone circularization of Bacillus subtilis family 11 xylanase increases its thermostability and its resistance against aggregation
Molecular BioSystems, 2015