Timezone: »

TrajectoryNet: A Dynamic Optimal Transport Network for Modeling Cellular Dynamics
Alexander Tong · Jessie Huang · Guy Wolf · David van Dijk · Smita Krishnaswamy

Wed Jul 15 05:00 AM -- 05:45 AM & Wed Jul 15 04:00 PM -- 04:45 PM (PDT) @ Virtual #None

It is increasingly common to encounter data in the form of cross-sectional population measurements over time, particularly in biomedical settings. Recent attempts to model individual trajectories from this data use optimal transport to create pairwise matchings between time points. However, these methods cannot model non-linear paths common in many underlying dynamic systems. We establish a link between continuous normalizing flows and dynamic optimal transport to model the expected paths of points over time. Continuous normalizing flows are generally under constrained, as they are allowed to take an arbitrary path from the source to the target distribution. We present {\em TrajectoryNet}, which controls the continuous paths taken between distributions. We show how this is particularly applicable for studying cellular dynamics in data from single-cell RNA sequencing (scRNA-seq) technologies, and that TrajectoryNet improves upon recently proposed static optimal transport-based models that can be used for interpolating cellular distributions.

Author Information

Alexander Tong (Yale University)
Jessie Huang (Yale University)
Guy Wolf (Université de Montréal; Mila)
David van Dijk (Yale University)
Smita Krishnaswamy (Yale University)

Smita Krishnaswamy is an Assistant Professor in the department of Genetics at the Yale School of Medicine and Department of Computer Science in the Yale School of Applied Science and Engineering. Smita’s research focuses on applying machine learning methods to high-throughput high dimensional biological data. Smita has been focusing on using manifold learning and deep learning to develop unsupervised algorithmic approaches to naturally process data, visualize it, understand progressions , find phenotypic diversity, and infer patterns. Some of the key projects developed in her Lab include MAGIC (a tool for imputation and denoising of data), PHATE (a powerful new visualization method for high dimensional data that can unveil progression and cluster structures, and SAUCIE (an autoencoder-based deep learning approach for automatically batch correcting, visualizing, denoising and clustering data). These methods have been applied to a variety of biological applications including embryoid body differentiation, the epithelial-to-mesenchymal transition in breast cancer, lung cancer immunotherapy, infectious disease data, gut microbiome data and population genetics data. At Yale, Smita teaches two CS/Genetics/Computational Biology cross-listed courses. Advanced Topics in Machine Learning & Data Mining (Spring), and Machine Learning for Biology (Fall). She completed her postdoctoral training at Columbia University in the systems biology department where she focused on learning computational models of cellular signaling from single-cell mass cytometry data. She was trained as a computer scientist with a Ph.D. from the University of Michigan’s EECS department where her research focused on algorithms for automated synthesis and verification of nanoscale logic circuits that exhibit probabilistic effects.

More from the Same Authors