BIMSA >
Seminar on Statistics and Data Analysis
Learning complex cellular dynamics from time-series scRNA-seq data
Learning complex cellular dynamics from time-series scRNA-seq data
Organizers
Speaker
Lin Wan
Time
Tuesday, May 6, 2025 4:00 PM - 5:00 PM
Venue
A3-2-303
Online
Zoom 435 529 7909
(BIMSA)
Abstract
The emergence of time-series single-cell RNA sequencing (scRNA-seq) data provides unprecedented opportunities to study the complex dynamics of heterogeneous cellular systems. However, analyzing temporal single-cell transcriptome snapshots remains challenging, particularly in linking gene expression measurements across time points. To address this, there is a need for advanced mathematical models and machine learning methods that can reconstruct collective cell population behaviors, including cell-cell interactions. In this talk, I will present our recent work on a mean-field modeling and dynamic learning framework designed for temporal scRNA-seq data. This framework leverages neural networks to resolve interacting mean-field systems, enabling the reconstruction of intrinsic cell population dynamics and the characterization of cell-cell interactions.