Modeling Microenvironment Trajectories on Spatial Transcriptomics with NicheFlow

2025-11-02

AI Summary

Existing methods struggle to jointly model the spatiotemporal co-evolution of cellular microenvironments in spatial transcriptomics, failing to integrate local neighborhood structure with global spatial dynamics. To address this, we propose the first streaming generative model for spatial transcriptomics, unifying point-cloud representation, optimal transport, and variational flow matching to simultaneously capture continuous evolution of both cellular states and spatial coordinates. Our method explicitly infers spatiotemporal trajectories of cellular microenvironments across multiple time points, enabling concurrent reconstruction of tissue-level spatial architecture and dissection of local cell–cell interaction dynamics. Validated on real spatiotemporal datasets—including embryonic and brain development—it accurately recapitulates microenvironmental evolutionary pathways, significantly outperforming existing single-cell or purely spatial modeling approaches. The framework establishes a novel, interpretable, and generalizable paradigm for spatiotemporal modeling, advancing developmental biology and disease mechanism studies.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Web Mining and Content Analysis: Models for Web evolution
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: User model development and evaluation

Abstract

Understanding the evolution of cellular microenvironments in spatiotemporal data is essential for deciphering tissue development and disease progression. While experimental techniques like spatial transcriptomics now enable high-resolution mapping of tissue organization across space and time, current methods that model cellular evolution operate at the single-cell level, overlooking the coordinated development of cellular states in a tissue. We introduce NicheFlow, a flow-based generative model that infers the temporal trajectory of cellular microenvironments across sequential spatial slides. By representing local cell neighborhoods as point clouds, NicheFlow jointly models the evolution of cell states and spatial coordinates using optimal transport and Variational Flow Matching. Our approach successfully recovers both global spatial architecture and local microenvironment composition across diverse spatiotemporal datasets, from embryonic to brain development.

Problem

• Modeling temporal trajectory of cellular microenvironments
• Inferring coordinated evolution of cell states spatially
• Recovering spatial architecture across developmental datasets

Innovation

• Flow-based generative model for microenvironment trajectories
• Uses optimal transport and Variational Flow Matching
• Models cell state evolution with spatial coordinates