Reproduce “STAGED: A Multi-Agent Neural Network for Learning Cellular Interaction Dynamics” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Joao F. Rocha, Ke Xu, Xingzhi Sun, Ananya Krishna, Dhananjay Bhaskar, Blanche Mongeon, Morgan Craig, Mark Gerstein and 1 more
STAGED is a multi-agent deep neural network that infers dynamic cell-cell interactions and intracellular gene regulatory networks directly from spatial transcriptomics data. This framework models cells as interacting agents, successfully recovering population dynamics in tumor simulations, reconstructing oscillatory GRNs in synthetic data, and identifying distinct regulatory programs in microglia during Alzheimer's disease.
Key insights
- Integrating agent-based modeling with Graph Neural ODEs enables data-driven learning of complex, dynamic intercellular and intracellular interactions.
- Modeling cells as interacting agents, rather than independent entities, allows for a more comprehensive understanding of how their internal states are shaped by their neighbors.
- The framework simultaneously infers both dynamic gene regulatory networks within cells and the communication pathways between them.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.