Reproduce “Horizontal Flow Matching for Single-Cell Spatiotemporal 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
Zihan Wang, Jianzhe Li, Zhenyi Zhang, Peijie Zhou
HORIZON models spatial transcriptomics time courses by jointly aligning tissue snapshots, matching cell populations, and learning trajectories. It removes shared translation and rotation from spatial motion during training and generation, while modeling molecular-state change and abundance. On a simulated severe-frame benchmark, its direct spatial distance was 0.115 ± 0.003 versus 1.213 ± 0.027 for WFR-FM; rigid tissue motion remains unidentifiable.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.