Back to publications
Multimodal alignments of in vivo imaging and spatial biology datasets at cellular resolution

Authors:

Lun Wang, Xiqian Jiang, Xiaochen Sun, Gaurav Chattree, Ali Cetin, Xiaochun Cai, Elijah Paul, Radosław Chrapkiewicz, Oscar Hernandez, Yuxi Ke, Tomohisa Yoda, Fatih Dinc, Bariscan Kurtkaya, Yanping Zhang, Zhengji Zhang, Mark J. Schnitzer

Publication:

bioRxiv

Publication Date:

01/05/2026

The parallel revolutions in spatial biology and in vivo microscopy have transformed our ability to record cellular dynamics and profile molecular signatures—yet linking activity patterns to molecular identity in the exact same cells has remained a major bottleneck. The TRU-FACT Workflow:
  • Align: An optomechanical strategy physically preserves the orientation of in vivo imaging planes, ensuring postmortem tissue sections are cut precisely parallel—critical for reliable cell matching across modalities.
  • Fingerprint: The Soma-print algorithm registers cells by their unique geometric relationship to neighbors, delivering robust matching even with dense populations, thin sections, or distorted tissue.
  • Integrate: A statistical framework that provides for each cell an a posteriori probability of correct registration, enabling true multimodal integration—molecular identity, connectivity, and function—for the same individual cells.
Validated across 13 mice which gave 10,522 matched cells, the pipeline works with both low-plex (HCR-FISH) and high-plex (MERFISH, 500-gene) spatial biology methods. The TRANSVISTA miniature two-photon microscope enabled the complete TRU-FACT workflow in freely behaving mice, seamlessly connecting naturalistic behavior, cellular activity, and molecular identity in the exact same cells.