Neural Noise

Neuron signal extraction suite for one-photon calcium imaging across sessions

2 min read

  • A comprehensive suite exists for extracting neuron signals across multiple sessions in one-photon calcium imaging. 1
  • The work falls under neuroinformatics and neural data analysis; published in Nature (tier-1). 1
  • Modality is optical imaging, not electrophysiology (e.g. EEG/ECoG), and is down-weighted relative to electrophysiology for BCI. 1 1

Weekly enrichment (2026-07-20)

  • The suite is named CaliAli (Calcium Imaging intersession Alignment), published in Nature Communications on 2025-04-11 (DOI 10.1038/s41467-025-58817-z) by 16 authors across 6 affiliations, including the University of Tsukuba (WPI-IIIS), the Icahn School of Medicine at Mount Sinai, and the University of Tokyo; note the journal is Nature Communications, not the flagship Nature.2
  • CaliAli extracts neuronal signals from one-photon calcium imaging recorded in free-moving mice, and uniquely incorporates information from both blood vessels and neurons to correct inter-session misalignments, making it robust to non-rigid brain deformation and large field-of-view changes across sessions.2
  • The pipeline has two stages: a multi-level, group-wise registration that aligns each session to a common atlas via non-rigid displacement fields, followed by a customized CNMF-E (Constrained Nonnegative Matrix Factorization for one-photon data) module that extracts denoised, demixed signals from the concatenated video stack.2
  • CaliAli runs in a memory-optimized batch mode to overcome computational constraints on long recordings, enhances detectability of weak calcium signals via concatenation, and includes a post-processing interface for flagging false-positive components.2
  • Reported validations: improved spatial-coding accuracy of hippocampal CA1 neurons across sessions; an optogenetic-tagging experiment showed better neuronal trackability in the dentate gyrus over a timescale of weeks; and dentate-gyrus neurons tracked with CaliAli exhibited stable population activity for 99 days.2
  • For BCI relevance, this is an optical-imaging analysis method (rodent, calcium) rather than electrophysiology, so its direct BCI decoding applicability is limited; its value is longitudinal ensemble tracking that could inform stability assumptions for chronic neural interfaces.2
  • Context: CaliAli builds on the CNMF-E lineage; an alternative longitudinal pipeline, SCOUT (Single-Cell SpatiOtemporal LongitUdinal Tracking), adds a spatial filter plus a predictor-corrector for cross-session cell registration, tracked hippocampal ensembles up to 60 days, and reported outperforming CellReg on simulated and in vivo data.34
  • Context: Minian is an open-source miniscope analysis pipeline with five stages (background/vignetting correction, motion correction, seed-based initialization, CNMF, and optional cross-session registration), representing the open-source baseline that suites like CaliAli aim to improve on for multi-session work.5

Footnotes

  1. https://news.google.com/rss/articles/CBMiX0FVX3lxTFBXOUhQRnFyTmlFb1cwZ3lBemdEekk5bFFBck1pVnhLNXlrUkViNHNXbkY1NEFOUWhPTHRILURhdjVYM3VBSXc0TzRLcllZX2xEUnB5NXdBcDdOdmlsWWRz?oc=5 ↩ ↩2 ↩3 ↩4

  2. https://www.nature.com/articles/s41467-025-58817-z ↩ ↩2 ↩3 ↩4 ↩5 ↩6

  3. https://doi.org/10.1101/2020.08.26.268151 ↩

  4. https://github.com/kgj1234/SCOUT ↩

  5. https://www.biorxiv.org/content/10.1101/2021.05.03.442492v3 ↩

Graph View