Summary: Following the same neurons over days and weeks has been a persistent challenge for neuroscientists using calcium imaging in freely moving mice. Researchers at the University of Tsukuba have developed an analytical framework called CaliAli that addresses these limitations by aligning imaging data across sessions with high accuracy, reconstructing continuous video of neural activity, and extracting clear neuronal signals while removing redundant detections and noise.
CaliAli rebuilds concatenated imaging sessions into a single coherent video, corrects shifts in the imaging field and tissue deformations, and applies an optimized signal-extraction algorithm. Using this approach with a standard ultra-miniature microscope, the team was able to reliably track the same neurons for as long as 99 days, enabling new opportunities to study long-term brain dynamics, memory processes, and the progression of neurological disorders.
Key Facts:
- Precision Tracking: CaliAli enables individual neurons to be tracked across sessions for up to 99 days.
- Advanced Alignment: The framework corrects field-of-view shifts and compensates for non-rigid tissue deformation between sessions.
- Noise Reduction: Automated extraction yields cleaner neural signals while filtering redundant or noisy detections.
Source: University of Tsukuba
Background: One-photon calcium imaging with ultra-small head-mounted microscopes has become a widespread method for visualizing neuronal activity in freely behaving mice, including during natural behaviors such as exploration and sleep. These miniature systems make it possible to record neural population dynamics in realistic settings, but they also introduce challenges when researchers attempt to identify and follow the same neurons over long time periods.
Typical analytical pipelines struggle with inter-session variability: small changes in how the microscope sits, shifts in the field of view, and subtle non-rigid deformations of brain tissue can all prevent reliable matching of neuronal footprints between sessions. These issues reduce the ability to trace how individual neurons change their activity or connectivity over time, limiting studies of memory consolidation, learning, aging, and disease progression.
To address these problems, the research team created CaliAli (Calcium Imaging inter-session Alignment), a comprehensive suite tailored for aligning and analyzing multi-session one-photon calcium imaging data. CaliAli leverages structural cues such as blood vessel patterns alongside neuronal features to compute robust inter-session alignments, which makes the method resilient to shifts and deformations that commonly occur in longitudinal experiments.
After alignment, CaliAli reconstructs a continuous video spanning multiple sessions. From this concatenated video, an optimized signal extraction algorithm automatically detects neuronal footprints and extracts calcium traces while removing redundant detections and suppressing noise. The integrated pipeline enhances sensitivity to weak signals and improves the reliability of neuron identification when populations overlap or when active neuron subsets change between sessions.
The authors validated CaliAli on hippocampal datasets and demonstrated improved spatial coding accuracy for CA1 neurons across sessions. In optogenetic tagging experiments, the tool increased the trackability of neurons in the dentate gyrus over time scales of weeks. Using a standard ultra-miniature microscope, the team tracked dentate gyrus neuronal populations continuously for up to 99 days, reporting stable population-level activity across that period.
The ability to follow the same neuronal ensembles over months opens new experimental possibilities. Researchers can now more reliably examine how specific neurons participate in memory encoding and retrieval, how representations evolve with learning, and how gradual pathological changes unfold in models of neurological disease. By enabling long-term, single-cell resolution tracking in freely behaving animals, CaliAli helps bridge short-term observations and long-term neurobiological processes.
Funding: This work was partially supported by the Japan Agency for Medical Research and Development (JP21zf0127005, JP23wm0525003), Japan Society for the Promotion of Science (JSPS) (24H00894, 23H02784, 22H00469, 16H06280, 20H03552, 21H05674, 21F21080), Takeda Science Foundation, Uehara Memorial Foundation, The Mitsubishi Foundation, and G-7 Scholarship Foundation to M.S.; JSPS (23K19393, 24K18212) to I.K.; and the Japan Science and Technology Agency (JPMJSP2124) to Y.W.
About this neurotech and neuroscience research news
Author: YAMASHINA Naoko
Source: University of Tsukuba
Contact: YAMASHINA Naoko – University of Tsukuba
Image: The image is credited to Neuroscience News
Original Research: Open access. “A comprehensive suite for extracting neuron signals across multiple sessions in one-photon calcium imaging” by SAKAGUCHI, Masanori et al. Nature Communications
Abstract
A comprehensive suite for extracting neuron signals across multiple sessions in one-photon calcium imaging
We developed CaliAli, a comprehensive suite designed to extract neuronal signals from one-photon calcium imaging data collected across multiple sessions in freely moving mice. CaliAli integrates information from blood vessels and neuronal features to correct inter-session misalignments, demonstrating robustness to non-rigid brain deformations and substantial changes in the field of view.
CaliAli also handles scenarios with high neuronal overlap and shifts in active neuron populations across sessions. By performing computationally efficient signal extraction on concatenated video sessions, the suite improves detection of weak calcium transients and increases the accuracy of extracted spatial coding patterns.
Evaluation in hippocampal datasets showed that CaliAli enhanced the spatial coding fidelity of CA1 neuron activity across sessions. Optogenetic tagging confirmed improved trackability of dentate gyrus neurons over weeks, and dentate gyrus populations tracked with CaliAli displayed stable activity over a 99-day period.
Overall, CaliAli advances our capacity to study the dynamics of neuronal ensembles over extended timescales, a critical step toward understanding the neural mechanisms that underlie natural behaviors, learning, memory, and the progression of neurological disorders.