cFos SWR Reactivation

Overview

This repository contains code associated with the study "Inhibitory and excitatory cFos engram neurons are preferentially reactivated by sharp wave ripples" by Javed, Robles-Hernandez et al. from the Memory Circuits Lab. Here we investigated how cFos-tagged hippocampal CA1 neurons are recruited during sharp wave ripples.

In this project, we combined cFos-dependent tagging, chronic in vivo electrophysiology, local field potential recordings, optotagging, spatial coding analyses, interneuron characterization, and sharp wave ripple analyses to ask whether experience-associated cFos-tagged neurons are preferentially recruited during post-behavior rest.

The study focuses on both excitatory and inhibitory components of a cFos-tagged spatial ensemble, including cFos-tagged place cells, non-place principal cells, and interneurons.

System requirements

The source code is written in MATLAB. It was tested with a MATLAB 2023 release.

• MATLAB: MATLAB R2023 (tested).
• MATLAB toolboxes: Signal Processing Toolbox and Statistics and Machine Learning Toolbox.
• Third-party MATLAB dependencies: MClust/Neuralynx readers and CellExplorer (for CCG classification).
• Operating system: The code is intended for standard desktop operating systems supported by MATLAB.
• Hardware: No non-standard hardware is required to run the analyses. Memory and storage requirements depend on the size of the electrophysiology recordings.

Installation

1. Download or clone this repository.
2. Install MATLAB R2023 and the required toolboxes.
3. Install the third-party dependencies listed above and record their local paths in Matlab_code/classification/classification_config.json. At minimum, configure cellExplorerPath and mclustPath; pass the detectHFOs and findpeaksmine folders through the additionalPaths setting when running oscillation analyses.
4. Start MATLAB and add the repository code to the MATLAB path:

addpath(genpath('/path/to/cfos-swr-reactivation/Matlab_code'))

The MATLAB-path setup takes less than one minute. Once MATLAB and the third-party dependencies are already installed, configuration normally takes less than 10 minutes; installing external dependencies may take longer.

Demo data

The demo_data folder documents how to run the packaged SWR cell-metrics example. It calculates per-cell SWR participation, firing-rate increase, and spikes-per-ripple fields from a small real sleep-session subset with precomputed SWR events.

Running the analyses

The electrophysiology analyses use a sessionInfo.mat file containing sessInfo and an All_Cells_combined.mat file. Pass both locations explicitly when they are not in the repository root or Data folder:

settings = struct( ...
    'sessionInfoPath', '/path/to/sessionInfo.mat', ...
    'allCellsPath', '/path/to/All_Cells_combined.mat');

Sharp-wave ripple and phase analyses

Run SWR detection first, followed by the desired cell-level analyses:

detect_swr_events(settings)
compute_swr_cell_metrics(settings)
compute_swr_phase_analysis(settings)

For open-field theta-phase encoding, run:

compute_open_field_theta_phase_encoding( ...
    'sessionInfoPath', settings.sessionInfoPath, ...
    'allCellsPath', settings.allCellsPath)

See Matlab_code/oscillations/README.md for required session files, output filenames, and optional settings.

Cell classification and place-cell metrics

Compute the GMM features before running the GMM classifier:

compute_gmm_cell_properties(settings)
GMM_based_classifications(settings)
Classifications_Spatial_Info(settings)
compute_place_cell_reuse_metrics(settings)

The GMM classifier requires curated CCG ground-truth labels. To generate and review these labels, run CCG_based_classifier_modified(settings), curate the generated review sessions with review_CCG_connection_labels, and then run rebuild_All_Cells_combined_CCG_classification_from_curated. The complete sequence is described in Matlab_code/classification/README.md.

Optotagging and CSV exports

Run the SALT optotagging analysis before exporting its summaries:

SALT_opto_analysis(settings)
CSV_file_export_first(settings)

Image quantification

Provide the path to the quantification workbook and an output folder:

plot_CB_quantification_figures( ...
    '/path/to/Quantifications.xlsx', ...
    'OutputFolder', '/path/to/output/CB')

plot_interneuron_quantification_figures( ...
    '/path/to/Quantifications.xlsx', ...
    'OutputFolder', '/path/to/output/interneurons')

The figure functions write figures and summary tables to their specified output folders. The cell-property workflow and its input expectations are documented in Matlab_code/cell_properties/README.md.

Preprint

bioRxiv: https://www.biorxiv.org/content/10.1101/2024.12.17.628897

Lab

This work was carried out in the Memory Circuits Lab.

At the German Center for Neurodegenerative Diseases (DZNE)
and
Charité - Universitätsmedizin Berlin
Berlin, Germany

Contact

For questions about the code or analyses, please contact:

Silvia Viana da Silva
Memory Circuits Lab
German Center for Neurodegenerative Diseases (DZNE)
Berlin, Germany

or
Matthias Haberl
Charité - Universitätsmedizin Berlin
Berlin, Germany

License

Please see the repository license for terms of use.