Critical_reorganization_of_adolescent_prefrontal_circuitry/README.md

Reorganization of adolescent prefrontal cortex circuitry is required for mouse cognitive maturation

This repository contains the ephys data and custom code related to the article "Reorganization of adolescent prefrontal cortex circuitry is required for mouse cognitive maturation". We provide in vivo recordings from L2/3 of the prefrontal cortex of 26 chronically recorded head-fixed mice (LFP, n=195 recordings; SUA, n=7230 units) and 4 chronically recorded freely-moving mice (LFP, n=130). Following the structure of the article, the dataset is structured according to the mouse ID. An overview of recordings can be found in the table "Recording summary of chronically recorded mice".

Head-fixed recordings:

Each mouse ID folder contains data of analyzed LFP periods (broadband 0.1-9000 Hz, LFP.dat file), timestamps of analyzed LFP data (Timestamps_LFP_Baseline.mat file), movement of the mouse (Movement_Baseline.mat file), and sorted single unit information and timestamps (SUA.mat file). Note, for LFP-SUA analysis single unit timestamps corresponding to the LFP timestamps have to be used. Head-fixed LFP data have a sampling rate of 32000 Hz and movement is a logical vector with a sampling rate of 1 Hz. Spike sorting has been done using klusta (https://klusta.readthedocs.io/en/latest/), and manually curated with Phy (https://phy.readthedocs.io/en/latest/). The SUA.mat structure provides following information for every sorted unit:

• ClusterID: ID of the unit
• Channel: The channel on which the unit has been detected
• Timestamps: Timestamps of detected action potential with a sampling frequency of 32000 Hz
• Waveform: the average waveform of the unit

Freely moving recordings:

Each mouse ID folder contains data of analyzed LFP periods (broadband 0.1-9000 Hz, LFP.dat file) and movement of the mouse (Movement_Baseline.mat file. Freely-moving LFP data have a sampling rate of 32552 Hz and movement is a vector of the distance moved (in pixel) with a sampling rate of 201 Hz. No spike sorting was performed for freely-moving data.

Electrode channels:

All recordings were recorded with a 16 channel Neuronexus 1x16 1-shank electrode with a channel interspace of 100 µm. Channel 1 is the most superficial (closer to brain surface), and 16 the deepest. The position of electrodes has been post-mortem histologycally reconstructed.

Custom code:

• statistics: The folder contains Jupyter Notebooks with the implementation of restricted (natural) cubic spline model fits, detecting extrema of obtained fits, and visualization of the data together with the models. Each file corresponds to one figure and can be run locally and in google colaboratory environment.

• morphology: The folder contains all subfunctions for quantifying overlays with Matlab in confocal stacks with 2 or 3 channels. The “MasterPruning” calls all subfunctions and provides explanation.

• LFP: The folder contains Matlab scripts used for LFP analysis. The measures “gPDC” and “PAC” have subfolders to separate their subfunctions. The main function locates outside the subfunction folder.

• SUA: The folder contains Matlab scripts used for SUA analysis.

All Matlab code was written in MATLAB R2021a. However, we have proven that the code also runs on previous versions up to MATLAB R2017a.

For every question and/or further data request, please don't hesitate in contacting me (jastyn.poepplau@zmnh.uni-hamburg.de)