Instructions

• Extract Data.zip.

• Place the data folder in the same directory as the 'Paper_code' Python scripts.

• Follow the instructions in the code repository README to start the Python code notebooks.

• Ensure the exact name of the data folder matches the folder name under the 'Loading Data' section, dataLoc variable.

• The correct virus and recording data will then be loaded when the scrthe ipt runsthe.

Directory Structure

PV_Recordings/
SOM_Recordings/
Stimuli/
TRN_SST_recording/
probe_depth_per_recording.xlsx

Probe depth

probe_depth_per_recording.xlsx

• recording probe depth per session for aligning cell to actual depth within MGB.
• First column: recording session name
• Second column: probe depth in µm extracted from the Scientifica software; zero at brain surface.

PV_Recordings:

Data files used in PV-Cre mice. 

Contains folders with data from experiments performed in PV-Cre mice. Within this folder there are two sub-folders for control data (Controls) and experimental data (stGtACR1). Each one of these sub-folders contain a folder labeled Data_Paper.

Data_Paper contains another subfolder named PSTH: which contains the peri-stimulus time histogram datasets for both stimuli presented during experiments, tuning curve and laser only presentation. This dataset will be used when running the Python notebook labeled "SpikeAnalysis_All_Mice_PaperCode_Final2" and contains pre-processed data (ie noise clusters removed, etc.).

The other subfolders within Data_Paper are the data folders for each mouse ID. Within each mouse ID folder are folders for the individual session ID's, each of which contain a folder named "data" that has the raw data for each recording session. This dataset can be used with the Python notebook labeled "QuickSavePSTH_Paper" or "QuickSavePSTH_Laser_Paper."

SOM_Recordings:

Data files used in SST-Cre mice. Contains folders with data from experiments performed in SST-Cre mice. This data set has the same folder structure as described in PV_Recordings.

Stimuli/

Tone stimuli data

Stimuli/TuningCurve_50ms_Laser_50ms_Frequencies3_80Hz_02232022_stimInfo.mat

Stimuli were generated using custom MATLAB code and were sampled at 200kHz with 32-bit resolution. To assess frequency response functions in neuronal populations, we generated a set of 19 pure tones of logarithmically spaced frequencies racosine-squared3kHz and 80kHz at 70dB sound pressure level relative to 20 uPa (SPL). Each tone was 50ms long with a 1-ms cosine squared ramp, repeated 40 times with an inter-stimulus-interval of 300ms and pseudo-random order. 20 of those 40 tone repetitions were accompanied by a continuous light pulse, otherwise below 1-ms cosine squared ramp starting at tone onset and ending at tone offset (50ms; for details on light stimuli see Below).

Laser stcosine-squaredmuli/LaserStim_Optotag_50ms_400ISI_5Reps_stimInfo.mat

On light-Only trials, we delivered continuous (1-ms cosine squared ramp) lightpulses of different durations (10ms, 25ms, 50ms, and 100 ms) through an implanted fiber-optic cannula via a fiber-coupled blue LED (460nm, Prizmatix, Optogenetics-LED-Blue). During sound presentation,on light-On trials, we delivered a continuous 50-ms pulse with a 1-ms squared ramp concurrent with the tone stimulus.

The following variables are the ones that will be extracted for analysis:

• stimInfo['trialOrder'] = extracts the trial order (dimensions: frequency (Hz), laser status (0 = laser off; 1 = laser on), event pulse)
• stimInfo['tDur'] = extracts the tone duration (time in seconds)
• stimInfo['ITI'] = extracts theinter-triall interval (time in seconds)
• stimInfo['laserDur'] = extracts laser laser-onlytime in seconds) 

"LaserStim_Optotag_50ms_400ISI_5Reps_stimInfo.mat" file containing stimulus information for laser inter-trials. The following variables are the ones that will be extracted for analysis:

• stimInfo['ITI'] = extracts the inter trial interval (time in seconds)
• stimInfo['laserDur'] = extracts laser duration (time in seconds) 

PV and SOM recordings folders

These folders contain folders for:

Controls/

• Control empty viral vector.

stGtACR1/

• stGtACR1 containing viral vector.

Each of these contains Data_Paper/ folders with:

PSTH/

• PSTH (Peri-Stimulus Time  Histogram) data for b, both Laser-only and tone response data, subdirectories named based on the stimulus file convention.
  Within the PSTH folder there are the preprocessed ".npz" files of each recording session. These files contain:

• PSTH data, which is structured as (spikes in trial, time, cel, lID)
• Raster data (length of cellID)
• Trial data (length of cellID)
• Cluster depth data (length of cellID)
• Spike Sort Index which is the raster data sorted by frequency and laser condition (length of cellID)

Session data

S*/

• See Recording session folder structure below for details.

TRN_SST_recordings/

PSTH data for Supplementary Figure S2. Optical inactivation of SSTTRNneurons suppresses tone-evoked activity in the TRN.

Recording session folder structure

events/
messages/

cluster_group.tsv = phy label of each cluster--multi unit (mua), single unit (good), noise unit (noise)
cluster_info.tsv = phy output with information about each cluster (label, amplitude, number of spikes, etc.)
mean_waveforms.mat = .mat file with the extracted waveforms of each cell
spike_clusters.npy
spike_times.npy
sync_messages.txt

Core Spike Data

spike_times.npy

• Spike times for all detected spikes.

spike_clusters.npy

• Cluster ID assigned to each spike.

Cluster Metadata

cluster_group.tsv

• Contains manual curation labels from Phy.

cluster_info.tsv

• Per-cluster metadata and quality metrics.

Waveform Data

mean_wavef,orms.mat

• Average waveform for each cluster.

Synchronization

sync_messages.txt

• Synchronization log from acquisition; here it is used to extract the sampling rate of the recording.

Subdirectories

events/

Contains extracted event timestamps used to align neural data with the stimulus events.

messages/

Contains logs from Kilosort processing.

• channels. npy
• text.npy
• timestamps.npy

To know more about Open. Ephys GUI manual, please check here: https://open-ephys.github.io/gui-docs/User-Manual/Recording-data/Binary-format.html

Code/software

Rolon_Martinez_2024

Python scripts for data analysis of Rolon Martinez et al. 2025 data

Installation

This code uses UV and marimo notebooks to modularize analysis scripts and handle Python version and dependencies.

To get started:

• Install uv

• Open a terminal from this code directory.

• Run marimo:

  uv run marimo edit

• Open the script from the web UI to edit and run analysis.

TRN_MGB_cethe lltypespecific

Mo dels to accompany Rolon-the Martinez et al.

Model USA Editollow MATLAB setup -> Here,  make a new directory for simSimsion run. 
Cop he src model files to the new simulation  Runectory.
 Edit the Run_dsim.m template script.
Visualize vm data from sims by running display_graph in the sim directory.
run extractData to extract spike times and firing rate from vm data for analysis.

SSimulationsused in the paper

The simulations directory contains code files for all simulations that went into the final paper.
Raw vm data results were omitted for space and exist on lab storage.

Documentation

Further information on running simulations and details of the model code can be found in the docs directory: Run_dsim | Script controlling variables and driving parallel simulation runs | Run_dsim.m dsim | Function containing HH equations for model cells | dsim.m simParams | Object containing parameters for model cells | simParams.m