MATLAB

MATLAB scripts currently provide the main preprocessing path for WILD recordings.

Scripts

Script	Purpose
WILD_PreProcess.m	Generate corrected headers, timing files, and event outputs.
WILD_PreProcessFolder.m	Process multiple recordings.
WILD_ReadHeader.m	Read WILD parameter headers.
WILD_processIMU.m	Process IMU data.
WILD_scaleIMU.m	Scale IMU signals.
WILD_genIntanHeader.m	Generate Intan-compatible header output.
WILD_PreProcess_Multi.m	Preprocess multiple synchronized devices and write a merged sorter-ready binary.

Minimal Workflow

cd("C:\code\github\Neurologger\Code");
recording_path = "C:\WILD\example_recording";
WILD_PreProcess(fullfile(recording_path, "amplifier.dat"));
WILD_processIMU(fullfile(recording_path, "analogin.dat"), 100);

Adjust paths and sampling assumptions for your recording mode.

Expected Folder Layout

Input

example_recording/
  amplifier.dat
  analogin.dat
  CE_params.bin
  adc.dat           # optional
  misc.dat          # optional

Outputs generated by the public MATLAB path

example_recording/
  time.dat
  info.rhd
  IMU.mat
  device_event.d01.evt   # if digital events are present
  device_event.s01.evt   # if stimulation-related event files are generated

Known-Good First Pass

1. Run WILD_PreProcess on amplifier.dat.
2. Confirm that time.dat and info.rhd appear.
3. Run WILD_processIMU on analogin.dat.
4. Confirm that IMU.mat is generated.
5. Keep the raw export unchanged and write downstream results into a separate analysis copy when reprocessing.

Multi-Device Workflow

Use WILD_PreProcess_Multi.m when multiple WILD logger exports should be merged before spike sorting. The current default path detects the device folders below a selected subepoch folder and calls WILD_channelMerger.m, which aligns two logger recordings from the neural common-mode signal rather than requiring generated .evt sync files.

addpath("C:\code\github\Neurologger\Code");

% Opens a folder-selection GUI for multiple WILD devices.
result = WILD_PreProcess_Multi();

For scripted processing, pass the folders explicitly:

addpath("C:\code\github\Neurologger\Code");

parent_folder = "D:\WILD\experiment01";
result = WILD_PreProcess_Multi(parent_folder, ...
    "MasterIndex", 1, ...
    "OutputFolder", "D:\WILD\experiment01\merged");

For the standard multi-device animal folder layout, select the subepochName directory:

RatID/
  sessionName/
    subepochName/       <-- select this folder
      Device name 1/
        datetime/
          amplifier.dat
          analogin.dat
          CE_params.bin
      Device name 2/
        datetime/
          amplifier.dat
          analogin.dat
          CE_params.bin

WILD_PreProcess_Multi recursively detects child folders that contain both amplifier.dat and CE_params.bin. The generated channel layout records the device folder name and recording datetime folder for each merged channel. You can also pass device folders directly:

device_folders = {
    "D:\WILD\experiment01\master"
    "D:\WILD\experiment01\slave01"
};

result = WILD_PreProcess_Multi(device_folders, ...
    "MasterIndex", 1, ...
    "OutputFolder", "D:\WILD\experiment01\merged", ...
    "MergeMethod", "channelMerger", ...
    "Overwrite", true);

Expected merged outputs:

first_device_recording_folder/
  amplifier_merged.dat
  analogin_merged.dat
  amplifier_merged_mergeInfo.mat

selected_output_folder/
  multi_device_channel_merger_result.mat
  multi_device_channel_layout.tsv

WILD_channelMerger.m currently expects exactly two logger recordings. It writes the merged binary next to the first detected device recording, while WILD_PreProcess_Multi.m also writes summary metadata into the selected output folder.