https://doi.org/10.5061/dryad.ht76hdrwn

Date of Data Collection: 2025-12-01

Related Repository: https://doi.org/10.5281/zenodo.18395537

Contributors

• Esben Schroeder†: Department of Human Physiology, University of Oregon, Eugene, OR 97403
• Helia G. Megowan†: Department of Human Physiology, University of Oregon, Eugene, OR 97403
• Madeline Luu†: Department of Computer Science & Department of Data Science, University of Oregon, Eugene, OR 97403
• Adam Shuaib: Department of Neuroscience, University of Oregon, Eugene, OR 97403
• Adam Fries: Genomics & Cell Characterization Core Facility, University of Oregon, Eugene, OR 97403
• Jake Searcy: Department of Data Science, University of Oregon, Eugene, OR 97403
• Hans C. Dreyer (Corresponding Author): Department of Human Physiology, University of Oregon, Eugene, OR 97403. Email: hcdreyer@uoregon.edu

† Co-first authors

Overview

This dataset serves as a comprehensive collection of example input data designed for the Single-Fiber Myonuclear Analysis Pipeline. It is intended to facilitate the reproduction of results, testing of pipeline functionality, and validation of the algorithms described in our associated research paper.

The data provided here represents the two primary stages of data acquisition required before running the automated Python quantification:

1. Raw Microscopy Data: Compressed archive containing the original .lif project file from the Leica microscope.
2. Processed Macro Output: The intermediate image files (skeletons, projections, and stacks) generated by the FIJI/ImageJ segmentation macro, which serve as the direct input for the Python analysis script (SF_analysis_pipeline.py).

By providing these files, we allow users to test the full workflow—from raw image extraction to final morphological quantification. The custom ImageJ macro code required to process the raw .lif data, as well as the Python analysis script (SF_analysis_pipeline.py) and its associated documentation, are available in the linked GitHub repository.

Software & Compatibility

To utilize this data effectively, the following software environment is recommended:

• Primary Python Pipeline: Single-Fiber Myonuclear Analysis Pipeline (v1.0).
• Image Processing: FIJI / ImageJ. (The .ijm macro script and instructions for running it are provided in the GitHub repository listed above).
• Python Dependencies: Python 3.10+, numpy, pandas, scikit-image, scikit-learn, scipy, openpyxl, Pillow.

Available at .

Dataset Structure

The dataset consists of two primary compressed archives:

1. Raw Microscopy Data (example_LIF_microscopy_project.zip)

File Format: .zip (Compressed Leica Image File)

Description:
This archive contains the original, multi-channel 3D confocal microscopy data (23_Wu_20_D14_R_02.lif) acquired directly from the instrument. It includes the Z-stacks of single muscle fibers before any segmentation or processing has occurred.

Usage:
Unzip this file to access the .lif project. This file is the starting point for the FIJI/ImageJ macro. Users wishing to test the segmentation portion of the workflow should download the macro from the GitHub repository and follow the instructions provided there to process this file.

2. Example Python Script Input (example_IJMacro_output.zip)

File Format: .zip (Compressed Directory)

Description:
This compressed archive contains the output of the FIJI/ImageJ macro. These files are structured specifically to serve as the input for the Python analysis pipeline. The directory structure mimics the exact hierarchy required by the SF_analysis_pipeline.py script.

Directory Hierarchy:
When unzipped, the folder structure is organized by Subject, Timepoint, Side, and Image Type:

example_IJmacro_output/
└── 20/                     <-- Subject ID
    ├── BL/                 <-- Timepoint (Baseline)
    │   └── L/              <-- Side (Left Leg)
    │       ├── STDIP/      <-- Standard Deviation Z-Projections
    │       ├── Skel/       <-- Binary Skeletons of the Fiber
    │       └── TIFs/       <-- Raw Z-Stacks (Converted from .lif)
    └── D14/                <-- Timepoint (Day 14)
        └── R/              <-- Side (Right Leg)
            ├── STDIP/
            ├── Skel/
            └── TIFs/
...

File Contents within Subfolders:
For each fiber analyzed, three corresponding files are present across the STDIP, Skel, and TIFs folders. The filenames share a common identifier to link them during processing.

1. STDIP Folder (.tif):

• Example: STDIP_23_Wu_20_D14_R_02.lif_-_Fiber1-1_Merged.tif
• Content: A standard deviation intensity projection used for nuclear segmentation masks.

2. Skel Folder (.tif):

• Example: Skel_23_Wu_20_D14_R_02.lif_-_Fiber1-1_Merged.tif
• Content: A binary skeleton image representing the centerline of the muscle fiber, used for orientation and distance calculations.

3. TIFs Folder (.tif):

• Example: 23_Wu_20_D14_R_02.lif_-_Fiber1-1_Merged.tif
• Content: The raw 3D Z-stack of the fiber, used for Z-position inference and intensity analysis.

Usage and Accessibility

These files are designed to be fully compatible with the open-source release of the Single-Fiber Myonuclear Analysis Pipeline.

Instructions for Use:

Option A: Run the Python Pipeline Only (Quick Start)
Use the pre-processed data to test the Python analysis script immediately.

1. Download and unzip example_IJMacro_output.zip.
2. Clone the GitHub repository: git clone https://github.com/DreyerLabUO/23Wu-SingleFiber.
3. Install dependencies as listed in the repository README.md.
4. Run the Python pipeline pointing to the unzipped directory:

python SF_analysis_pipeline.py "path/to/example_IJmacro_output" \
    --pixel_size_xy_um 0.329 \
    --z_scale_um_per_index 2.7

Option B: Full Workflow Reproduction
Start from the raw microscopy data, run the segmentation macro, and then analyze the results.

1. Download and unzip example_LIF_microscopy_project.zip to retrieve the .lif file.
2. Download/Clone the GitHub repository to access the ImageJ macro file 23_Wu-SF_Analysis.ijm and the Python script.
3. Run ImageJ Macro: Open FIJI, load the macro from the repository, and execute it on the extracted .lif file. This will generate a new output directory structured identically to example_IJmacro_output.
4. Run Python Pipeline: Execute the Python script, pointing it to the newly generated output folder from the previous step:

python SF_analysis_pipeline.py "path/to/YOUR_MACRO_OUTPUT_FOLDER" \
    --pixel_size_xy_um 0.329 \
    --z_scale_um_per_index 2.7

For detailed troubleshooting, parameter adjustments, or macro installation guides, please refer to the docs/ folder in the available repository.

Human subjects data

Our research group received explicit consent from participants to publish de-identified data publicly. This dataset has been de-identified in accordance with HIPAA standards by removing all 18 direct identifiers and specific quasi-identifiers, ensuring no reasonable basis exists to re-identify any individual.