Published July 12, 2026 | Version v1

Data and code for "Loss of Paneth cell contact starts a WNT differentiation timer in intestinal crypts"

  • 1. Universitair Medisch Centrum Utrecht
  • 2. ROR icon Hubrecht Institute for Developmental Biology and Stem Cell Research
  • 3. ROR icon Institute for Atomic and Molecular Physics
  • 4. ROR icon Delft University of Technology
  • 5. AMOLF

Description

This dataset contains the tracking data and analysis scripts used for the publication "Loss of Paneth cell contact starts a WNT differentiation timer in intestinal crypts".

System requirements

No special system requirements are necessary, a standard work laptop/desktop is sufficient.

Running the scripts

The analysis scripts have been tested against OrganoidTracker v2.1 on Windows 11. It should also run on other operating systems.

First, install OrganoidTracker ( https://github.com/jvzonlab/OrganoidTracker/releases/tag/v2.1.0 ). Then, configure your Python environment such that OrganoidTracker can be found. One way to do this is to open OrganoidTracker in PyCharm (File -> Open...), then open the Analysis Scripts folder in PyCharm too, attaching it to the OrganoidTracker project. Then, in File -> Settings... -> Project: OrganoidTracker -> Project Dependencies, make sure that the Analysis Scripts depend on OrganoidTracker (and not the other way around). Another way to do this is to set the `PYTHONPATH` environment variable to include the OrganoidTracker folder. The total installation time depends on the users' familiarity with the command line, Miniforge, etc., but should be under 15 minutes in the optimistic case.

Then, each script can be run from the Analysis Scripts folder. If you're using PyCharm, you can simply press the Run button (Play icon), or otherwise you can use a command like `python figure_1b_axin2_by_paneth_distance.py`. The scripts should normally run in under a minute. Assuming you kep the relative directory structure intact, the scripts should pop up a Matplotlib GUI at the end, recreating the display item from the paper. You can then save this graph as a PDF, PNG, etc. using the save button.

Running the figure_s6_rna_sequencing.Rmd script

One exception is the script `figure_s6_rna_sequencing.Rmd`, which was written in R, and requires the following packages to be installed:

  • Seurat 5
  • SeuratObject
  • dplyr
  • patchwork
  • ggplot2


Inspecting the cell tracking data

You can run OrganoidTracker to inspect the cell tracking data. Due to space reasons, we could not include all images, but a selection (organoids grown under homeostatic conditions, converted to compressed 8bit TIFF files) is available in the Images folder. Other images (Wnt pertubation, FRAP, Paneth cell ablation, in vivo crypts, Palbociclib treatment) as well as the original Leica Image Files (LIF files) are available upon request to the corresponding authors of the paper.

To load the images, simply look in the Images folder, and drag the first image of a folder into the OrganoidTracker GUI. The corresponding tracking file can be found in `Tracked data/Axin2 and staining`.

Format of the tracking files (AUT files)

AUT files are simply JSON files, and the format is mostly self-explanatory. However, for storing the position and links two varations exist.

There is the older "v1" format, in use from 2018-2024. You can recognize these files by the "version": "v1" entry in the root of the file. It stored positions and links, plus their metadata, in a kind of cumbersome and duplicative way:


// Old format
"positions": {
 "1": [[56.4, 54.1, 4], ...],
 ...
},
"links": { // Links i nthe d3js graph format
    "directed": false,
    "multigraph": false,
    "graph": {},
    "nodes": [
      // All positions again, so that we follow the d3js format
      // Difference with above is that we now also list position metadata
      // Note how "_time_point_number" and the metadata keys are repeated over and over,
      // further bloating file size
      {"id": {"x": 136.92842303555807, "y": 186.91887053353733, "z": 12.0,
        "_time_point_number": 1, "blah": ...}},
      {"id": {"x": 157.02603447837174, "y": 184.34746761541615, "z": 12.0,
        "_time_point_number": 1, "blah": ...}}
      ...
    ],
    "links": [
       // All individual links
       // The save format doesn't know about tracks, so those are reconstructed when loading tracking data
       {
         "source": {"x": 132.0952578173535, "y": 249.18141069746628, "z": 12.0, "_time_point_number": 1},
         "target": {"x": 125.84057341732412, "y": 253.73027207930585, "z": 12.0, "_time_point_number": 2},
         "blah": ...  // Link metadata
       },
       ...
    ]
}

There is also the newer "v2" format, in use from 2025 onwards. You can recognize these files by the "version": "v2" entry in the root of the file. It directly stores tracks, and also stores metadata in a more efficient way, without repeating the metadata keys over and over:

// New format
"positions": [
  { // Holds all positions and metadata for a single time point
    "time_point": 0,
    "coords_xyz_px": [[56.4, 54.1, 4], ...],
    "position_meta": { // Position meta now stored among the positions (instead of in the links graph)
       "blah": [...]  // One entry per position, null if not defined
    }
  },
  ...
]
"tracks": [
  { // Holds all positions and metadata for a single track
    // Coords of mother cell, in time point 17. Usually one entry, multiple in case of cell merges
    "coords_xyz_px_before": [[52.4, 51.1, 3]],  
    "time_point_start": 18,  // Time point of first position of track, 18 in this case
    // Coords of all positions, first for time point 18, then time point 19, etc.
    "coords_xyz_px": [[56.4, 54.1, 4], [58.4, 55.1, 4], ...],  
    "link_meta": { // Link metadata
       "blah": [...],
       ...
    },
    "link_meta_before": { // Link metadata for the links connecting this track to the previous track(s)
       "blah": [...],
       ...
    },
    
    // Metadata for the entire lineage (so also daughter tracks), only defined for tracks without a previous track
    "lineage_meta": { 
        "blah": ...  
    }
  },
  ...
],

Files

KokZhengWntDifferentationTimer.zip

Files (5.0 GB)

Name Size
md5:b29e01498c4b877b02ce68d331574951
5.0 GB Preview Download

Additional details

Software

Programming language
Python , R