Dataset DOI: 10.5061/dryad.612jm64k7

Description of the data and file structure

This dataset accompanies the article “Chemotaxis-driven instabilities govern size, shape and migration efficiency of multicellular clusters” and contains experimental and simulation data quantifying collective chemotactic migration, shape deformation, orientation, and breakup of multicellular clusters. Experimental data were obtained from time-lapse microscopy of chronic lymphocytic leukemia (CLL) cell clusters exposed to controlled CCL19 gradients (0–100 ng ml⁻¹ mm⁻¹). Processed measurements extracted using Fiji/ImageJ include cluster size, centroid position, shape metrics, and the orientation of the major axis relative to the chemical gradient, reported in pixel units. Simulation data consist of processed outputs from agent-based models spanning multiple cluster sizes and gradient strengths, including forward migration index, entropy, correlations, shape metrics, rotation, and breaking frequencies. Together, these data enable analysis of chemotaxis-driven instabilities and provide reusable benchmarks for studies of collective migration and cluster dynamics. The dataset contains no personal or identifiable information and has no ethical or legal restrictions beyond citation of the associated publication.

Files and variables

File: DataForPaper.zip

Description: 

Folder Structure

The deposited dataset contains two main folders:

1. ExperimentalData/

Processed measurements extracted from time-lapse microscopy of CLL cell clusters migrating under CCL19 gradients. These experimental data are used to calculate the orientation of the cluster’s major axis relative to the direction of the chemical gradient, as a function of cluster size.

Contents:

• AngleAspect_Orientation_of_Cluster/expt.csv

  Processed breaking probability for a representative experiment.

• G0-100Angle/combined_angles.csv

  Combined angle data aggregated across eight experiments (9Apr2014, 14Feb2014, 17Oct2013, 22Dec2014, 23Oct2013 Nikon Gradient Slopes, 26Nov2013 Nikon, Position4_9Dec2013).

• Subfolders for individual experiments, each containing processed CSV along with the actual experimental file:

  • 9April2014/
  • 14Feb2014/
  • 17Oct2013/
  • 22dec2014/
  • 23ct2013NIKON Gradient Slopes/
  • 26nov2013NIKON/
  • Position4_9Dec2013/

Each folder may contain the following files:

• *.tif — original image file
• *(0–100).tif — cropped experimental region (0–100), if available
• Results(0–100).csv* — measurement results for the cropped region, if generated

The CSV file contains measurements exported from Fiji/ImageJ (Analyze → Measure) for segmented intensity clusters.

Units

• Lengths & positions: pixels (px)
• Area: square pixels (px²)
• Intensity: raw grayscale values (arbitrary units)

Columns

• Area (px²): Cluster size
• Mean / Min / Max: Pixel intensity statistics
• XM, YM (px): Cluster centroid coordinates
• Major / Minor (px): Ellipse major & minor axes
• Angle (deg): Orientation of major axis (from x-axis)
• Circ.: Circularity (1 = perfect circle)
• Slice: Image slice index
• AR: Aspect ratio (Major/Minor)
• Round: Roundness (based on major axis)
• Solidity: Area/convex area (1 = solid)

2. SimulationsData/

Processed summary outputs extracted from simulations of cluster migration under varying gradients and cluster sizes.

Files include:

• averaged_data_angle_with_cluster_fraction_size.csv
• BreakingFrequency_size-grad_gdep_250.txt
• ClusterQuantification_size-grad_gdep_250.txt
• Entropy_con-gradHeatmap.txt
• entropy_contours_all_levels_plus_extra.csv
• entropy_contours_all_levels.csv
• entropy_contours_thresholds.csv
• Entropy_size-gradHeatmap.txt
• FMI_con-grad_gdep_37_Heatmap.txt
• FMI_size-grad_gdep_250_Heatmap.txt
• full_data_angle_with_cluster_fraction_size.csv
• ROTation_size_grad_gdep_250.txt
• ROTationdata.txt

These represent the processed results used to generate the figures in the paper, including FMI, entropy, correlations, shape metrics, and breaking frequencies.

A. TimeSeriesData/XXX/RUN_0001/ to RUN_0010/

TimeSeriesData/N_XX_pcore_XX_noff_XX_grad_X.XXXX_gdep_250/RUN_0001/ to RUN_0010/

In the folder name:

• N_XX: size of the cluster
• pcore_XX: active propulsion strength of the core particles
• noff_XX: noise offset
• grad_X.XXXX: chemical gradient strength
• gdep_250: concentration at the center of mass (CM) of the cluster

Each RUN_XXXX folder contains:

data/

Reduced simulation frames or processed particle positions.

• Ovito.xyz

  Main data file containing time-resolved particle information (load using the OVITO application).

  Columns:

  ParticleTypeID, ClusterID, x, y, r, vx, vy, nx, ny, gx, gy

• velocityvst.txt

  Time-resolved center-of-mass data:

  time, x_cm, y_cm, v_x, v_y

analysis/

Summary outputs including cluster size, velocities, shape evolution, breaking probability, and orientation over time.

• CMFrame.xyz

  Ovito.xyz with particle positions shifted to the cluster center-of-mass (CM) frame.

• ZeroToBoxLx.xyz

  Ovito.xyz shifted to a 0 → L box reference (from −L/2 → L/2).

• colorCodedOvito.xyz

  Ovito.xyz with particles color-coded by cluster ID.

• CM_with_layers.xyz

  Cluster in the CM frame, color-coded into three layers, used to calculate mixing entropy.

• clusterNumber_vs_time.txt

  Number of clusters and particles per cluster at each frame.

• entropy_time_data.txt

  Mixing entropy as a function of time.

• clusterRotationData.txt

  Cluster dynamical properties (rotation and motion) as a function of time.

• output.txt

  Time-resolved shape metrics of the cluster:

  time, major_axis_length, minor_axis_length, angle_with_x

• ROG.txt

  Radius of gyration and shape tensor–based quantities:

  t, xx, xy, yy, radiusOfGyration, asymmetry, compactness, principal_dir_x, principal_dir_y

Data Variables

Experimental Data Variables

• Angle – Cluster orientation relative to the +x cdirection (degrees).
• Frame / Time – Time index of the movie (frames).
• x, y – Cluster centroid position (pixels or µm).
• Size – Cluster size or projected area (cell count or pixels).

All files include column names as descriptions that explain the variables contained in each dataset.

Simulation Data Variables

• FMI – Forward migration index (dimensionless).
• Entropy – Mixing entropy of cluster layers (dimensionless).
• AspectRatio – Ratio of major/minor cluster axes.
• BreakFrequency – Number of breaking events.
• CorrelationCrossing – Distance where velocity correlation crosses zero (cell diameters).
• Gradient (g_r) – Chemical gradient strength (model units).
• Velocity – Cell or cluster speed (model units).
• Position_x, Position_y – Position coordinates (model units).

Missing Values

Missing values are indicated as blank cells or NaN. No special placeholders are used.

Files and Data Notes

Raw simulation trajectories and raw video files are not deposited due to size limits, but are available upon request.

Code/software

1. Experimental videos were opened in Fiji, where clusters were segmented, and measurements (angle, size, position) were extracted.
2. Simulation trajectories (*.xyz files) were visualized in OVITO .
3. Processed CSV/TXT files were analyzed in Python, where all quantitative analysis and plotting were performed using the provided scripts.