camps

Analysis software for the CAMP cell-motility study.

Cell detection and tracking from timelapse movies, per-cell mean squared displacement (MSD), and the code that produced the published figures.

Raw movies and spot tables are archived separately in a Zenodo repository. The derived tables the figure code needs are included here, so the figures rebuild from a fresh clone.

Code behind each figure

Paper element	Code
Detection and tracking	pipeline/stage1_trackmate.py
Per-cell MSD	pipeline/stage2_compute_msd.py
Figure 4b	figures/code/build_fig4b.py
Figure 4c	figures/code/build_fig4c.py
Figure 6b	figures/code/build_fig6b.py
Supplementary Figure 8	figures/code/build_figS8.py
Supplementary Figure 9	figures/code/build_figS9.py
Supplementary Figure 10	figures/code/build_figS10.py
Supplementary Table S2	figures/panel_data/tableS2_significance.csv

Pipeline

Requires Fiji with TrackMate, and Python 3 with numpy and pandas.

cd pipeline
make PROFILE=phase     # phase contrast: density and PMA experiments
make PROFILE=gfp       # GFP: doxycycline titration
make print             # show all settings for the selected profile

The two imaging channels use different detector and tracker settings, so a profile must be selected; make print shows the resolved values. If Fiji is not on the default search path, pass FIJI=/path/to/fiji, or run make install-fiji.

Figures

conda env create -f environment.yml
conda activate c04-panels
cd figures/code
PYTHONPATH=. python build_fig4b.py

Rebuilds are written to figures/rebuilt/; the published renders are in figures/output/. Each builder prints a text-overlap count, where zero is the pass condition.

build_fig4b.py and build_fig4c.py refit the significance model from the per-well data rather than reading stored values, so running them also reproduces figures/panel_data/tableS2_significance.csv.

si_confound_checks.py recomputes the values quoted in the Supplementary Figure 9 and 10 captions and exits non-zero if any of them fail to reproduce.

Figures in the paper were assembled and typeset from these renders, so a rebuild reproduces the data, statistics and panel geometry but not the final typography.

Layout

pipeline/     movies to tracks to per-cell MSD
figures/
  code/       figure builders and shared plotting code
  data/       derived inputs used by the builders
  panel_data/ per-panel values and the published tables
  output/     the published figure renders
docs/         software versions used for the published analysis

Notes

Significance uses a pooled-error model: one within-condition variance estimated from all eight wells of a subplot, giving four residual degrees of freedom, with Benjamini-Hochberg correction within each panel. An earlier Welch-based analysis is retained at figures/panel_data/supporting/ for comparison only.

Pixel size and MSD lag differ by channel: 0.68626 um/px and 3 frames for phase contrast, 1.02939 um/px and 5 frames for GFP.

PMA doses are in ng/mL.

Citation and archived versions

Each release is archived on Zenodo. Cite the DOI for the specific version you used; the concept DOI always resolves to the most recent release.

	DOI
v1.0.0 (published with the study)	10.5281/zenodo.22089115
All versions (concept)	10.5281/zenodo.22089114

The supporting data are archived separately at 10.5281/zenodo.22073097.

See CITATION.cff for author and citation metadata.

License

MIT. See LICENSE.