Optimized PAR-2 RING dimerization mediates cooperative and selective membrane binding for robust cell polarity

Complete analysis code for the paper "Optimized PAR-2 RING dimerization mediates cooperative and selective membrane binding for robust cell polarity" (Bland et al., 2024; EMBO Journal)

Code structure

Notebooks used to perform the analysis in the paper and generate most of the figure panels are found in scripts. This is split into two sections:

• Run: notebooks used to perform quantification on raw images, and save results in the data folder (e.g. ph_quantify.ipynb outputs the data file ph_quantification.csv). Running these scripts requires access to the raw data repository (i.e. image files, see below).
• Analysis: notebooks used to perform downstream analysis on the quantification results from above and generate figure panels (which are saved in a series of Figs folders). Most figures can be created without access to the raw data repository, but some do require access to raw data (e.g. panels displaying specific images). Also includes SEC-MALS analysis and modelling.

Notebooks for creating specific figure panels:

Figure 1: D-E, F, G
Figure 2: A, B-C, F-G, H, I, J, K, L
Figure 3: B-D, F
Figure 4: A-B, C, D, E
Figure 5: A, C-D, E-F, G-H
Figure EV1: A-B, C-F
Figure EV3: B-C
Figure EV4: B-D
Figure EV5: A, C-D, F, G
Figure S1: D
Figure S2: B-F, G
Figure S3: A-D

Core code is found in src. Also relies heavily on the par-segmentation package.

Data availability

The entire dataset will be made publicly available (~7.5 GB).

The vast majority of the data is image data, found in the Imaging folder. This is organised into a series of folders representing different experiments, and subfolders representing experimental conditions (worm strain, date, RNAi, acquisition settings). Data for individual embryos are further separated into folders. Within each embryo folder you will find:

• Raw images (one for each channel including DIC)
• An autofluorescence-corrected image (af_corrected.tif), generated from the raw images using SAIBR
• A preliminary manual ROI (ROI_manual.txt) generated using the matplotlib-polyroi package
• An optimised ROI (ROI_fit.txt) generated using the par-segmentation package
• An nd file containing metadata

Also includes the following datasets:

• AlphaFold: a PDB file for the PAR-2 RING domain dimer
• Sequence alignments: FASTA and Jalview files for the RING domain sequence alignments
• SEC MALS: Raw SEC-MALS data for the PAR-2 RING domain

Installation

This project requires Python 3.11.6. Package requirements are listed in requirements.txt (or requirements-windows.txt if using Windows).

To ensure reproducibility, it is best to run the project in a Docker container. To do this, perform the following steps:

⁣1. Make sure Docker is installed and open on your machine

⁣2. Download/clone the code and navigate to the project directory

⁣3. Build the Docker image (~ 3 GB)

docker-compose build

⁣4. Once the build has completed, run the Docker container

docker-compose run -p 8888:8888 app

⁣5. Open Jupyter

jupyter notebook --ip 0.0.0.0 --no-browser --allow-root

This will print a couple of URLs at the bottom of your terminal. Copy and paste the second of these into your web browser to open up Jupyter.

Notes:

• If cloning the repository from GitHub, use the --depth 1 option to save disk space (75 MB vs 2.5 GB). i.e. git clone --depth 1 https://github.com/goehringlab/2024-Bland-EMBO.git
• If you're using the raw data, you must either place it inside the project directory as a folder named raw_data, or change the path in docker-compose.yaml where indicated. Otherwise, you should delete this line.
• To delete the Docker image once you've finished, run docker image rm -f 2024-bland-embo-app

License

This work is licensed under a Creative Commons Attribution 4.0 International License.