Subcellualar spatialproteomic screen of SARS-CoV-2 infection

This repository contains the code used the analysis of spatial proteomics screen in the manuscript "Subcellular mapping of the protein landscape of SARS-CoV-2 infected cells for target-centric drug repurposing".

Steps to preprocess plates for annotation-tool:

Step 1. Convert file names and max-projection

python generate_covid_dataset-max-projection.py

Paths need to be updated:

• ACQUIRED_DATA_PATH: Raw data acquisition path with Images folder (input folder)
• FORMATTED_DATA_PATH: Images with formatted names (intermediate folder)
• IMG_FOLDER: Images with max-projection (output folder)
• ANNOTATION_FOLDER: Folder for annotation-tool with Imjoy (output folder, only used when manual annotation is later needed)

Step 2. Segmentation with HPACellSegmentator and convert mask to geojson file

python generate_covid_dataset-max-projection_segmentation.py
python mask_to_json.py

Paths need to be updated:

• IMG_FOLDER: Images with max-projection
• SEGMENTATION_FOLDER: path to save segmentation masks
• NUC_MODEL: nuclei model path
• CELL_MODEL: cell model path

If signals are too low in a plate, they need to be magnified for the segmentation algirthm to work. Note: The increased intensity is only used to create masks, subsequent analysis rely on raw intensity.

python generate_covid_dataset-max-projection.py
python generate_covid_dataset-max-projection_segmentation.py

If the experiment needs to be uploaded to annotation tool later, change SEGMENTATION_FOLDER and run

python mask_to_json.py

to create annotation.json files for each FOV.

Steps to analyse intensity hits:

Step 3. Covid and protein quantification, Violin plot

python s1_covid_quantify_well.py
python s2_combine_quantification_df.py
python s3_define_cellsize_and_virus_threshold.py
python s4_violinplot_adjustment.py

Paths need to be updated:

• IMG_FOLDER: Images with max-projection
• SAMPLE_DEST: sample destination
• (s4_)NONINFECTED: Wells or well pattern with non-infected cells
• (s4_)INFECTED: Wells or well pattern with non-infected cells

Step 4. Fold change, volcano and circos plot

python covid19plots.py

Volcano and circos in R: covid19plots_<cellline>.R.