Imaging mass cytometry reveals early β-cell dysfunction and changes in immune signatures during type 1 diabetes progression in human pancreata

T1D IMC Analysis Deposited Data Single-cell Data: Single Cell objects are provided separately for both Islet + Immune panel. spe_Islet_full.rds + spe_Immune_full.rds Please note that for reproducibility purposes these data include 7 Pancreatitis samples, which have not been analyzed in the Manuscript. Raw IMC Data raw: Raw data is too large to fit into this repo and is therefore deposited in the following zenodo repositories. Immune Panel: I: 10.5281/zenodo.15074971 II: 10.5281/zenodo.15079318 III: 10.5281/zenodo.15082431 Islet Panel: I: 10.5281/zenodo.15084757 II: 10.5281/zenodo.15087441 III: 10.5281/zenodo.15088286 Processed images (.tiffs): Data is available in the following repositories: Immune panel: https://zenodo.org/records/15971803 Islet panel: https://zenodo.org/records/15919088 Pre-processing List of subfolders: masks_cells: cell segmentation masks, generated with the`03_CellSegmentation.ipynb` notebook using the *whole-cell* segmentation type. masks_islets: islet segmentation masks, generated with the`02_IsletSegmentation.ipynb` notebook. The model used for islet segmentation isalso provided (`islet_segmentation_model.pt`). spillcomp: measurements of spillover compensation slides. One slide wasgenerated for each panel (Islet and Immune), with 45 spots spiked with the 45antibodies in the panel. Each slide was measured once per batch, generating 45.txt files that are used in the second script of the data analysis workflow(02_SpilloverCompensation_cells_*.Rmd). The spillover matrices generated bythat script are also provided as .csv files in the `spillcomp` folder. Running T1D_preprocesing (see publication-associated GitHub) on the raw data will further generate the following subfolders: data_cells: single cell data. These .csv files are used as input fordownstream analysis in R. The data include marker intensities (`intensities`),cell neighbors (`neighbors`), and spatial properties (`regionprops`) andwere generated with the `04_Measurements.ipynb` notebook. data_islets: islet-level data, including marker intensities(intensities), and spatial properties (regionprops). These data weregenerated with the `04_Measurements.ipynb` notebook. img: multichannel IMC images formatted as .tiff files and generated bythe `01_Preprocessing.ipynb` notebook. In addition, metadata files are provided. These include: Antibody panels, as `panel_Islet.csv` and `panel_Immune.csv`. These panelfiles are used as input to the `01_Preprocessing.ipynb` notebook. The `cases.csv` file contains clinical metadata about pancreas donors. It isused for downstream analysis in R. The `cases_full.csv' file contains further clinical metadata, as provided by nPOD. The `order.csv` indicates in which order samples should be plotted, accordingto their clinical T1D stage. This file is also used for downstream analysis inR. # Pilot study: SCEs and Images of the PIlot study can be found at: https://zenodo.org/uploads/15072596