The kinase ERK plays a conserved dominant role in the heterogeneity of epithelial-mesenchymal transition in pancreatic cancer cells
Data files
Aug 06, 2025 version files 8.56 GB
-
HPAF_growth_factor_panel_days1to5.zip
2.14 GB
-
HPAF_growth_factors_chemotherapy_7pathway_panel.zip
4.03 GB
-
KPCY_7160_cells_growth_factors_chemo.zip
718.08 MB
-
PDX395_cells_baseline_chemo.zip
909.44 MB
-
PDX395.zip
769.16 MB
-
README.md
3.12 KB
Abstract
Epithelial-mesenchymal transition (EMT) occurs heterogeneously among malignant carcinoma cells to promote chemoresistance. Identifying the signaling pathways involved will nominate drug combinations to promote chemoresponse, but cell population-level studies are inherently fraught, and single-cell transcriptomics are limited to indirect ontology-based inferences. To understand EMT heterogeneity at a signaling protein level, we combined iterative indirect immunofluorescence imaging of pancreas cancer cells and tumors and mutual information (MI) analysis. Focusing first on MAP kinase pathways, MI indicated that cell-to-cell variation in ERK activity dominated control of EMT heterogeneity in response to diverse growth factors and chemotherapeutics, but that JNK compensated when MEK was inhibited. Population-level models could not capture these experimentally validated MI inferences. The dominant role of ERK was persistently indicated by MI even when analyzing seven potential EMT-regulating signaling nodes. More generally, this work provides an approach for studying multivariate signaling/phenotype relationships based on protein measurements in any setting.
Dataset DOI: 10.5061/dryad.2280gb653
Description of the data and file structure
Fluorescence intensity measurements from CellProfiler analysis of iterative indirect immunofluorescence imaging experiments measuring pathway activation proxies and cell phenotype markers. These data are used for mutual information analysis in our manuscript "The kinase ERK plays a conserved dominant role in the heterogeneity of epithelial-mesenchymal transition in pancreatic cancer cells."
Each row is one object (cell) that was measured. Similar files ending in _Cell, _Cytoplasm, _Nuclei, etc., can be merged using the Image and Object numbers from the dataset.
VIP variables:
- ImageNumber: Unique identifier for each image processed in the pipeline.
- ObjectNumber: Unique identifier for each object found in the associated image.
- Metadata_Well: Well number associated with a specific treatment condition, detailed in the experiment README files.
- Intensity_IntegratedIntensity_{Target}: Total intensity of all of the pixels in the area of the segmented object for a given immunofluorescence image (arbitrary intensity units)
- Example: Intensity_Integrated_Intensity_Vimentin
Files and variables
File: HPAF_growth_factor_panel_days1to5.zip
Description: Data associated with Figure 1.
Files:
- Days1to5_GFs_Plate1_Cell.csv
- Days1to5_GFs_Plate1_Cytoplasm.csv
- Days1to5_GFs_Plate1_Nuclei.csv
- Days1to5_GFs_Plate2_Cell.csv
- Days1to5_GFs_Plate2_Cytoplasm.csv
- Days1to5_GFs_Plate2_Nuclei.csv
- README_HPAF_growth_factor_panel_days1to5
File: KPCY_7160_cells_growth_factors_chemo.zip
Description: Data associated with Figure 2
Files:
- 7160_4i_Nuclei.csv
- 7160_4i_Cell.csv
- 7160_4i_Cytoplasm.csv
- README_KPCY_derived_7160_growth_factors_gemcitabine
File: PDX395_cells_baseline_chemo.zip
Description: Data associated with Figure 2
Files:
- PDX395_cells_chemo_Nuclei.csv
- PDX395_cells_chemo_Cell.csv
- PDX395_cells_chemo_image_qc.csv
- README_PDX_395_baseline_chemo_inhibitors
File: HPAF_growth_factors_chemotherapy_7pathway_panel.zip
Description: Data associated with Figure 6
Files:
- Full_panel_4i_Nuclei.csv
- Full_panel_4i_Cell.csv
- Full_panel_4i_Cytoplasm.csv
- Full_panel_4i_image_qc.csv
- README_HPAF_full_pathway_panel_chemo_growth_factor
File: PDX395.zip
Description: Data associated with Figure 7
Files:
- 395full_Nuclei.csv
- 395full_Ecad.csv
- 395full_Cytoplasm.csv
- Pilot_395_Z_Mesmr_Nuclei.csv
- Pilot_395_Z_Mesmr_Ecad.csv
- Pilot_395_Z_Mesmr_Cytoplasm.csv
- README_PDX_395
Code/software
Code for analysis available on Zenodo: https://doi.org/10.5281/zenodo.16042277
Access information
Other publicly accessible locations of the data:
- Data also available through the University of Virginia Dataverse: https://doi.org/10.18130/V3/KJCIGC
- Barbeau, Michelle C.; Brown, Brooke A.; Adair, Sara J. et al. (2025). The kinase ERK plays a conserved dominant role in the heterogeneity of epithelial-mesenchymal transition in pancreatic cancer cells. Science Signaling. https://doi.org/10.1126/scisignal.ads7002