Data for: A Morphology-Based Machine Learning Model for Scoring Epithelial-Mesenchymal Plasticity using Organelle Dynamics
https://doi.org/10.48723/m1cg-v223
Project Description:
This project is aimed at modelling dynamic organelle morphology changes in a cellular model of epithelial-mesenchymal transition (EMT). EMT is a developmental process, which is re-activated in cancer and promotes invasion, metastasis and resistance to various therapies.
The repository contains raw microscopy images (.nd2 files) and numerical data outputs of all the experiments performed on a variety of cell lines (breast cancer, lung cancer). Cells were stained with the CellPainting kit. Experiments codes are listed below.
Categories:
Translational and applied bioinformatics; Cellular interactions (incl. adhesion, matrix, cell wall)
Numerical Data Source_wet lab + database (782.78 KB)
(1 excel file with numerical data source for MS experiments; 1 pdf file with WB uncropped membranes images)
Numerical Source Data_all CP experiments (microscopy): (14.47 GB)
- CP015/016/018: NMuMG training set (cellprofiler outputs)
- CP017: NMuMG reversal experiment
- CP019:NMuMG model validation
- CP020: breast cancer cells
- CP_INT_03_Normoxia: A549 Normoxia
- CP_INT_03_Hypoxia: A549 Hypoxia
Images (.ND2 files):
- CP015_nd2 (37.24 GB): NMuMG training set.1
- CP016_nd2 (37.24 GB): NMuMG training set.2
- CP018_nd2 (37.24 GB): NMuMG training set.3
- CP017_nd2 (37.24 GB): NMuMG reversal experiment
- CP019_nd2 (37.24 GB): NMuMG model validation
- CP020_nd2 (37.24 GB): breast cancer cells
- CP_INT_03_Normoxia_nd2 (18.62 GB): A549 Normoxia
- CP_INT_03_Hypoxia_nd2 (18.62 GB): A549 Hypoxia
Image files are in nd2 format, which is Nikon's proprietary format, but can be viewed with open source tools such as ImageJ.
Data files
Data files
Citation and access
Citation and access
Data access level:
Creator/​Principal investigator(s):
Research principal:
Data contains personal data:
No
Citation:
Language:
Method and outcome
Method and outcome
Unit of analysis:
Population:
This study was performed with established cell lines and did not involve human subjects or animals.
Time method:
Study design:
- Experimental study
- Preclinical study
Description of study design:
A morphology-based machine learning approach was developed to score EMT based on changes in organelle dynamics. Using the Cell Painting assay and high-throughput microscopy, we trained a histogram gradient boosting classifier to identify stage-specific organelle remodeling during a time course of TGF-β1-induced EMT in mammary epithelial cells.
Time period(s) investigated:
Data format/​data structure:
Geographic coverage
Geographic coverage
Geographic location:
Administrative information
Administrative information
Responsible department/​unit:
Department of Laboratory Medicine [H5]
Other research principals:
Contributor(s):
Funding
Funding
Funding agency:
- Swedish Cancer Society
Opens a new window at ror.org.
ROROpens in a new tab
Award number:
21 1739 Pj; 24 3842 Pj
Funding agency:
- Radium Hemmets Research Funds
Opens a new window at ror.org.
ROROpens in a new tab
Award number:
211092; 231172
Topic and keywords
Topic and keywords
Standard för svensk indelning av forskningsämnen 2025:
Publications
Publications
Citation:
Fuxe, J., Slager, J., Gatto, F., Frey, B., Porebski, B., Carreras-Puigvert, J., Parniewska, M., & Shi, W. (2025). A Morphology-Based Machine Learning Model for Scoring Epithelial-Mesenchymal Plasticity using Organelle Dynamics. In Research Square. https://doi.org/10.21203/rs.3.rs-5859608/v1Opens in a new tab
