Mitocheck Data Supplementary Files

This dataset contains all supplementary files related to the mitocheck_data repository that were too large to upload on GitHub.

The compressed folders include extracted and normalized single-cell features from the following subsets of the original screen by MitoCheck Consortium:

• negative_control_data/ : Nuclei features from negative control cells transfected with scrambled siRNA.
• positive_control_data/ : Nuclei features from positive control cells transfected with siRNA targeting genes used during mitosis (INCENP, KIF11, COPB1).
• training_data/ : Nuclei features from cells manually labeled with a phenotypic class by MitoCheck Consortium.

More details for the compressed folders are as follows:

• 1.idr_streams/extracted_features/ : All IDR_stream-extracted single-cell features (CellProfiler, DeepProfiler, merged) for the data subsets of the original MitoCheck Consortium screen. More information can be found in mitocheck_data/1.idr_streams.
• 3.normalize_data/normalized_data/ : All IDR_stream-extracted single-cell features (CellProfiler, DeepProfiler, merged) normalized with sklearn.preprocessing.StandardScaler. One StandardScaler was derived from the negative control features and applied to all data subsets. More information can be found in mitocheck_data/3.normalize_data.

The Image Data Resource (IDR) hosts all data for the original screen by MitoCheck Consortium under accession idr0013 (screenA). Dr. Jean-Karim Heriche and Dr. Thomas Walter of the MitoCheck Consortium shared the manually-assigned phenotypic class labels used in training_data.