MOAProfiler_IN: Preprocessed Cell Painting MOA plate (images and metadata) for MOAProfiler_Feature_Extractor

This dataset contains preprocessed Cell Painting images and metadata used as input to the MOAProfiler_Feature_Extractor pipeline.

The data correspond to a single JUMP-MOA plate acquired on a high-content screening system and processed into a ready-to-use format for MOAProfiler-based feature extraction. The archive MOAProfiler_IN_v0.1.tar.gz expands into the directory MOAProfiler_IN/ with the following main contents:

– Preprocessed 16-bit TIFF images (5 fluorescence channels, 1080×1080 px), organised by well and site, in the format expected by the embedding script.– manifest.csv describing all images (plate, well, site, channel, and full paths), used by the MOAProfiler_Feature_Extractor code to locate the input files.– Plate-level metadata and auxiliary files needed to reproduce the embedding and post-processing steps described in the associated manuscript and GitHub repository.

The dataset is intended to allow full reproducibility of the MOAProfiler-based morphological profiling experiments in the accompanying paper and to provide an external example dataset for evaluating deep learning models on Cell Painting images.

Code for running the embedding pipeline on this dataset is available in the GitHub repository “MOAProfiler_Feature_Extractor”.