MOAProfiler_IN_MOA: Preprocessed Cell Painting MOA plate (images and manifest.csv) for MOAProfiler_Feature_Extractor

Summary

MOAProfiler_IN_MOA is a preprocessed Cell Painting dataset built from a single MOA plate and prepared specifically for running the MOAProfiler_Feature_Extractor model in an embedding-only regime. The dataset contains five-channel Cell Painting images transformed into the input format required by the MOAProfiler embedding scripts, together with a manifest file (manifest.csv) that links each image to well, site, channel, and perturbation metadata.

This dataset corresponds to the same experimental plate as the CellPaintSSL_IN_MOA dataset, but uses a preprocessing and manifest format tailored to MOAProfiler_Feature_Extractor rather than CellPaintSSL.

Contents

images/ – preprocessed single-plane 16-bit TIFF images (five fluorescence channels, 1080×1080 px), organised by plate, well, and site in the format expected by MOAProfiler_Feature_Extractor (one TIFF per channel and field of view, e.g. ch1.tif … ch5.tif).

manifest.csv – manifest linking each image to plate, well, site, channel and file paths, used by the MOAProfiler_Feature_Extractor code to locate input files and assemble per-field five-channel stacks.

How the data were generated

Raw Cell Painting images from the MOA plate were acquired as separate 16-bit TIFF files per channel using a high-content screening system. Illumination correction was performed and the fields of view were then cropped and padded to a standardized size of 1080×1080 pixels, matching the expected input of the MOAProfiler embedding pipeline. For each field of view and channel, the corrected images were saved as 16-bit TIFF files with a consistent naming scheme (ch1.tif … ch5.tif) and organised into the images/ directory by plate, well, and site.

The manifest.csv file was generated from the plate map and image metadata and contains, for every channel and site, the relative path to the corresponding TIFF together with plate, well, site and perturbation annotations. The resulting MOAProfiler_IN_MOA dataset therefore consists of five separate 16-bit TIFF images per field of view (one per channel) and a manifest that enables direct use with the MOAProfiler_Feature_Extractor embedding scripts.