Dataset title: UT-GPCR004 CHO-K1 cell line bright-field and fluorescence microscopy and corresponding segmentation ground truth Authors: Tahk, Maris-Johanna; Torp, Jane; Ali, Mohammed A.S.; Fishman, Dmytro; Parts, Leopold; Grätz, Lukas; Müller, Christoph; Keller, Max; Veiksina, Santa; Laasfeld, Tõnis; Rinken, Ago Contact information: laasfeld@ut.ee; ago.rinken@ut.ee Chair of Bioorganic chemistry, Institute of Chemistry, University of Tartu General dataset description: The dataset was gathered as a part of Tahk MJ, Torp J, Ali MA, Fishman D, Parts L, Grätz L, Müller C, Keller M, Veiksina S, Laasfeld T, Rinken A. Live-cell microscopy or fluorescence anisotropy with budded baculoviruses-which way to go with measuring ligand binding to M4 muscarinic receptors?. bioRxiv. 2021 Jan 1. https://doi.org/10.1101/2021.12.22.473643 The dataset consists of two zip files containing raw and processed intermediate data (background corrections, manual ground truth and results of intermediate models) corresponding to Ilastik and Deep-learning pipelines respectively Individual file naming convention: All raw data files use the following naming convention: ______.tif Binary segmentation masks follow the same naming convention such that the binary mask has the same name as the original image that was considered as "in focus" plane by the corresponding segmentation analysis. "Ilastik training" folder RF-FL-1 output : Output predictions of the RF-FL-1 model as hdf5 images as simple two class segmentation of Intracellular area of CHO-K1-hM4R cell line. Class balanced RF-FL-1 output : Output predictions of the RF-FL-1 model as png images modified to represent three classes: Near membrane background (NMBG, pixel value 1), Intracellular (IC, pixel value 2) and Membrane (MB, pixel value 3). Pixel value 0 corresponds to label not given to this pixel. Class balanced RF-FL-1 output with added ground truth and background class : Output predictions of the RF-FL-1 model with added pixel labels to the areas where the RF-BF-1 model made most errors as png images modified to represent four classes: Near membrane background (NMBG, pixel value 1), Intracellular (IC, pixel value 2) and Membrane (MB, pixel value 3) and Background (BG, pixel value 4). Pixel value 0 corresponds to label not given to this pixel. Contrast enhanced bright-field Z-stack projection images : Input images for the RF-BF-1 and RF-BF-2 models. The contrast enhanced bright-field Z-stack projection images were generated using Aparecium software (https://github.com/laasfeld/Aparecium and https://gpcr.ut.ee/aparecium.html) using stackLinearRegPartial.m function as described in "Allikalt A, Laasfeld T, Ilisson M, Kopanchuk S, Rinken A. Quantitative analysis of fluorescent ligand binding to dopamine D3 receptors using live‐cell microscopy. The FEBS journal. 2021 Mar;288(5):1514-32." Manual ground truth for Ilastik : Manually drawn ground truth from the fluorescence images. The annotations contains two classes: Cells (pixel value 1) and background (pixel value 2). Pixel value 0 corresponds to label not given to this pixel. Raw bright-field and fluorescence images : Contains the raw fluorescence and bright-field images (all Z-planes) aligned with the manual ground truth available for RF-FL-1 model "U-Net3 training" folder Flat-field corrected fluorescence images : Contains 191 flat-field corrected tiff images of CHO-K1-hM4 cells with added plasmamembrane label DiI imaged in the Red fluorescence protein (RFP) channel. Manual ground truth : Contains 10 binary png images corresponding to ground truth cell cytoplasm segmentation. Images were drawn on the flat-field corrected RFP images. The image names correspond to the fluorescence images that were used to generate this ground truth. U-Net3-FL-1 predictions : Contains 191 binary tif images corresponding to the RFP images. The images were generated by the developed U-Net3-FL-1 Keras model from the Flat-field corrected fluorescence images. Bright-field images : Contains 191 in-focus bright-field tiff images of CHO-K1-hM4 aligned with the "Flat-field corrected fluorescence images". Methodologial information: All data has been gathered using Cytation 5 imaging multimode plate reader. Each raw data tiff file contains detailed metadata about specific acquisition conditions. Processed images were generated as described in the connected publication (https://doi.org/10.1101/2021.12.22.473643). Aparecium and Ilastik software were used for generating and analyzing the images available at https://github.com/laasfeld/Aparecium and https://gpcr.ut.ee/aparecium.html and https://www.ilastik.org/download.html respectively Data specific information: All raw data uses arbitrary (relative) intensity units. Abbreviations: RFP - Red fluorescent protein RF - random forest BF - Bright-field License information: "UT-GPCR004 CHO-K1 cell line bright-field and fluorescence microscopy and corresponding segmentation ground truth" is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Funding: This dataset was supported by the University of Tartu ASTRA Project PER ASPERA, financed by the European Regional Development Fund, by the Enterprise Estonia Applied research programme 2021, financed by the European Regional Development Fund, by the Estonian Research Council grant (PSG230), by the COST action CA 18133 ERNEST, by the Research Training Group GRK1910 of the Deutsche Forschungsgemeinschaft (DFG), Wellcome (206194), and the Estonian Centre of Excellence in IT (EXCITE) (TK148).