Dataset title: ArtSeg-CHO-M4R, artifact segmentation in microscopy of ligand binding to M4 muscarinic receptor in live CHO-K1-hM4 cells. Authors: Mohammed A. S. Ali, Kaspar Hollo, Tõnis Laasfeld, Jane Torp, Maris-Johanna Tahk, Ago Rinken, Kaupo Palo, Leopold Parts, Dmytro Fishman Contact information: dmytro.fishman@ut.ee Department of Computer Science, University of Tartu, Narva mnt 18, 51009, Tartu, Estonia. & leopold.parts@ut.ee Wellcome Sanger Institute, Wellcome Genome Campus, Hinxton, Cambridgeshire, CB10 1SA, United Kingdom. General dataset description: This dataset's acuisition details were previously described in: Tahk, Maris-Johanna; Torp, Jane; Ali, Mohammed A.S.; Fishman, Dmytro; Parts, Leopold; Grätz, Lukas; Müller, Christoph; Keller, Max; Veikšina, Santa; Laasfeld, Tõnis; Rinken, Ago. 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 three zip files for `train`, `validation` and `test` splits. Each zip file contains `clean` and `artifact` folders, which include images without and with artifacts, respectively. In the `artifact` folder there are two subfolders `x` and `y` containing input images and artifact segmentation ground truth masks, respectively. On the other hand, the `clean` folders have one subfolder which contains input images with no artifacts on them. To get clean ground truth segmentation masks, one has to generate images with the same dimention of the input images with value of `1` in all pixels. Therfore, we did not include the clean segmentation masks here. Images naming conventions and data types: The images in the subfolders names by numbers. For example, in the folder `x`, the input image with name `1` has corresponding ground truth mask with name `1` in the folder `y` in the same level. The input images' datatype is unsigned integer (uint8) normalized between 0 and 255. The normalization is as following: Pixel_value = ((Pixel_value-min)/(max-min))*255. Where 'min' and 'max' are the minimum and the maximum pixel vlue in the corresponding image, respectively. The ground truth pixel values are 0 and 1. Ground truth mask generation: The ground truth segmentation masks is generated manually by an expert the MembraneTools module of Aparecium software 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. FEBS J. 288, 1514–1532 (2021). Additional information This dataset is part of UT-GPCR001 datset available here: Tahk, Maris-Johanna; Torp, Jane; Ali, Mohammed A.S.; Fishman, Dmytro; Parts, Leopold; Grätz, Lukas; Müller, Christoph; Keller, Max; Veikšina, Santa; Laasfeld, Tõnis; Rinken, Ago. https://datadoi.ee/handle/33/432 http://dx.doi.org/10.23673/re-306 License information: "ArtSeg-CHO-M4R, artifact segmentation in microscopy of ligand binding to M4 muscarinic receptor in live CHO-K1-hM4 cells" 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 the Estonian Research Council (IUT34-4) and the Estonian Center of Excellence in IT (EXCITE) (TK148); Estonian Research Council grants (PRG1095, PSG59 and ERA-NET TRANSCAN-2 (BioEndoCar)); Project No 2014-2020.4.01.16-0271, ELIXIR and the European Regional Development Fund through EXCITE Center of Excellence. The University of Tartu ASTRA Project PER ASPERA, financed by the European Regional Development Fund, COST action CA 18133 ERNEST; Wellcome (206194); and PerkinElmer Cellular Technologies.