For GIT dataset, formalin-fixed paraffin-embedded (FFPE) tissue microarray (TMA) slides were obtained from Pantomics (Pantomics, DID381) Tissue TMA samples for Supplementary Figures 12, 13, and 14 were obtained from Department of Pathology at University of Pittsburgh Medical Center Presbyterian Hospital. The slides went through cyclic immunofluorescence antigen retrieval protocol10 (link). The corresponding figure slides were stained in cycles with 1:200 dilution of Anti-Sodium Potassium ATPase antibody (Abcam ab198367, clone EP1845Y), 1:100 dilution of P53 antibody (Abcam ab270192, clone SP5), 1:50 dilution of EZH2 antibody (CST 45638, clone D2C9), and 1:100 dilution of LAMIN A/C antibody (CST 8617, clone 4C11) overnight at 4°C in the dark, followed by staining with Hoechst 33342 (CST 4082S) for 10 minutes at room temperature in the dark. TMA images were acquired using a 0.95 NA and a 40X objective on a Nikon Ti2E microscope.
Seventy five, 1000 × 1000 high quality regions were identified and extracted from the TMA images and saved as tiff images. Expert pathologists independently annotated these images. The annotations were done using Cellthon, a python based cell annotation graphical user interface (GUI) we created using Tkinter toolkit55 . Together these 75 images and their cell and nucleus annotations comprise the GIT dataset.
Free full text: Click here