# Segmentation_shape_vhfi The folder segmentation_shape_vhfi contains the segmentation shapes used in the RNA2seg paper with the dataset fromhttps://vizgen.com/data-release-program/ at MERSCOPE FFPE Human Immuno-Oncology Data Release -- Cellbound1, Cellbound3 and DAPI are the segmentations done by cellpose of the corresponding staining -- VHFI is the cell segmentation provided by vizgen -- annotation is the annotation done for RNA2seg paper --RNA2seg is the segmentation done by RNA2seg each segmentation can be loaded with : each segmentation can be load with : ``` import geopandas as gpd gdf = gpd.read_file(path2save_shape_organ / "shape.shp")``` Staining and transcripts are available at https://vizgen.com/data-release-program/ at MERSCOPE FFPE Human Immuno-Oncology Data Release # Segmentation_shape_cosmx The folder segmentation_shape_cosmx contains the segmentation shapes used in the RNA2seg paperwith the dataset from this study : He, S., Bhatt, R., Brown, C. et al.High-plex imaging of RNA and proteins at subcellular resolution in fixed tissue by spatial molecular imaging.Nat Biotechnol 40, 1794–1806 (2022). https://doi.org/10.1038/s41587-022-01483-z -- CD3, CD45, DAPI and Membrane are the segmentations done by cellpose of the corresponding staining -- cosmx_output is the cell segmentation provided by vizgen --RNA2seg is the segmentation done by RNA2seg each segmentation can be loaded with : ``` import geopandas as gpd gdf = gpd.read_file(path2save_shape_organ / "shape.shp")``` Staining and transcripts were provided by the authors of the study # mouse_ileum The folder mouse_ileum contains the segmentation shapes used in the RNA2seg paper with the dataset from :Petukhov, V., Xu, R.J., Soldatov, R.A. et al.Cell segmentation in imaging-based spatial transcriptomics.Nat Biotechnol 40, 345–354 (2022). https://doi.org/10.1038/s41587-021-01044-w -- cellpose is the segmentation done with cellpose by the original authors of the study -- annotation is the annotation done for RNA2seg paper # benchmark_staining_methodThe folder benchmark_staining_method contains the benchmarking results forsegmentation from merscope (VHFI) and cellpose. # benchmark_point_cloud_method The folder point_cloud_method contains the script and the benchmarking resultson our manual annotation for ComSeg and Baysor # CNN_trained_model model RNA2seg-cnn used for the benchmarck contact : alice.blondel@minesparis.psl.eu, thomas.defard@minesparis.psl.eu