You can access our data online via: https://spatch.pku-genomics.org
- First benchmarking of the most adavanced spatial transcriptomics technologies with subcellular resolution and large panel of genes.
- Benchmarking of multiple ST analysis tools on our datasets.
- Using spatial proteomics (CODEX) of adjacent slides of spatial transciptomics.
- The code we used for analyzing spatial transcriptomics of multiple platform, i.e., data reading, spatial clustering, cell annotation and so on.
-
1_load_data.py: load raw expression or transcripts data of different platforms using scanpy at 8-μm bin or cell resolution and save in h5ad format
-
2_8um_bin.py: process transcriptomic data from Xenium or CosMx into 8 µm resolution and remove signals outside tissue region.
-
3_registeration.py: perform the registration for paired images and apply the transformation to other channels of CODEX or spatial coordinates of ST data
-
4_diffusion.py: calculate the relative diffusion effects and minimum distance between spots inside and outside tissue for sST platforms.
-
5_correlation_with_codex.py: calculate the correlation between ST data and CODEX data over the spatial grids.
-
6_scrna.r: perform preprocessing for scRNA-seq data.
-
7_cluster.py: perform clustering for ST data.
-
8_cell_shape.py: calculate the statistics for describing the cell shape.
-
9_st_annotation.py: transfer the annotations from scRNA-seq to ST data.
-
9_st_annotation_consistency.r: assess the consistency across different annotation tools.
-
10_spatial_cluster.py: perform spatial clustering for ST and CODEX data, and assess their consistency.