Spatial-Human-Kidney-Map

A. Exporting Spatial Data

• Exporting_CosMx_Script1.py
  Using exported data from AtoMx for the spatial datasets, this script generates an h5ad object of the entire slide for further usage.
  Container used: docker://10jll/spatial:version2

• Exporting_CosMx_Script2.py
  Partitions the slide data into individual samples and generates QC information.
  Container used: docker://10jll/spatial:version2

• Exporting_Xenium.py
  Exports Xenium spatial datasets from the raw output and generates QC information.
  Container used: docker://10jll/xenium:version4

---

B. Integration

Integration_CosMx

• Integration_CosMx_Script1.py
  Prepares all CosMx files for integration and concatenates with the reference SN atlas.
  Container used: docker://10jll/spatial:version2

• Integration_CosMx_Script2.py
  Runs SCVI and SCANVI to filter the CosMx dataset.
  Container used: docker://10jll/scvi_cuda12:version4

Integration_Xenium

• Integration_Xenium_Script1.py
  Prepares all Xenium files for integration and concatenates with the reference SN atlas.
  Container used: docker://10jll/spatial:version2

• Integration_Xenium_Script2.py
  Runs SCVI and SCANVI to filter the Xenium dataset.
  Container used: docker://10jll/scvi_cuda12:version4

Final_Combined_Integration

• Final_Combined_Integration_Script1_Concat.py
  Concatenates filtered CosMx, Xenium, and SN reference atlas.
  Container used: docker://10jll/spatial:version2

• Final_Combined_Integration_Script2_SCVI_SCANVI.py
  Runs final integration using SCVI and SCANVI.
  Container used: docker://10jll/scvi_cuda12:version4

• Final_Combined_Integration_Script3_Annotation.py
  Performs Leiden clustering and annotation.
  Container used: docker://10jll/spatial:version2

• Final_Combined_Integration_Script4_Imputation.py
  Imputes spatial dataset using SCANVI latent space.
  Container used: docker://10jll/spatial:version2

---

C. SCIB (after Luecken et al. 2021)

• SCIB_Script1_Harmony_Scanorama.py
  Integrates spatial dataset with SN-Seq using Harmony and Scanorama.

• SCIB_Script2_Harmony_Pyliger.py
  Integrates using Pyliger.

• SCIB_Script3_scvi_scanvi.py
  Integrates using SCVI and SCANVI.

• SCIB_Script4_benchmarking.py
  Calculates benchmarking parameters.

These scripts were run on Google Colab.

---

D. Niche Analysis

• Niche_Analysis_Script1_call_neighbors.py
  Calls neighbors in CosMx and Xenium datasets.
  Container used: docker://10jll/spatial:version2

• Niche_Analysis_Script2_Neighbor_Dataframe.py
  Generates the neighbor dataframe.
  Container used: docker://10jll/spatial:version2

• Niche_Analysis_Script3_Kmeans.py
  K-means clustering of the neighbor dataframe.
  Container used: docker://10jll/spatial:version2

• Niche_Analysis_Script4_niche_annotation.py
  Niche annotation.
  Container used: docker://10jll/spatial:version2

• Niche_Analysis_Script5_cellphoneDB.py
  Prepares and runs CellPhoneDB using imputed expression and microenvironment key.
  Container used: docker://10jll/cellphonedb:version1

• Niche_Analysis_Script6_GFR_correlation.py
  Correlates CellPhoneDB interaction scores with disease severity in the glomerular niche.
  Container used: docker://10jll/spatial:version2

• Niche_Analysis_Script7_DEGs.py
  Calculates DEGs between DKD and control patients at niche level.
  Container used: docker://10jll/spatial:version2

• Niche_Analysis_Script8_spatial_genesignatures.py
  Generates pseudobulk counts and spatial gene signatures.
  Container used: docker://10jll/spatial:version2

• Niche_Analysis_Script9_DESeq2.R
  Runs DESeq2 on pseudobulk counts.
  Run on: Google Colab

---

E. NicheCompass

• Nichecompass_Script1_setup.py
  Sets up model and computes adjacency matrices.
  Container used: docker://10jll/nichecompass:version1

• Nichecompass_Script2_model.py
  Runs the NicheCompass model.
  Container used: docker://10jll/nichecompass:version1

• Nichecompass_Script3_clustering_annotation.py
  Performs clustering, annotation, and gene program comparison.
  Container used: docker://10jll/nichecompass:version1

---

F. COVET

• COVET_Script1.py
  Runs COVET on the CosMx dataset.

• COVET_Script2_Visualization.py
  Downstream visualizations.

• COVET_Script3_Hotspot.py
  Hotspot analysis based on latent niche representation.
  Run in local environment

• COVET_Script4_Hotspot_Visualization.py
  Visualizes gene-gene correlation matrices.
  Run on: Google Colab

---

G. Injured Tubular Microenvironments

• Tubular_ME_Script1_Kmeans.py
  K-means clustering of neighbor dataframe.
  Container used: docker://10jll/spatial:version2

• Tubular_ME_Script2_annotation.py
  Annotation of microenvironments.
  Container used: docker://10jll/spatial:version2

• Tubular_ME_Script3_20um_Env.py
  Infers 20um neighborhood for each ME.
  Container used: docker://10jll/spatial:version2

• Tubular_ME_Script4_cellphoneDB.py
  Calculates ligand-receptor interactions using imputed expression.
  Container used: docker://10jll/cellphonedb:version1

• Tubular_ME_Script5_correlation.py
  Correlates iPT–Fibroblast interaction scores with disease severity.
  Container used: docker://10jll/spatial:version2

• Tubular_ME_Script6_cellfractions.py
  Statistical comparison of cell types across injured tubular MEs.
  Container used: docker://10jll/spatial:version2

---

H. Immune Cell Atlas

• Script1_Prepare_SCVI_Xenium.py
  Prepares Xenium and SN-Seq input for SCVI/SCANVI.
  Container used: docker://10jll/spatial:version2

• Script2_Xenium_scvi_scanvi.py
  Runs SCVI/SCANVI on Xenium and SN-Seq.
  Container used: docker://10jll/scvi_cuda12:version4

• Script3_Add_CosMx.py
  Adds CosMx to integrated Xenium and SN for SCVI/SCANVI.
  Container used: docker://10jll/spatial:version2

• Script4_scvi_scanvi.py
  Runs SCVI/SCANVI on CosMx, Xenium, and SN-Seq.
  Container used: docker://10jll/scvi_cuda12:version4

• Script5_Annotation_Imputation.py
  Performs clustering, annotation, and gene expression imputation from SN-Seq.
  Container used: docker://10jll/spatial:version2

---

I. Immune Microenvironments

• Immune_ME_Script1_call_neighbors.py
  Generates neighbor dataframe using immune cell annotations.
  Container used: docker://10jll/spatial:version2

• Immune_ME_Script2_Kmeans.py
  Clusters immune cells into microenvironments.
  Container used: docker://10jll/spatial:version2

• Immune_ME_Script3_20um_Env.py
  Infers 20um environments.
  Container used: docker://10jll/spatial:version2

• Immune_ME_Script4_prepare_cellphoneDB.py
  Prepares CellPhoneDB with imputed expression.
  Container used: docker://10jll/spatial:version2

• Immune_ME_Script5_cellphoneDB.py
  Runs CellPhoneDB with a microenvironment key.
  Container used: docker://10jll/cellphonedb:version1

• Immune_ME_Script6_correlation.py
  Correlates interaction scores in immune MEs with disease severity.
  Container used: docker://10jll/spatial:version2

• Immune_ME_Script7_call_neighbors.py
  Visualizes cell neighborhoods and compares immune cell fractions between glomerular and tubular immune MEs.
  Container used: docker://10jll/spatial:version2

J. B Cell Atlas

• Script1_prepare_SC.py
  Prepares SC data.
  Container used: docker://10jll/spatial:version2

• Script2_integration_SC.py
  Integrating SC data.
  Container used: docker://10jll/scvi_cuda12:version4

• Script3_add_Xenium.py
  B and plasma cells extracted from Xenium 5k dataset are concatenated with the SC data.
  Container used: docker://10jll/spatial:version2

• Script4_integration_Xenium_SC.py
  Integrating Xenium and SC using SCVI and SCANVI.
  Container used: docker://10jll/scvi_cuda12:version4

• Script5_annotation.py
  Annotation of B cell subsets.
  Container used: docker://10jll/spatial:version2

K. Miscellaneous