Figure and Table Code

This page maps the bioRxiv manuscript figures and supplementary tables to the curated notebooks in the GitHub repository. Each notebook is intentionally output-free and contains comments describing the original data paths, provenance, and expected generated files.

Main Figures

Result	Topic	Curated notebook	Reproduction notes
Figure 1	Synthetic modular, nested, and random spatial benchmarks; COSTE, Squidpy, Giotto, and ANE comparison.	Figure_1_synthetic_benchmark.ipynb	Canonical source came from A100 benchmark notebooks. Synthetic data generation is included, but some final panels were assembled from PDFs.
Figure 2	Neonatal mouse pup Xenium StructureMap and hierarchical structures.	Figure_2_mouse_pup_xenium.ipynb	Requires mouse pup Xenium outputs and t-by-c result tables.
Figure 3	Lung fibrosis COSTE analysis, SSS, regional TRS, and TRU remodeling.	Figure_3_lung_fibrosis_TRU.ipynb	Requires Vannan lung fibrosis processed data and regional annotations.
Figure 4	Fibrosis progression and DST-GNN modeling from SSS matrices.	Figure_4_lung_fibrosis_DST_GNN.ipynb	Requires flattened SSS tables and recovered DST-GNN workspace.
Figure 5	Segment-free SSc pleura transcript/cell StructureMap, circular hierarchy, and selected gene panels.	Figure_5_SSc_pleura_segment_free.ipynb	Requires SSc pleura Xenium/transcript data and t-by-c outputs.

Supplementary Figures

Result	Topic	Curated notebook	Notes
Supplementary Figure 1	Synthetic modular benchmark.	Supp_Fig_01_synthetic_modular.ipynb	A100 modular benchmark source.
Supplementary Figure 2	Synthetic nested spatial patterns.	Supp_Fig_02_synthetic_nested_patterns.ipynb	A100 nested benchmark source.
Supplementary Figure 3	COSTE nested-pattern heatmaps.	Supp_Fig_03_COSTE_nested_heatmaps.ipynb	Uses generated COSTE benchmark outputs.
Supplementary Figure 4	Squidpy nested-pattern heatmaps.	Supp_Fig_04_Squidpy_nested_heatmaps.ipynb	Uses generated Squidpy benchmark outputs.
Supplementary Figure 5	Giotto nested-pattern heatmaps.	Supp_Fig_05_Giotto_nested_heatmaps.ipynb	Uses Giotto benchmark source.
Supplementary Figure 6	ANE nested-pattern heatmaps.	Supp_Fig_06_ANE_nested_heatmaps.ipynb	Uses analytical neighborhood enrichment source.
Supplementary Figure 7	Mouse pup hierarchical structures.	Supp_Fig_07_mouse_pup_structures.ipynb	Requires mouse pup t-by-c and spatial outputs.
Supplementary Figure 8	Mouse pup method comparison.	Supp_Fig_08_mouse_pup_method_comparison.ipynb	Uses runtime/method benchmark code.
Supplementary Figure 9	Squidpy parameter sensitivity.	Supp_Fig_09_squidpy_parameter_sensitivity.ipynb	Compares neighbor/radius settings.
Supplementary Figure 10	Human lymph node StructureMap and spatial map.	Supp_Fig_10_lymph_node_structuremap.ipynb	Direct source is a Y-drive R script plus Xenium lymph node outputs.
Supplementary Figure 11	Lung fibrosis unclustered SSS heatmaps.	Supp_Fig_11_lung_fibrosis_unclustered_heatmaps.ipynb	Requires lung fibrosis SSS matrices.
Supplementary Figure 12	SSc pleura transcript/cell panels.	Supp_Fig_12_SSc_pleura_transcript_cell.ipynb	Uses SSc transcript and cell-level outputs.
Supplementary Figure 13	TNBC spatial biomarker statistics.	Supp_Fig_13_TNBC_spatial_biomarkers.ipynb	Requires Ali TNBC clinical and spatial data.
Supplementary Figure 14	TNBC subgroup heatmaps.	Supp_Fig_14_TNBC_subgroup_heatmaps.ipynb	Per-patient and subgroup Searcher/Findee heatmaps.

Supplementary Tables

Result	Topic	Curated notebook	Notes
Supplementary Table 1	Runtime and memory benchmarking for COSTE, Squidpy, Giotto, and ANE.	Supp_Table_1_runtime_memory.ipynb	Uses A100 mouse pup benchmark scripts and performance counters.
Supplementary Table 2	SSc transcript-by-cell SSS.	Supp_Table_2_SSc_transcript_by_cell.ipynb	Requires SSc transcript-by-cell result tables.
Supplementary Table 3	SSc landmark transcript SSS.	Supp_Table_3_SSc_landmark_transcript_SSS.ipynb	Requires landmark/segment-free output tables.

Detailed Provenance

For a longer inventory of original local, Y-drive, and A100 locations, see docs/cellgps_science_manuscript_code_inventory.md in the GitHub repository:

https://github.com/hutaobo/cellgps/blob/main/docs/cellgps_science_manuscript_code_inventory.md