Welcome to S2-omics’s documentation!
S2-omics is an end-to-end workflow for designing smart spatial omics experiments using histology images. It automatically selects optimal Regions of Interest (ROIs) for spatial omics acquisition and utilizes the resulting data to virtually reconstruct spatial molecular profiles across entire tissue sections. This minimizes experimental cost while preserving critical spatial molecular variations.
Key Features
Histology image-guided ROI selection using foundation models (UNI, Virchow2, Prov-GigaPath, HIPT).
Whole-slide spatial reconstruction from partial spatial omics data.
3D multi-slice ROI selection capability.
Modular pipeline for preprocessing, segmentation, clustering, ROI selection, and label broadcasting.
Paper link: https://www.biorxiv.org/content/10.1101/2025.09.21.677634v1
Accepted in principle by Nature Cell Biology.
Contents
Documentation
- Installation Guide
- Tutorials
- Tutorial 1: Design VisiumHD experiment for a colorectal cancer section
- Step 1: Preprocess the H&E image
- Step 2: Quality control for all superpixels
- Step 3: Histology feature extraction
- Step 4: Histology segmentation
- Step 5: Merge over-clusters
- Step 6: Select best ROI for VisiumHD experiment
- (Optional) Step 7: Cell type label broadcasting
- Tutorial 2: Design CosMx experiment for 2 kidney sections
- Step 1: Preprocess the H&E image
- Step 2: Quality control for all superpixels
- Step 3: Histology feature extraction
- Step 4: Histology segmentation
- Step 5: Merge over-clusters
- Step 6: Select best ROI for CosMx experiment
- Step 6: Select best FOV for CosMx experiment
- Tutorial 3: Design spatial omics experiment for consecutive breast cancer sections
- Step 1: Preprocess the H&E image
- Step 2: Quality control for all superpixels
- Step 3: Histology feature extraction
- Step 4: Joint histology segmentation
- Step 5: Select best ROI for spatial omics experiment
- Tutorial 4: Design Tissue Micro Array experiment for multiple breast biopsies
- Step 1: Preprocess the H&E image
- Step 2: Quality control for all superpixels
- Step 3: Histology feature extraction
- Step 4: Histology segmentation
- Step 5: Merge over-clusters
- Step 6: Select best ROI for TMA experiment
- Usage
- API Reference
- s2omics.p1_histology_preprocess
- s2omics.p2_superpixel_quality_control
- s2omics.p3_feature_extraction
- s2omics.single_section.p4_get_histology_segmentation
- s2omics.single_section.p5_merge_over_clusters
- s2omics.single_section.p6_roi_selection_rectangle
- s2omics.single_section.p6_roi_selection_circle
- s2omics.single_section.p7_cell_label_broadcasting
- s2omics.multiple_sections.p4_get_histology_segmentation
- s2omics.multiple_sections.p6_roi_selection_rectangle
- s2omics.single_section.p6_roi_selection_circle
- s2omics.multiple_sections.p6_cell_label_broadcasting
Citation
If you use this tool in your research, please cite our work.