Welcome to S2Omics’s documentation!
S2Omics 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 (S2Omics-3D).
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