Machine Learning for Spatial Omics
What it does: Automated cell-type discovery and tissue-scale mapping of cell-type-specific spatial architectures from cycleHCR data using machine learning.
ML-guided spatial omics pipeline (Lian, Adjavon et al., 2026, bioRxiv):
https://github.com/liulabspatial/Omics-cycleHCR-QuAC
cycleHCR — Deep-Tissue Spatial Transcriptomics
What it does: Multiplexed imaging of RNA and protein targets in thick tissue volumes (>300 µm) with subcellular resolution.
Data processing pipeline (Gandin, Kim et al., 2025, Science) — GitHub: github.com/liulabspatial/CycleHCR-Pipeline — Archive: doi: 10.5281/zenodo.14780420
Gene Co-Expression Analysis
What it does: Quantification of single-cell gene co-expression patterns from combined imaging and genomics data.
Data and code (Dong et al., 2024, Nature Genetics) — Archive: doi: 10.5281/zenodo.11406939
3D ATAC-PALM — Super-Resolution Genome Imaging
What it does: Nanoscale 3D maps of the accessible genome in single cells using ATAC-seq chemistry combined with lattice light-sheet PALM microscopy.
Analysis code (Xie, Dong et al., 2020, Nature Methods) — GitHub: github.com/ammondongp/3D_ATAC_PALM
Molecular simulation code (Xie, Dong et al., 2022, Nature Genetics) — GitHub: github.com/ZhangGroup-MITChemistry/Chrom_cluster
Created by Dall-E