🌌 Spatial Transcriptomics Tools

Other resource → awesome_spatial_omics

• Spatial-omic SIB tutorial

Table of Contents

• General Tools
  • Nextflow Pipelines
• Viewers & Interactive Annotation
• Probe & Panel Design
• Analysis Pipeline Steps
  • ROI Selection
  • QC
  • Normalization
  • Gene Imputation & Denoising
  • Bias Correction
  • Cell Segmentation
  • Cell Annotation
  • Cell Deconvolution
  • Differential Expression
  • Spatially Variable Genes
  • Integration
  • Cell Niches & Tissue Domains
  • Cell Distances & Neighborhood
  • Spatial Trajectories
  • Cell-Cell Communication
  • Metacells & Scalability
  • Subcellular Analysis
  • Copy Number Variations
  • Isoform Analysis
  • Transcription Factors & Gene Regulatory Networks
• Technical Enhancements
  • Slide Alignment
  • Super Resolution
  • Transcripts + Histology
• Benchmarks
  • Simulators & Ground Truth
• Datasets & Foundation Models

General Tools

• Best practices Bioconductor - [R] - Principles for statistical analysis of spatial transcriptomics data
• squidpy - [Python] - Spatial single cell analysis toolkit from scverse
• Giotto - [R/Python] - Comprehensive spatial data analysis suite
• Vitessce - [JavaScript] - Visual integration tool for exploration of spatial single cell experiments
• Voyager - [R] - Spatial transcriptomics visualization from Pachter lab
• BASS - [R] - Multiple sample analysis
• SpaVAE - [Python] - All-purpose tool for dimension reduction, visualization, clustering, batch integration, denoising, differential expression, spatial interpolation, and resolution enhancement | Implementations
• sopa - [Python] - Spatial omics processing and analysis
• SpatialAgent - [Python] - An autonomous AI agent for spatial biology
• ChatSpatial - [Python] - MCP server enabling spatial transcriptomics analysis via natural language, integrating 60+ methods including SpaGCN, Cell2location, LIANA+, CellRank for Visium, Xenium, MERFISH | Paper | Docs | PyPI
• STAgent - [Python] - Autonomous multimodal LLM agent (Claude/GPT/Gemini) for end-to-end ST analysis of .h5ad data, with code generation, visual reasoning over tissue images, literature retrieval, and report synthesis via a Streamlit interface | Paper
• LazySlide - [Python] - Framework for whole slide image (WSI) analysis
• pasta - [R] - Point pattern and lattice data analysis from Robinson lab
• rakaia - [JavaScript] - Scalable interactive visualization and analysis of spatial omics including spatial transcriptomics, in the browser (Website)
• semla - [R] - Useful tools for Spatially Resolved Transcriptomics data analysis and visualization | Docs
• sosta - [Python] - Spatial Omic Structure Analysis
• SPATA2 - [R] - Spatial transcriptomics analysis toolkit
• spatial-omics-tutorials - [Python/R] - Tutorials and best-practices for spatial omics data analysis from BIML 2025
• MGC-BioSB-Spatial-Omics-Analysis-2025 - [Python] - Workshop materials for spatial omics analysis
• SMINT - [Python] - Spatial Multi-Omics Integration Toolkit for transcriptomics and metabolomics | Docs
• Thor - [Python] - Comprehensive platform for cell-level analysis with anti-shrinking Markov diffusion and 10 modular tools paired with Mjolnir web interface
• VR-Omics - [GUI] - Free platform-agnostic software with end-to-end automated processing of multi-slice spatial transcriptomics data through biologist-friendly GUI | Windows | MacOS | GitHub
• CosMx-Analysis-Scratch-Space - [R/Python] - Analysis resources and tools for CosMx SMI spatial transcriptomics | GitHub
• SpaceSequest - [R] - Unified pipeline for analysis, visualization, and publication of spatial transcriptomics data from Visium, Visium HD, Xenium, GeoMx, and CosMx | Tutorial
• VST-DAVis - [R Shiny] - Browser-based GUI for end-to-end Visium HD spatial transcriptomics analysis including QC, clustering, cell annotation, pathway enrichment, CellChat, and trajectory analysis
• BrainConnect - [Python] - Integrative analysis of mouse brain connectivity and whole-brain spatial transcriptomics using LSTM networks to predict connectivity strength from regional gene expression
• stLearn - [Python] - Combines expression, spatial distance and tissue morphology for clustering, pseudo-time-space trajectories and cell-cell interaction
• gsMap - [Python] - Integrates GWAS summary statistics with spatial transcriptomics to map cells and spatial regions associated with human complex traits
• VoltRon - [R] - Multi-resolution, multi-omic spatial toolbox with built-in image registration across assays and resolutions
• spatialLIBD - [R] - Bioconductor package and Shiny app for interactive visualization of Visium and spot-based spatial data

Nextflow / Pipelines

• nf-core/spatialaxe - [Nextflow] - Nextflow pipeline for Xenium and Artera spatial transcriptomics analysis (renamed from spatialxe)
• nf-core/sopa - [Nextflow] - Spatial Omics Pipeline Analysis (SOPA) for processing spatial transcriptomics data
• Allen Immunology Xenium Pipeline - [Web] - HISE platform pipeline for Xenium data processing
• SCALPEL - [Nextflow] - Allen Institute pipeline for large-scale ST atlas construction with 3D segmentation, doublet detection (SOLO), MapMyCells label transfer, and CCF registration
• NNclinSSOAP - [Nextflow/R] - GxP-ready clinical pipeline for 10x Xenium spatial transcriptomics and scRNA-seq analysis with Docker/Apptainer containerization
• nf-core/spatialvi - [Nextflow] - nf-core pipeline for Visium data: spot counts, spatial coordinates and image data through QC, normalization and clustering
• WebAtlas - [Python/Nextflow] - Converts h5ad, SpaceRanger, Xenium and MERSCOPE output into Zarr + OME-TIFF for browser-based Vitessce atlases | Paper

Viewers & Interactive Annotation

• QuPath - [Java] - De-facto open-source whole-slide image viewer and annotator: ROI drawing, cell detection, stain deconvolution, scripting, Xenium/Visium image overlays | Paper
• InstanSeg - [Python] - Embedding-based nucleus and cell segmentation, shipped as the official QuPath extension; faster and more accurate than Cellpose/StarDist on six public datasets | Paper
• napari-spatialdata - [Python] - Desktop napari viewer and annotator for spatial data: multi-scale images, transcripts and polygons with a live scanpy round-trip
• TissUUmaps - [JavaScript/Python] - GPU-accelerated browser viewer for 10^7+ transcripts over multi-resolution tissue images; Jupyter-embeddable, with in-situ decoding QC modules | Paper

Probe & Panel Design

Panel Selection

• Spapros - [Python] - Gene panel selection for targeted spatial transcriptomics (Visium HD, Xenium, MERFISH, etc.) from an scRNA-seq reference, optimizing for cell-type ID and within-cell-type variation
• scGPD - [Python] - Deep-learning gene panel design for targeted ST; a two-stage, gene-gene correlation-aware binary gating mechanism selects compact, nonredundant marker panels from an scRNA-seq reference | Paper
• ReconST - [Python] - Gated-autoencoder panel design that selects the gene subset best reconstructing the whole transcriptome (rather than relying on known markers) for MERFISH/seqFISH/Xenium/MERSCOPE | Paper

Probe Sequence Design (Visium / Visium HD / FFPE)

• gene2probe - [Python] - Pipeline for designing custom probes against human genes, tuned to VisiumHD / Visium FFPE constraints
• ProbeST - [Snakemake] - Custom Visium-compatible probe design pipeline for any species (including prokaryotes and host–pathogen)

Off-target QC

• Off-Target Probe Tracker - [Python] - Pipeline to predict off-target binding from probe sequences, useful for auditing Xenium and other targeted ST panels

Analysis Pipeline Steps

ROI Selection

• S2Omics - [Python] - Designing smart spatial omics experiments with S2Omics
• SOFisher - [Python] - Reinforcement-learning agent that chooses the next field-of-view position from previously sampled FOVs, cutting the acquisition needed to reach regions of interest | Paper

QC

• SpaceTrooper - [R] - Quality control for spatial transcriptomics
• GrandQC - [Python] - Comprehensive solution for quality control in digital pathology
• SpotSweeper - [R] - Spatially aware quality control for spatial transcriptomics
• MerQuaCo - [Python] - A computational tool for quality control in image-based spatial transcriptomics
• SpatialQM - [R] - Standardized QC-metric suite for imaging-based ST (Xenium/CosMx/MERSCOPE): transcripts/cell, global FDR, signal-to-noise, Moran's I and more; the software arm of the Spatial Touchstone reproducibility framework | Paper
• SpatialArtifacts - [R] - Identification and classification of spatial artifacts (edge and interior) in Visium and Visium HD data via a two-step outlier-detection plus image-processing workflow | GitHub
• destriping-GLM - [Python] - Removes Visium HD stripe artifacts by GLM modelling of per-bin nuclear counts | Preprint

Normalization

• Cell volume normalization - [R] - Recommended for imaging-based techniques, especially with small probe lists
• SpaNorm - [R] - First spatially-aware normalization method that concurrently models library size effects and underlying biology | Bioconductor
• TranspaceR - [R] - Statistical framework for imaging-based (Xenium/CosMx) and Visium HD ST: novel QC, scalable spatially-variable-gene selection, and data-driven optimal normalization plus low-dimensional embedding | Paper

Gene Imputation & Denoising

• Note: Gene imputation is not recommended for deconvolution tasks
• SpaGE - [Python] - Spatial gene expression prediction with best overall performance
• SpaGCN - [Python] - Spatial graph convolutional network for gene correlation analysis
• Tangram - [Python] - Transcript distribution prediction and spatial mapping
• SpaOTsc - [Python] - Spatial imputation via optimal transport
• Seurat integration workflow - [R] - Transfer gene expression from scRNA-seq reference
• Sprod - [Python] - Spatial denoising method
• TISSUE - [Python] - Transcript imputation with spatial single-cell uncertainty estimation
• SpaIM - [Python] - Single-cell Spatial Transcriptomics Imputation via Style Transfer
• Cellpin - [Python] - Lightweight probabilistic model for reference-based imputation and denoising of spatial transcriptomes from an scRNA-seq reference, enabling transcriptome-wide imputation and atlas-to-spatial label transfer for targeted-panel and full-transcriptome data | Paper

Bias Correction

• ResolVI - [Python] - Bias correction method
• Statial - [R] - Correction of spill-over effects
• ovrl.py - [Python] - A python tool to investigate vertical signal properties of imaging-based spatial transcriptomics data
• SPLIT - [R] - SPLIT effectively resolves mixed signals and enhances cell-type purity
• cellAdmix - [R] - From Kharchenko lab - Evaluating and correcting cell admixtures in imaging-based spatial transcriptomics data.
• DenoIST - [R] - Denoising Image-based Spatial Transcriptomics data
• MisTIC - [Python] - A probabilistic model for correcting mis-assigned transcripts due to cell segmentation errors
• TRACER - [Python] - Tissue Reconstruction via Associative Clique Extraction and Relation-mapping
• RESCUE - [R] - Negative-selection method that partitions ST expression into reference-explained "canonical" and sparse "idiosyncratic" components, recovering biological signal (fragile cell types, neurites, extracellular transcripts) lost by reference-based deconvolution/segmentation | Paper
• XeniumClean - [R] - Spatial neighbour-aware transcript cleanup for imaging-based ST (Xenium, MERSCOPE, CosMx, Atera): removes biologically implausible transcripts by combining a single-cell RNA-seq reference with spatial neighbourhood information to erase genes that cannot plausibly originate from a cell's own type
• SPARKLE - [Python] - Evidence-constrained kernel estimator that removes local ambient RNA leakage in imaging-based ST | Preprint
• CLEAR-ST - [Python] - Physics-informed probabilistic decontamination modelling mRNA lateral diffusion between neighbouring cells | Preprint
• DeSpotX - [Python] - Identifiability-based deep generative decontamination of single-cell-resolution ST | Preprint

Cell Segmentation

Imaging-based Segmentation

• Baysor - [Julia] - Bayesian segmentation of spatial transcriptomics data
• Cellpose - [Python] - Generalist algorithm for cellular segmentation
  • Cellpose 3 - With supersampling/restoration capabilities
  • Cellpose-SAM - Cell and nucleus segmentation with superhuman generalization, works in 3D with various image conditions
• DeepCell - [Python] - Deep learning library for single cell analysis
• Bo Wang's method - [Python] - Better than SOTA segmentation (Nature Methods 2024)
• Proseg - [Rust] - Probabilistic segmentation method
• ComSeg - [Python] - Transcript-based point cloud segmentation
• FICTURE - [Python] - Feature-based image segmentation
• Xenium cell boundary - [Web] - Alternative when interior staining fails
• Bioimage.io - [Web] - Repository of AI models for segmentation
• ST-cellseg - [Python] - Segmentation for spatial transcriptomics
• CelloType - [Python] - Cell type detection and segmentation
• SAINSC - [Python] - Segmentation for sequencing-based spatial data
• BIDCell - [Python] - Biologically-informed deep learning for subcellular spatial transcriptomics segmentation
• FastReseg - [R] - Using transcript locations to refine image-based cell segmentation results
• Segger - [Python] - Fast and accurate cell segmentation of imaging-based spatial transcriptomics data
• Bering - [Python] - Graph deep learning for joint noise-aware cell segmentation and molecular annotation in 2D and 3D spatial transcriptomics
• STP - [Python] - Single-cell Partition for subcellular spatially-resolved transcriptomics integrating data with nuclei-stained images
• Deep learning-based segmentation - [Python] - Extensively trained nuclear and membrane segmentation models for precise transcript assignment in CosMx SMI data
• CellSAM - [Python] - Foundation model for cell segmentation achieving state-of-the-art performance across cellular targets (bacteria, tissue, yeast, cell culture) and imaging modalities (brightfield, fluorescence, phase, multiplexed) | Paper | Web App
• DISSECT - [Python] - Diffusion-based cell segmentation combining cytological image segmentation with transcriptome-guided boundary refinement for Xenium, CosMx, and Stereo-seq | PyPI

Segmentation-free methods:

• SSAM - [Python] - Subcellular segmentation-free analysis by multidimensional mRNA density (repo archived 2024-10, reference implementation)
• Points2Regions - [Python] - Transcript-based region identification without segmentation
• Cellist - [Python] - Multi-modal segmentation combining image and expression signal, benchmarked across Stereo-seq, Seq-Scope, seqFISH+, STARmap and Xenium | Paper
• RNA2seg - [Python] - Generalist segmentation model trained on 4M MERFISH/CosMx cells, fusing RNA point clouds with membrane and nuclear stains; zero-shot capable | Paper

VisiumHD Segmentation

• Bin2Cell - [Python] - Segmentation for VisiumHD data
• ENACT - [Python] - Enhanced accuracy for VisiumHD segmentation
• STHD - [Python] - Cell annotation for VisiumHD
• Segmentation and annotation pipeline for VisiumHD with Proseg and Novae - [Python] - Tutorial with a proposed VisiumHD pipeline using Proseg and Novae
• SMURF - [Python] - Soft-segmentation and manifold-unrolling framework that assigns 2 µm Visium HD bins to StarDist/Cellpose nuclei (fractionally distributing transcripts to optimize per-cell cluster similarity) and "unrolls" cells onto Cartesian coordinates for single-cell reconstruction | Paper | PyPI

Cell Annotation

• STEM - [Python] - Cell type annotation method
• TACIT - [Python] - Automated cell type identification
• moscot - [Python] - Optimal transport-based cell mapping
• CELLama - [Python] - Cell annotation model
• TACCO - [Python] - Transfer of annotations between single-cell datasets
• Tangram - [Python] - Label transfer by mapping an scRNA-seq reference onto spatial coordinates (also listed under Gene Imputation for its expression-prediction use)
• MMoCHi - [Python] - Cell annotation method
• CytoSPACE - [Python] - High-resolution alignment of single-cell and spatial transcriptomes
• ABCT - [R] - Anchor-based Cell Typer
• STELLAR - [Python] - Annotation of spatially resolved single-cell data with STELLAR
• Vesalius - [R] - Multi-scale and multi-context interpretable mapping of cell states across heterogeneous spatial samples
• STALocator - [Python] - ST-Aided Locator using deep learning to localize cells from single-cell RNA-seq data onto tissue slices
• CMAP - [Python] - Cellular Mapping of Attributes with Position, maps large-scale individual cells to precise spatial locations using divide-and-conquer strategy
• TransST - [Python] - Transfer learning framework leveraging cell-labeled information from external sources for cell-level heterogeneity inference | GitHub
• InSituType - [R] - Cell typing for CosMx SMI spatial transcriptomics
• HieraType - [R] - Hierarchical cell typing using RNA + protein for CosMx SMI
• CosMx-Cell-Profiles - [R] - Collection of reference datasets for CosMx SMI
• GARDEN - [Python] - Graph-based dynamic attention framework for identifying rare pathogenic cell populations (disease-driving cells often missed by standard methods), enables 3D tissue reconstruction
• Spatial-ID - [Python] - Supervision-based cell typing for high-throughput cell-level SRT via transfer learning from scRNA-seq + spatial embedding | Paper
• HiCAT - [Python] - Hierarchical atlas-guided annotation transfer for cohort-scale spatial omics | Preprint

Cell Deconvolution

• RCTD - [R] - Robust cell type decomposition
• rctd-py - [Python] - Python reimplementation of the RCTD algorithm with GPU acceleration
• Cell2location - [Python] - Mapping scRNA-seq to spatial data
• SPOTlight - [R] - Seeded NMF regression to deconvolute spatial spots
• CARD - [R] - Spatially informed cell-type deconvolution
• FlashDeconv - [Python] - Atlas-scale spatial deconvolution via structure-preserving sketching with linear O(N) scalability | Paper | PyPI
• UCASpatial - [R] - Ultra-precision spatial deconvolution using entropy-based weighting of cell-identity genes to robustly map low-abundance and transcriptionally heterogeneous cell subpopulations | Paper | Docs
• RETROFIT - [R] - Bayesian reference-free deconvolution requiring no single-cell reference or marker genes; effective down to Visium HD near-single-cell resolution | Paper | Bioconductor
• SpaCET - [R] - Reference-free tumour-aware deconvolution that infers malignant cell fractions from CNV signal | Paper
• spacedeconv - [R] - Unified interface to 30+ spatial deconvolution tools behind one input/output contract

Differential Expression

• C-SIDE - [R] - Cell type-Specific Inference of Differential Expression in spatial transcriptomics
• Niche-DE - [R] - Niche-differential gene expression analysis identifying context-dependent cell-cell interactions
• smiDE - [R] - Spatial differential expression method | GitHub
• spatialGE - [R] - Spatial gene expression analysis
• Vespucci - [R] - Prioritize spatial regions involved in the response to an experimental perturbation in spatial transcriptomics
• CSDE - [Python] - Corrected Spatial Differential Expression using Prediction-Powered Inference to account for preprocessing uncertainties (segmentation, quantification, cell typing)
• SpNeigh - [R] - Boundary- and gradient-aware spatial differential expression for high-resolution ST (Xenium, MERFISH, Visium HD): neighborhood extraction, distance-weighted/spline DE, and spatial enrichment scoring | Paper
• SpaceMarkers - [R] - Infers interaction-driven molecular changes from overlapping latent-space patterns: genes that change because two cell populations are adjacent | Paper

Spatially Variable Genes

• PROST - [Python] - Detection of spatially variable genes
• SpatialDE - [Python] - Spatial differential expression analysis
• SPARK-X - [R] - Detection of spatially variable genes, best performing
• Hotspot - [Python] - Identify informative gene modules with lowest false positive rate
• SOMDE - [Python] - Self-organizing map for spatially variable gene detection with optimization
• trendsceek - [R] - Identification of spatial expression trends
• nnSVG - [R] - Scalable identification of spatially variable genes using nearest-neighbor Gaussian processes
• SLOPER - [Python] - Score-based learning of Poisson-modeled expression rates for spatial gene modules and tissue organization patterns
• FlashS - [Python] - Frequency-domain Gaussian kernel testing for SVG detection, where expression sparsity accelerates rather than hinders computation | PyPI
• MERINGUE - [R] - Spatial autocorrelation and cross-correlation analysis robust to non-uniform cell density | Paper

Integration

• PRECAST - [R] - Probabilistic embedding, clustering, and alignment for integrating spatial transcriptomics data
• MISO - [Python] - MultI-modal Spatial Omics for versatile feature extraction and clustering (Coleman Lab)
• SpatialMETA - [Python] - Joint analysis of spatial transcriptomics and metabolomics via CVAE | Docs
• pyWNN - [Python] - Weighted Nearest Neighbors (WNN) implementation for Scanpy
• MOFA-FLEX - [Python] - Factor model framework for integrating omics data with prior knowledge | Docs
• SIMO - [Python] - Spatial Integration of Multi-Omics through probabilistic alignment integrating spatial transcriptomics with multiple single-cell modalities
• GSI - [Python] - Gene Spatial Integration using deep learning with representation learning to extract spatial distribution of genes | GitHub | Zenodo
• SPACE-seq - [Paper] - Unified molecular approach for spatial multiomics enabling simultaneous analysis of chromatin accessibility, mitochondrial DNA mutations, and gene expression on standard 10× Genomics Visium CytAssist platform
• LLOKI - [Python] - Cross-platform spatial transcriptomics integration using optimal transport and scGPT foundation models for unified features across different gene panels (RECOMB 2025)
• LYNX - [Python] - Deep generative model integrating paired spatial multi-modal data (RNA, protein, metabolomics, H&E) from adjacent sections to infer microenvironmental gradients and cell-state transitions | Docs
• INSPIRE - [Python] - Adversarial GNN plus NMF integration of many ST datasets, yielding interpretable spatial factors and gene programs; scales to Stereo-seq | Paper
• SpaMosaic - [Python] - Mosaic integration of spatial multi-omics with only partially overlapping modalities, with missing-modality imputation | Paper

Cell Niches & Tissue Domains

(Smaller) Cell types → Cell modules/neighborhoods → Niches/tissue domains (Larger)

• BANKSY - [R/Python] - Unified cell typing and tissue domain segmentation
• CellCharter - [Python] - Hierarchical niche detection
• SpatialGLUE - [Python] - Multi-omics cell niche identification
• smoothclust - [R] - Spatial clustering
• SpaTopic - [R] - Spatial topic modeling
• hdWGCNA - [R] - Weighted gene correlation network analysis
• GASTON - [Python] - Graph-based spatial domain detection
• PAST - [Python] - Prior-based self-attention method for spatial transcriptomics tissue domain identification
• SpatialMNN - [R] - Identification of shared niches between slides
• NicheCompass - [Python] - End-to-end analysis of spatial multi-omics data
• Proust - [Python] - Spatial domain detection using contrastive self-supervised learning for spatial multi-omics technologies (multi-modal domains)
• STAMP - [Python] - Spatial Transcriptomics Analysis with topic Modeling, provides interpretable dimension reduction through deep generative modeling discovering tissue domains and cellular communication patterns
• DeepGFT - [Python] - Combines deep learning with graph Fourier transform for spatial domain identification | GitHub | Zenodo
• SpatialFusion - [Python] - A lightweight multimodal foundation model for pathway-informed spatial niche mapping
• SpaHDmap - [Python] - High-definition spatial embedding integrating expression NMF with histology image encoder-decoder for spatial domain detection at enhanced resolution
• SpatialEcoTyper - [R] - Discovers and recovers spatially distinct multicellular communities from ST, scRNA-seq, and bulk data | Paper | Docs
• STAGATE - [Python] - Graph-attention autoencoder for spatial domain identification; standard benchmark comparator (upstream inactive since 2022) | Paper
• GraphST - [Python] - Graph self-supervised contrastive learning for spatial domains, vertical/horizontal integration and deconvolution | Paper
• lisaClust - [R] - Clusters tissue into spatial regions using local indicators of spatial association between cell types
• SpaNiche - [R] - Graph-regularized joint NMF over cell abundance and ligand-receptor expression to call colocalization patterns and multi-sample ecotypes | Paper

Cell Distances & Neighborhood

• CRAWDAD - [R] - Cell relationship analysis with directional adjacency distributions
• HoodscanR - [R] - Neighborhood analysis
• SpicyR - [R] - Spatial analysis in R
• MISTy - [R] - Explainable multiview framework for dissecting spatial relationships from highly multiplexed data
• SpatialCorr - [Python] - Identifying gene sets with spatially varying correlation structure
• CatsCradle - [R] - Spatial analysis framework for tissue neighbourhoods
• RIPPLE - [R] - Replicate-aware detection of cell-type-anchored proximity gradients, with GPU permutation testing | Preprint

Spatial Trajectories

• spaTrack - [Python] - Spatial trajectory analysis
• scSpace - [Python] - Reconstruction of cell pseudo-space from single-cell RNA sequencing data
• SOCS - [Python] - Accurate trajectory inference in time-series spatial transcriptomics with structurally-constrained optimal transport
• STORIES - [Python] - Spatiotemporal Reconstruction Using Optimal Transport for cell trajectory inference from spatial transcriptomics profiled at multiple time points
• ONTrac - [Python] - Ordered Niche Trajectory Construction

Cell-Cell Communication

• Spatia - [Python] - Spatial cell-cell interaction analysis
• CellAgentChat - [Python] - Agent-based cell communication modeling
• SpaTalk - [R] - Knowledge-graph-based cell-cell communication inference
• SpaOTsc - [Python] - Infers signalling relationships between cells via optimal transport (also listed under Gene Imputation for its spatial-mapping use)
• MISTy - [R] - Multi-view modelling of ligand-receptor and pathway views to attribute expression to intercellular signalling (also listed under Cell Distances for its neighbourhood views)
• DeepLinc - [Python] - De novo reconstruction of cell interaction landscapes
• CellChat - [R] - Inferrence of cell-cell communication from multiple spatially resolved transcriptomics datasets
• COMMOT - [Python] - Screening cell-cell communication in spatial transcriptomics via collective optimal transport
• NicheNet - [R] - Linking ligands to downstream target gene regulation
• DeepTalk - [Python] - Single-cell resolution cell-cell communication using deep learning
• CellNEST - [Python] - Cell–cell relay networks using attention mechanisms on spatial transcriptomics
• FlowSig - [Python] - Inferring pattern-driving intercellular flows from single-cell and spatial transcriptomics
• CytoSignal - [R] - Detects locations and dynamics of ligand-receptor signaling at single-cell resolution from spatial transcriptomic data (VeloCytoSignal captures temporal signaling velocity) | Paper
• LIANA+ - [Python] - All-in-one CCC framework: 20+ ligand-receptor methods, consensus ranking, spatially-aware bivariate scores and MOFA-based multi-condition analysis | Paper
• SpatialDM - [Python] - Bivariate Moran's I on ligand-receptor pairs with an analytical null, giving per-spot interaction calls | Paper
• FineST - [Python] - Fuses histology foundation-model features with ST for nuclei-resolved imputation and ligand-receptor interaction discovery | Paper
• MaskTalk - [Python] - Cell-identity-gated spatial lag model for target-aware cell-cell communication at single-cell resolution | Preprint

Metacells & Scalability

• SuperSpot - [R] - Metacell analysis for spatial data
• SEraster - [R] - Rasterization method for spatial data processing

Subcellular Analysis

• Sprawl - [Python] - Subcellular transcript localization
• Bento - [Python] - Python toolkit for subcellular analysis of spatial transcriptomics data
• FISHfactor - [Python] - Analysis of subcellular transcript patterns
• InSTAnT - [Python] - Intracellular spatial transcript analysis
• troutpy - [Python] - Analysis of transcripts outside segmented cells in spatial transcriptomics data
• smoppix - [R] - Nonparametric probabilistic-index tests for uni- and bivariate single-molecule localization patterns, avoiding segmentation and density estimation | Paper
• pyTrance - [Python] - Finds co-localizing RNAs in subcellular imaging-based ST via latent embeddings | Preprint

Copy Number Variations

• CalicoST - [Python] - CNV detection in spatial data
• inSituCNV - [Python] - Inference of Copy Number Variations in Image-Based Spatial Transcriptomics
• SpaCNA - [R] - Spatially aware CNA detection using morphology-matched neighbour aggregation and a hidden Markov random field; supports 3D | Paper
• fastCNV - [R] - Fast putative CNV detection in single-cell and spatial transcriptomics data
• SpatialInferCNV - [R] - Clone calling from Visium in cancer; underpins Erickson et al., Nature 608:360 (reference implementation, inactive since 2022)
• SPICE - [R] - Calls somatic copy-number events from spatially resolved transcriptomics | Preprint

Isoform Analysis

• SPLISOSM - [Python] - Spatial isoform statistical modeling for detecting isoform-resolution patterns (alternative splicing, polyadenylation) from spatial transcriptomics data | Paper | Docs
• Sicelore-2.1 - [Java] - Barcode and UMI assignment plus isoform quantification for Nanopore long-read Visium; produces the isoform-level spatial counts that downstream tests consume | Paper

Transcription Factors & Gene Regulatory Networks

• STAN - [R] - Spatial transcription factor analysis
• SpaGRN - [Python] - Spatially aware GRN inference using ligand-receptor-mediated spatial co-expression (bivariate Moran/Geary) on Stereo-seq and imaging ST

Technical Enhancements

Slide Alignment

• PASTE/PASTE2 - [Python] - Probabilistic alignment of spatial transcriptomics experiments
• SPIRAL - [R] - Integrating and aligning spatially resolved transcriptomics data across different experiments, conditions, and technologies
• TOAST - [Python] - Topography Aware Optimal Transport for Alignment of Spatial Omics Data
• STalign - [Python] - Alignment of spatial transcriptomics data using diffeomorphic metric mapping (JEFworks Lab)
• SPCoral - [Python] - Spatial multi-modal alignment and integration
• SPOmiAlign - [Python] - Multi-modal spatial alignment and integration for transcriptomics and metabolomics
• MALDI-MSI Overlay - [Python] - Script for co-registration of MALDI-MSI and spatial transcriptomics from One Slide Two Worlds
• SpaMTP - [R] - Spatial multi-task prediction and alignment
• SANTO - [Python] - A coarse-to-fine alignment and stitching method for spatial omics
• 3d-OT - [Python] - Geometry-aware heterogeneous slice alignment of spatial multi-omics via soft-correspondence optimal transport, handling nonrigid deformation | Paper
• VALIS - [Python] - Automated rigid and non-rigid registration of brightfield/IF whole-slide image series and 3D serial-section reconstruction; registers the images rather than the coordinates | Paper

Super Resolution

• TESLA - [Python] - Super resolution for 10X Visium
• istar - [Python] - Super resolution for Visium
• BayesSPACE - [R] - Subspot resolution
• Spotiphy - [Python] - Super resolution tool for spatial data

Transcripts + Histology

• ST-Net - [Python] - Integrating spatial gene expression and tumor morphology via deep learning
• SpaceDIVA - [Python] - Integration of transcript data with histological images
• HEST - [Python] - Dataset for Spatial Transcriptomics and Histology Image Analysis
• CellLENS - [Python] - Cell Local Environment Neighborhood Scan
• DeepSpot - [Python] - Leveraging Spatial Context for Enhanced Spatial Transcriptomics Prediction from H&E Images | Paper
• SpatialWhisperer - [Python] - Trimodal contrastive foundation model embedding H&E images, spatial transcriptomes, and text into a shared 512-D space for zero-shot histopathology annotation and cross-modal retrieval (ICML 2026) | HuggingFace
• DeepSpotM - [Python] - Multimodal foundation model predicting transcriptome-wide (~19k genes) virtual ST from H&E tiles via a LoRA-adapted pathology backbone with a cross-attention gene decoder and a gene router hypernetwork that draws frozen embeddings from DNA/RNA/protein/single-cell/text FMs (Evo2, Orthrus, ProtT5, scGPT, Apertus), enabling zero-shot prediction of unseen genes | Paper | HuggingFace
• DeepSpot2Cell - [Python] - Predicting virtual single-cell spatial transcriptomics from H&E images using spot-level supervision (NeurIPS 2025 Imageomics) | Paper
• SpotWhisperer - [Python] - Molecularly informed analysis of histopathology images using natural language
• STPath - [Python] - A Generative Foundation Model for Integrating Spatial Transcriptomics and Whole Slide Images
• STFlow - [Python] - Scalable generation of spatial transcriptomics from histology images via whole-slide flow matching
• AESTETIK - [Python] - Representation learning for multi-modal spatially resolved transcriptomics data (Bioinformatics) | Paper

Benchmarks

• Deconvolution benchmark - [Paper] - Comprehensive comparison
• RCTD and Cell2location benchmark - [Paper] - Claims these are the best methods
• Spatial clustering benchmark - [Paper] - Comparison of clustering methods
• Spatialbench - [Python] - Benchmark for evaluating AI agents on spatial biology analysis tasks
• Nature Communications review - [Paper] - Confirms Cell2location performance
• Open problems benchmark - [Web] - Cell2location is top performer
• Neighborhod benchmark - [Paper] - New COZI method top performer
• Kaiko.ai FM benchmark EVA - [Python] - WSI benchmark
• Benchmarking of spatial transcriptomics platforms across six cancer types - [Paper] - Comprehensive platform comparison
• PathBench - [Python] - Pathology benchmark
• SPATCH Benchmark - 2025 - [Paper] - Showing Xenium performs best
• Thunder - [Python] - Pathology benchmark
• Histoboard - [Web] - Pathology leaderboards
• Xenium_benchmarking - [Python] - Independent benchmark and best-practice analysis workflows for Xenium across 25 datasets: QC, segmentation (Baysor+Cellpose), preprocessing, SVG selection, gene imputation, and domain identification (Nature Methods 2025) | Paper
• SACCELERATOR - [Python] - Community benchmarking and consensus framework for spatially aware clustering across 22 methods, >170 samples and 8 platforms; argues manual anatomical labels are biased and unsuitable as ground truth | Paper
• spDDB - [Python] - Benchmarks 21 deconvolution and 18 domain-detection methods across 37 datasets and 5 platforms | Preprint
• Benchmarking-CCI - [R] - Nine spatial cell-cell interaction methods over simulations and nine datasets (Visium, Stereo-seq, Xenium); no method is optimal across resolutions, and rankings flip with the ligand-receptor database | Paper
• Cross-platform deconvolution benchmark - [Python] - Deconvolution under cross-platform technological bias; SpatialDecon and cell2location come out ahead | Paper
• ctSVGbench - [R] - Six cell-type-specific SVG methods over 46 real and 666 simulated datasets, with rotation and null controls; Celina best overall | Paper
• BEASTsim - [Paper] - Benchmarks four ST simulators across five datasets; SRTsim best reference-based, scDesign3 best reference-free
• Imputation benchmark - [Paper] - Seven imputation methods over 23 datasets and five platforms; the winner is platform-dependent (gimVI on Visium, MAGIC on Stereo-seq, Tangram on Slide-seqV2)
• Segmentation error impact - [Paper] - Shows segmentation error confounds differential expression, neighbour influence and ligand-receptor inference, frequently dominating the result; Proseg recommended, cellAdmix corrects

Simulators & Ground Truth

• scDesign3 - [R] - Unified simulator for single-cell and spatial omics with interpretable parameters; best reference-free simulator in the BEASTsim benchmark | Paper
• scCube - [Python] - De-novo simulation of spatial transcriptomes with controllable spatial patterns and resolutions | Paper
• SRTsim - [R] - Reference-based spatially resolved transcriptomics simulator; top reference-based performer in the BEASTsim benchmark

Datasets & Foundation Models

Datasets

• HISSTA - [Python/R] - Histopathology spatial transcriptomics dataset
• STOmicsDB - [Web] - Spatial transcriptomics database
• STHELAR - [Python] - Multi-tissue dataset linking spatial transcriptomics (Xenium) and histology for cell-type annotation
• DeepSpaceDB 2.0 - [Web] - Interactive web database for large-scale Xenium exploration (628 public datasets, ~1,045 samples / 129M cells) plus Visium, stored in gene- and coordinate-chunked Zarr for sub-second browser-based expression and ROI queries | Paper
• TCGA virtual spatial transcriptomics atlas - [Python] - DeepSpot-M predicted transcriptome-wide ST for TCGA H&E (FF + FFPE); ~28.7k slides / 32 cancer types / ~296M spots (gated) | Paper
• HEST Xenium virtual spatial transcriptomics - [Python] - DeepSpot-M predicted transcriptome-wide ST for 59 HEST-1k 10x Xenium H&E slides (~13.3M cells; gated) | Paper

Foundation Models

Expression-Centric (Transcriptomics)

• novae - [Python] - Deep learning foundation model for spatial domain assignments and tissue organization analysis | Paper | Docs
• scGPT-spatial - [Python] - Spatial-omic foundation model pretrained on 30M spatial profiles (SpatialHuman30M) across 821 slides
• Nicheformer - [Python] - Transformer-based foundation model pretrained on SpatialCorpus-110M containing over 110 million cells for spatial composition and label prediction | GitHub
• stFormer - [Python] - Foundation model generating contextual gene representations within spatial niches with ligand–receptor aware attention
• TERRA - [Python] - Self-supervised foundation model producing cell- and neighborhood-level embeddings via a Graph Transformer + JEPA on masked gene tokens, reusable across downstream tasks without retraining | Docs

Visual-Omics & Multimodal (H&E + ST)

• Loki / OmiCLIP - [Python] - Visual–omics foundation model and platform bridging H&E with ST | Docs
• SpaFoundation - [Python] - Visual foundation model for spatial transcriptomics using histology images alone for gene expression inference and super-resolution
• ST-Align - [Python] - Multimodal foundation model for image–gene alignment in spatial transcriptomics
• spEMO - [Python] - Framework unifying embeddings from pathology foundation models and LLMs for spatial multi-omic analysis
• FOCUS - [Python] - Foundational generative model for cross-platform unified ST enhancement conditioned on H&E images, scRNA-seq references, and spatial co-expression priors (trained on 1.7M H&E-ST pairs, 10 platforms)
• TITAN - [Python] - A multimodal whole-slide foundation model for pathology
• SEAL - [Python] - Spatial Expression-Aligned Learning that fine-tunes pathology foundation models (UNI, CONCH) using spatial transcriptomics data for gene expression-image alignment (Mahmood Lab)
• Phoenix - [Python] - Pan-cancer virtual spatial transcriptomics from routine histology via latent flow matching generative model predicting single-cell gene expression directly from H&E images, generalizing across cohorts, donors, organs, and tissues

Pathology & Histology

• Virchow - [Python] - Foundation model for computational pathology
• UNI and UNI2 - [Python] - Universal pathology foundation models (Mahmood Lab)
• UNI (KatherLab) - [Python] - General-purpose self-supervised pathology foundation model
• Google Path Foundation - [Python] - Self-supervised embedding model for H&E histopathology
• CONCH - [Python] - Contrastive learning for histopathology
• GIGApath - [Python] - Large-scale pathology foundation model
• CHIEF - [Python] - Clinical Histopathology Imaging Evaluation Foundation Model for cancer detection and prognosis
• Phikon-v2 - [Python] - Spatial biology foundation model
• Bioptimus H-optimus-1 - [Python] - Latest biology-focused foundation model from Bioptimus
• Atlas 2 - [Python] - Foundation models for clinical deployment
• DeepCell dataset - [Web] - CNN + human features embeddings
• NuSPIRe - [Python] - Self-supervised nuclear-morphology model pretrained on 15.5M DAPI nuclei, for cell typing, perturbation detection and ROI/FOV selection | Paper

Proteomics

• KRONOS - [Python] - Foundation Model for Multiplex Spatial Proteomic Images