bioRxiv Science⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/2025.01.25.634867

Building Foundation Models to Characterize Cellular Interactions via Geometric Self-Supervised Learning on Spatial Genomics

You, Y.·Wang, Z.·Fleisher, K.·Liu, R.·Thomson, M.
View original source ↗Export metadata ↓

Abstract

Cellular interactions form the fundamental/core circuits that drive development, physiology, and disease within tissues. Advances in spatial genomics (SG) and artificial intelligence (AI) offer unprecedented opportunities to computationally analyze and predict the behavior of cell intricate networks, and to identify interactions that drive disease states. However, challenges arise in both methodology and scalability: (i) how to computationally characterize complicated cellular interactions of multi-scale nature where chemical genes/circuits in individual cells process information and drive interactions among large numbers of diverse cell types, and (ii) how to scale up the pipeline to accommodate the increasing volumes of SG data that map transcriptome-scale gene expression and spatial proximity across millions of cells. In this paper, we introduce the Cellular Interaction Foundation Model (CI-FM), an AI foundation model functioning to analyze and simulate cellular interactions within living tissues. In the CI-FM pipeline, we explicitly capture and embed cellular interactions within microenvironments by leveraging the powerful and scalable geometric graph neural network model, and optimize the characterization of cellular interactions with a novel self-supervised learning objective - we train it to infer gene expressions of cells based upon their interacting microenvironment. As a result, we construct CI-FM with 100 million parameters by consuming SG data of 23 million cells. Our benchmarking experiments show CI-FM effectively infers gene expressions conditional on the microenvironmental contexts: we achieve a high correlation and a low mismatch error (MSE of 1.1% relative to the square median expression), with 79.4% of cells on average being annotated as the similar cell type based on their predicted and actual expressions. We demonstrate the downstream utility of CI-FM by: (i) applying CI-FM to embed tumor samples to capture cellular interactions within tumor microenvironments (ROC-AUC score of 0.76 on classifying sample conditions via linear probing on embeddings), and identifying shared signatures across samples; and (ii) using CI-FM to simulate changes in microenvironmental composition in response to T cell infiltration, which highlights how CI-FM can be leveraged to model cellular responses to tissue perturbations - an essential step toward constructing "AI virtual tissues". Our model is open source and publicly accessible at https://huggingface.co/ynyou/CIFM.

Explore related subjects

bioinformaticsPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

You, Y., Wang, Z., Fleisher, K., Liu, R., Thomson, M.. 2025-01-27. Building Foundation Models to Characterize Cellular Interactions via Geometric Self-Supervised Learning on Spatial Genomics. https://doi.org/10.1101/2025.01.25.634867

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

spatialMET: an open and scalable framework for spatial metabolomics analysis

Mass spectrometry imaging (MSI) enables spatially resolved metabolomics in intact tissue sections, but analysis remains challenging at scale. Existing MSI workflows often require users to combine multiple software tools, while others rely on proprietary vendor software that limits interoperability and reproducibility. To address these challenges, we developed spatialMET, an open-source framework that provides an end-to-end workflow for MSI analysis. spatialMET provides a unified platform for preprocessing, spatial domain detection, and visualization. Downstream analyses include differential abundance testing, spatial autocorrelation and gradient analysis, dimensionality reduction, and correlation network analysis. Spatial domain detection uses hcdist, a C-based hierarchical clustering implementation that substantially reduces runtime and memory use relative to existing R-based approaches. spatialMET can be run through an interactive R Shiny application or as a standalone command-line workflow for larger datasets or high-performance computing environments. Applied to mouse small cell lung cancer MALDI-MSI data containing 284,673 pixels, spatialMET identified tumor-associated, stromal, and adjacent lung spatial domains that aligned with matched histology. Differential abundance analysis identified 117 m/z features that differed between tumor and stromal regions, while spatial autocorrelation analyses revealed spatially structured abundance patterns. Applying spatialMET to mouse lung adenocarcinoma data from an entire lung lobe containing 338,477 pixels further demonstrated scalability and captured spatial heterogeneity across tumor and surrounding lung tissue. In summary, spatialMET provides a scalable, open-source framework for end-to-end spatial metabolomics analysis, and it is distributed as a Docker container for reproducible deployment. Source code and installation instructions are available at https://github.com/biodatalab/spatialMET.

bioinformatics↗
bioRxiv2026

Probing the transcriptome response to shivering in skeletal muscle using a multilayered bioinformatics approach

Cold acclimation holds therapeutic potential for improving metabolic health. We previously demonstrated that repeated cold-induced shivering enhances insulin sensitivity in humans. However, the molecular pathways that underlie the skeletal muscle shivering response, and how these relate to beneficial physiological effects, remain poorly understood. In this study, we combined complementary bioinformatics approaches to allow in-depth analysis of the transcriptomic response of human skeletal muscle to repeated shivering. We identified a robust transcriptional signature and show a sex-specific component in the shivering skeletal muscle response, which seemed to diminish following cold adaptation. Our findings provide mechanistic insights into cold-induced muscle adaptations, shed light on potential interesting molecular targets for further investigation, and emphasize the importance of including both sexes in future cold acclimation studies.

bioinformatics↗
bioRxiv2026

An Information Geometry approach to model topological trajectories and Gene Expression Radius from UMAP geometry.

Understanding the relationship between gene expression dynamics and cellular identity remains a central challenge in single cell biology. Here, we introduce a novel computational and mathematical framework that integrates information geometry, fuzzy topology, and UMAP analysis to model gene expression landscapes derived from single cell RNA sequencing data. We formalize gene expression data as a fuzzy topological space, where interactions between expression points are governed by probabilistic distributions inspired by manifold learning approaches such as UMAP. Within this framework, we define an information geometric structure through a Fisher metric induced by these distributions, enabling the computation of geodesic trajectories that capture cellular differentiation processes. A key contribution of this work is the derivation of analytical conditions, expressed as expression radius formulas, that characterize local neighborhoods in gene expression space. These conditions allow for the identification of genes associated with stem cell states and predictions in transitional cell types in future work. Application of the proposed framework to single cell datasets reveals biologically meaningful gene sets enriched in key regulatory pathways and transcription factors, demonstrating the capacity of our approach to uncover latent structure in complex gene expression data. Our results suggest that integrating differential geometry with statistical learning theory offers a powerful paradigm for modeling genotype and phenotype relationships and cellular state transitions, with potential implications for precision medicine and systems biology.

bioinformatics↗