2026

G2T: Tissue Reconstruction from Gene Expression via Embedding-Distance Flow Matching

S. Birk, F. J. Theis, M. Lotfollahi

Machine Learning in Computational Biology (MLCB) 20262026Conference Proceedings

Learning Discrete Cell and Niche Codes from Spatial Transcriptomics Using Dual Residual Vector Quantization

S. Birk, A. Merchant, A. Vahidi, F. J. Theis, M. Lotfollahi

Machine Learning in Computational Biology (MLCB) 20262026Conference Proceedings

Multi-scale modeling of human tissues from spatial transcriptomics with TERRA

S. Birk, M. V. Sanian, A. Vahidi, S. Ogden, D. J. Jafree, A. Miraki Feriz, et al.

bioRxiv2026Preprint

A graph-transformer foundation model that reads tissue structure from spatial transcriptomics across scales, from single cells to whole tissues.

Graph Tokenization Meets JEPA: Self-Supervised Learning on Spatial Cell Graphs

S. Birk, A. Vahidi, M. V. Sanian, A. Merchant, M. Lotfollahi

ICML 2026 Workshop on Graph Foundation Models2026Workshop

SQUINT: Spatial Quantization for Understanding and IN-painting Tissues

A. Merchant, S. Birk, A. Vahidi, D. Jafree, L. Steele, A. R. Foster, V. Baskar, et al.

ICLR 2026 Workshop on Machine Learning for Genomics Explorations2026Workshop

Hidden immune memory niches in inflammatory skin diseases

L. Steele, A. R. Foster, K. Roberts, C. Admane, S. Birk, P. V. Mazin, A. Akbarnejad, C. Tudor, et al.

bioRxiv2026Preprint

2025

SIGMMA: Hierarchical Graph-Based Multi-Scale Multi-modal Contrastive Alignment of Histopathology Image and Spatial Transcriptome

D. Jeong, A. Vahidi, C. Ramírez-Suástegui, M. Moullet, K. Ly, M. V. Sanian, S. Birk, Y. Chang, A. Boxall, D. Jafree, L. Steele, V. Baskar MS, M. Haniffa, M. Lotfollahi

arXiv2025Preprint

Mapping and reprogramming human tissue microenvironments with MintFlow

A. Akbarnejad, L. Steele, D. J. Jafree, S. Birk, M. R. Sallese, K. Rademaker, A. Boxall, B. Rumney, C. Tudor, M. Patel, M. Prete, S. Makarchuk, C. Y. C. Lee, J. Maaskola, T. Li, H. Stanley, A. R. Foster, K. Roberts, A. L. Trinh, C. E. Villa, G. Testa, S. Mahil, A. Mehrjou, C. Smith, S. Vakili, M. R. Clatworthy, T. Mitchell, O. A. Bayraktar, M. Haniffa, M. Lotfollahi

bioRxiv2025Preprint

Disentangles the gene expression a cell owes to its microenvironment from what is intrinsic to it, making tissue niches something you can steer rather than only observe.

Quantitative characterization of cell niches in spatially resolved omics data

S. Birk, I. Bonafonte-Pardàs, A. Miraki Feriz, A. Boxall, E. Agirre, F. Memi, A. Maguza, A. Yadav, E. Armingol, R. Fan, G. Castelo-Branco, F. J. Theis, O. A. Bayraktar, C. Talavera-López, M. Lotfollahi

Nature Genetics 57(4), 897–9092025Journal

NicheCompass — a graph deep-learning method that learns interpretable cell-niche representations from spatial omics, grounded in known cell–cell communication programs.