Spatially Informed and Direction Sensitive Representation Learning for Spatial Transcriptomics
Abstract
Cellular states are shaped by both intrinsic gene expression programs and the spatial microenvironment in which cells reside. Spatial transcriptomics makes this context measurable, but effective representation learning requires capturing both spatially informative molecular signals and intercellular organization. Existing approaches address these challenges only partially: conventional gene selection can overlook spatially structured signals, while methods for encoding cell-cell relationships often either reduce a cell's microenvironment to proximity alone, discarding direction, or rely on absolute coordinates tied to an arbitrary tissue frame. Consequently, cells with similar neighbor distances but distinct spatial organization may receive overly similar representations, despite experiencing different biological contexts that shape their cellular states. We introduce Tessera, a spatial representation learning framework that addresses both molecular selection and cellular context. Tessera combines spatially informed gene selection with a geometric inductive bias that contextualizes each cell within its local tissue environment. By modeling relative cell-cell relationships, Tessera is translation-invariant while preserving direction, allowing neighboring cells to contribute differently according to their spatial context and cellular state. Across held-out tissues and technologies, Tessera outperforms existing spatial foundation models in representation quality and generalizes zero-shot to contexts up to 16× larger than those seen during training.
Related
- SpaCoEx: Sparse Gene Selection for Spatially Varying Co-expression in Spatial Transcriptomics
- PathwayFinder enables linked exploration of cell types and gene modules in spatially resolved transcriptomics data
- Covariance Nonstationarity is Evident in Spatial Transcriptomics and Provides a New Categorization of Spatially Varying Genes
- Direction-sensitive neural correlates of voluntary pain regulation: regional response attenuation during down-regulation and exploratory prefrontal-valuation coupling
- Spatially resolved host-fungal transcriptomics profiling by fungal-DBiT
BibTeX
@misc{amirabad2026spatially,
title = {{Spatially Informed and Direction Sensitive Representation Learning for Spatial Transcriptomics}},
author = {Azim Dehghani Amirabad and Ali Khodabandeh Yalabadi and Artem Moskalev and Yun-Ching Chen and Mehdi Pirooznia and Scott Oloff and Rui Liao},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.28579},
eprint = {2610.28579},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.28579},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Spatially Informed and Direction Sensitive Representation Learning for Spatial Transcriptomics
AU - Azim Dehghani Amirabad
AU - Ali Khodabandeh Yalabadi
AU - Artem Moskalev
AU - Yun-Ching Chen
AU - Mehdi Pirooznia
AU - Scott Oloff
AU - Rui Liao
PY - 2026
DA - 2026/10/20
PB - arXiv
DO - 10.48550/arxiv.2610.28579
UR - https://arxiv.org/abs/2610.28579
AB - Cellular states are shaped by both intrinsic gene expression programs and the spatial microenvironment in which cells reside. Spatial transcriptomics makes this context measurable, but effective representation learning requires capturing both spatially informative molecular signals and intercellular organization. Existing approaches address these challenges only partially: conventional gene selection can overlook spatially structured signals, while methods for encoding cell-cell relationships often either reduce a cell's microenvironment to proximity alone, discarding direction, or rely on absolute coordinates tied to an arbitrary tissue frame. Consequently, cells with similar neighbor distances but distinct spatial organization may receive overly similar representations, despite experiencing different biological contexts that shape their cellular states. We introduce Tessera, a spatial representation learning framework that addresses both molecular selection and cellular context. Tessera combines spatially informed gene selection with a geometric inductive bias that contextualizes each cell within its local tissue environment. By modeling relative cell-cell relationships, Tessera is translation-invariant while preserving direction, allowing neighboring cells to contribute differently according to their spatial context and cellular state. Across held-out tissues and technologies, Tessera outperforms existing spatial foundation models in representation quality and generalizes zero-shot to contexts up to 16× larger than those seen during training.
ER -