Cross-Modal Imputation and Uncertainty Estimation for Spatial Transcriptomics

Xiangyu Guo, Ricardo Henao
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, PMLR 258:4375-4383, 2025.

Abstract

High-resolution spatial transcriptomics (ST) technologies can capture gene expression at the cellular level along with spatial information, but are limited in the number of genes that can be profiled. In contrast, single-cell RNA sequencing (SC) provides more comprehensive gene expression profiles but lacks spatial context. To bridge these gaps, existing methods typically focus on single-modality prediction tasks, leveraging complementary information from the other modality. Here, we propose an attention-based cross-modal framework that simultaneously imputes gene expression for ST and recovers spatial locations for SC, while also providing uncertainty estimates for the expression of the imputed genes. Our approach was evaluated on three real-world datasets, where it consistently outperformed state-of-the-art methods in spatial gene profile imputation. Moreover, our framework enhances latent embedding integration between the two modalities, resulting in more accurate spatial position estimates.

Cite this Paper

BibTeX
@InProceedings{pmlr-v258-guo25a, title = {Cross-Modal Imputation and Uncertainty Estimation for Spatial Transcriptomics}, author = {Guo, Xiangyu and Henao, Ricardo}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {4375--4383}, year = {2025}, editor = {Li, Yingzhen and Mandt, Stephan and Agrawal, Shipra and Khan, Emtiyaz}, volume = {258}, series = {Proceedings of Machine Learning Research}, month = {03--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v258/main/assets/guo25a/guo25a.pdf}, url = {https://proceedings.mlr.press/v258/guo25a.html}, abstract = {High-resolution spatial transcriptomics (ST) technologies can capture gene expression at the cellular level along with spatial information, but are limited in the number of genes that can be profiled. In contrast, single-cell RNA sequencing (SC) provides more comprehensive gene expression profiles but lacks spatial context. To bridge these gaps, existing methods typically focus on single-modality prediction tasks, leveraging complementary information from the other modality. Here, we propose an attention-based cross-modal framework that simultaneously imputes gene expression for ST and recovers spatial locations for SC, while also providing uncertainty estimates for the expression of the imputed genes. Our approach was evaluated on three real-world datasets, where it consistently outperformed state-of-the-art methods in spatial gene profile imputation. Moreover, our framework enhances latent embedding integration between the two modalities, resulting in more accurate spatial position estimates.} }
Endnote
%0 Conference Paper %T Cross-Modal Imputation and Uncertainty Estimation for Spatial Transcriptomics %A Xiangyu Guo %A Ricardo Henao %B Proceedings of The 28th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2025 %E Yingzhen Li %E Stephan Mandt %E Shipra Agrawal %E Emtiyaz Khan %F pmlr-v258-guo25a %I PMLR %P 4375--4383 %U https://proceedings.mlr.press/v258/guo25a.html %V 258 %X High-resolution spatial transcriptomics (ST) technologies can capture gene expression at the cellular level along with spatial information, but are limited in the number of genes that can be profiled. In contrast, single-cell RNA sequencing (SC) provides more comprehensive gene expression profiles but lacks spatial context. To bridge these gaps, existing methods typically focus on single-modality prediction tasks, leveraging complementary information from the other modality. Here, we propose an attention-based cross-modal framework that simultaneously imputes gene expression for ST and recovers spatial locations for SC, while also providing uncertainty estimates for the expression of the imputed genes. Our approach was evaluated on three real-world datasets, where it consistently outperformed state-of-the-art methods in spatial gene profile imputation. Moreover, our framework enhances latent embedding integration between the two modalities, resulting in more accurate spatial position estimates.
APA
Guo, X. & Henao, R.. (2025). Cross-Modal Imputation and Uncertainty Estimation for Spatial Transcriptomics. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 258:4375-4383 Available from https://proceedings.mlr.press/v258/guo25a.html.

Related Material