A spectral dimension reduction technique that improves pattern detection in multivariate spatial data

2026/01/28 by David Köhler, Niklas Kleinenkuhnen, Kiarash Rastegar +7 · 1 voice
Biochemistry, Genetics and Molecular Biology · Mathematics · #Single-cell and spatial transcriptomics #Gene expression and cancer classification #Statistical Methods and Inference

paper · doi:10.1093/bioinformatics/btag052

openalex publication_date 2026/01/28 · openalex created_date 2026/02/02 · openalex updated_date 2026/07/08

Abstract

MOTIVATION: We introduce a statistical approach for pattern recognition in multivariate spatial transcriptomics data. RESULTS: Our algorithm constructs a projection of the data onto a low-dimensional feature space which is optimal in maximizing Moran's I, a measure of spatial dependency. This projection mitigates non-spatial variation and outperforms principal components analysis for pre-processing. Patterns of spatially variable genes are well represented in this feature space, and their projection can be shown to be a denoising operation. Our framework does not require any parameter tuning, and it furthermore gives rise to a calibrated, powerful test of spatial gene expression. AVAILABILITY AND IMPLEMENTATION: The algorithm is implemented in the open source software R and is available at https://github.com/IMSBCompBio/SpaCo.

Citations

• NOTES ON CONTINUOUS STOCHASTIC PHENOMENA
• Integrated analysis of multimodal single-cell data
• Dictionary learning for integrative, multimodal and scalable single-cell analysis

Discussions

• SPACO, our tool for Spatial Component Analysis for Spatial Sequencing Data, is now available as R package on CRAN!🎉 Congratulations to the team!!! We also have a Docker image which is available on Ze [bsky, 3 points, 1 comments]

Related

• Nonnegative spatial factorization
• Categorization of 33 computational methods to detect spatially variable genes from spatially resolved transcriptomics data
• Spatial Transcriptomics Dimensionality Reduction using Wavelet Bases
• Spatially Regularized Gaussian Mixtures for Clustering Spatial Transcriptomic Data
• A Bayesian Modified Ising Model for Identifying Spatially Variable Genes from Spatial Transcriptomics Data