Back

SPRI: Spatial Pattern Recognition using Information based method for spatial gene expression data

Hu, J.-X.; Yuan, Y.; Shen, H.-B.

2023-11-27 bioinformatics
10.1101/2022.02.09.479510 bioRxiv
Show abstract

The rapid development of spatially resolved transcriptomics has made it possible to analyze spatial gene expression patterns in complex biological tissues. To identify spatially differential expressed genes, we propose a novel and robust nonparametric information-based approach, SPRI. SPRI converts the problem of identifying spatial gene expression patterns into the detection of dependencies between spatial coordinates with observed frequencies measured by read counts. It directly models spatial transcriptome raw count without assuming a parametric model. SPRI was applied to spatial datasets with different resolutions, suggesting that SPRI outperforms previous methods, by robustly detecting more genes with significant spatial expression patterns, and revealing biological insights that cannot be identified by other methods.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above