A comparative study of statistical methods for identifying differentially expressed genes in spatial transcriptomics.

Spatial transcriptomics (ST) provides unprecedented insights into gene expression patterns while retaining spatial context, making it a valuable tool for understanding complex tissue architectures, such as those found in cancers. Seurat, by far the most popular tool for analyzing ST data, uses the Wilcoxon rank-sum test by default for differential expression analysis. However, as a nonparametric method that disregards spatial correlations, the Wilcoxon test can lead to inflated false positive rates and misleading findings. This limitation highlights the need for a more robust statistical approach that effectively incorporates spatial correlations. To this end, we propose a Generalized Score Test (GST) in the Generalized Estimating Equations (GEEs) framework as a robust solution for differential gene expression analysis in ST. We conducted a comprehensive comparison of the GST with existing methods, including the Wilcoxon rank-sum test and the GEEs with the robust Wald test. By appropriately accounting for spatial correlations, extensive simulations showed that the GST demonstrated superior Type I error control and comparable power relative to other methods. Applications to ST datasets from breast and prostate cancer showed that the GST-identified differentially expressed genes were enriched in pathways directly implicated in cancer progression. In contrast, the Wilcoxon test-identified genes were enriched in non-cancer pathways and produced substantial false positives, highlighting its limitations for spatially structured data. Our findings suggest that the GST approach is well-suited for ST data, offering more accurate identification of biologically relevant gene expression changes. We have implemented the proposed method in R package "SpatialGEE", available on GitHub.