Quantification of gene expression changes in mouse disease models using a high-throughput spatial omics platform

Download Poster

Batch effects due to technical variability are a major problem in single cell transcriptomics. Spatial methods are no exception – their low throughput requires high numbers of technical replicates, reducing the statistical power needed to quantify differential gene expression across experimental conditions.

To overcome this problem, we took advantage of the large imaging area of the Rebus Esper spatial omics platform. We processed brain sections from three mouse genotypes in parallel – one wild type and two disease models. In addition, the Esper High Fidelity assay, based on single-molecule fluorescent in situ hybridization (smFISH), requires no amplification, yielding quantitative results with minimal batch effects. The combination of low technical variation and balanced experimental design allowed us to integrate more than 500,000 cells from multiple datasets for analysis without the need for batch correction. We were able to identify more than 12 neuronal and glial cell type clusters using 20 cell type-specific genes, and further dissect these cell types by anatomical structures utilizing the spatial information. We then performed differential gene expression of 10 disease-related genes on each cell-type subset.

These results demonstrate the ability of the Rebus Esper spatial omics platform to yield high-throughput spatial omics data with single cell resolution, low technical variation, and the sensitivity and specificity required for differential gene expression quantification.