Empowering Innovation: Our Success Stories with Leading Scientists

Good morning, everyone!

We are excited to announce our collaborative research with Dr. Jonathan Epp and his team, in which we employed the Nanostring GeoMx Digital Spatial Profiler (DSP) and 10X Genomics Xenium In Situ platforms to identify and validate molecular biomarkers associated with Alzheimer's disease.

The study is titled "Impaired Parvalbumin Interneurons in the Retrosplenial Cortex as a Cause of Sex-Dependent Vulnerability in Alzheimer's Disease." You can access the paper at DOI: 10.1126/sciadv.adt8976 and below is a detailed overview of how the GeoMx and Xenium platforms were employed in this study:

  Nanostring GeoMx DSP Xenium In Situ
 
Methodology
  • GeoMx was utilized for spatially resolved transcriptomics using the whole-transcriptome panel
  • Tissue sections were prepared and stained with fluorescence-labelled antibodies that targeted specific cell types, such as neurons and parvalbumin interneurons
  • This approach allowed GeoMx to capture gene expression profiles of each cell type in the retrosplenial cortex (RSC) selected based on antibody staining
  • Xenium represents a novel approach utilized for single-cell resolution spatial transcriptomics
  • Unlike traditional methods such as GeoMx, Xenium facilitates direct in situ hybridization of transcripts, thus enabling the precise subcellular localization of RNA molecules
  • This technique was employed to generate high-resolution maps that elucidate gene expression at the single-cell level
 
Application
  • GeoMx was employed to map transcriptomic differences between male and female mice, as well as between diseased models and control groups
  • GeoMx enabled the distinction between parvalbumin interneurons and neurons within the RSC
  • The findings were subsequently validated through the utilization of an alternative spatial platform, i.e., Xenium
  • Xenium served as a pivotal tool for validating and refining findings initially identified through GeoMx, offering an elevated spatial resolution
  • This method proved instrumental in uncovering cellular heterogeneity, allowing for the resolution of gene expression patterns both at the single-cell and subcellular tiers across diverse brain regions
 
Key Findings
  • The analysis led to the identification of down-regulated genes in the disease model group compared to the control group, highlighting potential early treatment targets
  • Gaining insights into the vulnerability of this region to Alzheimer's disease-related pathology is crucial for developing effective early diagnostic and intervention strategies
  • The application of Xenium data resulted in the identification of cell-type-specific expression signatures with remarkable precision
  • The enhanced approach contributed to a deeper understanding of the spatial relationships that exist between different cell populations
  • The findings corroborated previous results obtained from GeoMx, but with a higher degree of granularity, thereby enriching the overall comprehension of the studied biological systems

In summary, GeoMx provided a comprehensive regional overview, identifying larger-scale spatial transcriptomic patterns. At the same time, Xenium refined these insights by offering single-cell and subcellular resolution, allowing for precise gene expression mapping at the individual cellular level. Integrating these technologies effectively bridges the divide between tissue-wide transcriptomics and single-cell resolution data, culminating in a more holistic analysis of spatial biology. Check out the analysis workflows and results at ASOC.ucalgary.ca/publication/RSC_PV_AD/.

As shown above, spatial transcriptomics (ST) has significantly advanced our ability to study gene expression within tissues while preserving spatial context. Despite its transformative impact, ST technologies still face challenges in sensitivity, specificity, and technical variability. For example, we continue to encounter technical biases, notably a consistent pattern of genes exhibiting either high or low coverage irrespective of tissue type. To address this, we investigated and characterized the capabilities of pre-designed panels using publicly available ST datasets: seven Xenium In Situ with the Xenium Prime 5k panel and 16 GeoMx DSP datasets using the human whole transcriptome atlas (WTA) panel. Check out our paper on bioRxiv at DOI: 10.1101/2025.04.07.647642 for more information and below are the key findings and Figure 1 from the study:

  1. The study examines publicly available ST datasets from the 10X Genomics and Nanostring platforms
  2. The proposed coverage index (CI) quantitatively assesses gene representation across different panels, i.e., Xenium Prime 5k and GeoMx DSP WTA
  3. Cancer-related gene lists exhibit the highest CI values, suggesting that pre-designed panels are optimized for oncology research
  4. Genes encoding ligand/receptor and specific cellular markers show lower representation, indicating potential gaps in panel design
  5. The findings highlight the need for customized panel designs to ensure adequate gene coverage in diverse biological and clinical applications
 
Figure 1. Evaluation of the pre-designed gene panel utilizing the coverage index (CI). (a) A schematic representation outlining the calculation of the coverage index. (b) CI values for 23 cohorts examining ligand and receptor genes. (c) Mean CI values across cohorts for ten different gene databases.		  

In conclusion, the fundamental issue lies in the inherent limitations of commercial pre-designed gene panels, which may not adequately encompass all relevant biological signatures. Consequently, clients should consider developing an add-on panel tailored to include specific genes of interest.

Thank you very much for your ongoing trust and collaboration. We look forward to using these exciting new capabilities to support your research!

Spatially yours,

The ASOC team