GeoMx Digital Spatial Profiler: Region-Based Spatial Profiling

The GeoMx Digital Spatial Profiler measures RNA and protein from tissue regions that a researcher selects, rather than mapping every cell across a slide. That region of interest (ROI) based design fits within the wider spatial biology methods landscape alongside sequencing- and imaging-based approaches. It suits hypothesis-driven work where the biology already has a known location, such as a tumor margin or an immune infiltrate.

How the GeoMx Digital Spatial Profiler works

Digital spatial profiling separates where a measurement happens from how it is detected. A tissue section is stained with morphology markers, then labeled with RNA probes or antibodies that carry an oligonucleotide tag attached through a photocleavable linker.

The foundational description of this design showed that ultraviolet light projected onto the tissue can release those tags from an operator-defined area covering as few as a single cell or as many as roughly 5,000 cells at once, without physically dissecting the section. The released tags are then collected from that specific area for downstream counting.

Figure 1: A labeled schematic of the four-stage digital spatial profiling workflow, from tissue staining through region selection, barcode release, and expression counting. Credit: AI-generated image created using Google Gemini (2026).

Because the release step happens optically rather than through physical microdissection, the surrounding tissue stays intact and available for additional stains or serial-section work. That separation of spatial selection from molecular counting is the core idea behind every GeoMx workflow, whether the readout is RNA, protein, or both.

The instrument works on both fresh-frozen tissue and formalin-fixed, paraffin-embedded (FFPE) sections, the two preservation methods most common in research and clinical pathology archives. FFPE compatibility matters because many banked clinical samples relevant to translational research exist only in that fixed form.

Once collection is complete, the released oligonucleotide tags are quantified either through next-generation sequencing or a lower-plex optical counting method, depending on the size of the gene or protein panel in use. Sequencing-based counting is the standard route for larger RNA panels, including the WTA assay described later in this guide.

ROI selection and segmentation in GeoMx

Selecting ROIs is the step that determines what a digital spatial profiling experiment actually measures. A pathologist or researcher views the stained slide and draws each ROI around a tissue feature of interest, such as a tumor nest or an immune-rich zone, rather than the instrument sampling a predefined, unbiased grid.

Each ROI can be further divided into AOIs based on morphology staining, producing separate expression profiles for compartments within the same region, such as tumor cells vs surrounding stroma. Clinically sourced workflows typically target ROIs covering a minimum of 100 cells to keep the resulting profile statistically stable.

Because each ROI aggregates signal across every cell it contains, the output represents an average across that population rather than a single-cell measurement. That trade-off is deliberate. It allows one experiment to sample many anatomically distinct regions across multiple sections without the cell segmentation burden that single-cell imaging platforms require.

Morphology markers used to guide ROI and AOI selection typically pair a general nuclear stain with one or two antibodies that highlight a specific cell population, such as markers for epithelial cells or infiltrating immune cells. Those markers let a researcher draw a single tumor-focused ROI, then split it automatically into tumor and stromal compartments based on staining intensity rather than redrawing each boundary by hand.

This morphology-guided segmentation is particularly useful in oncology research, where comparing gene or protein expression between a tumor compartment and its adjacent stroma is often the central question. Drawing that boundary manually for every sample in a large cohort would otherwise be impractical.

RNA and protein readouts on the GeoMx platform

The GeoMx Digital Spatial Profiler was built from the outset as a dual-analyte platform, measuring RNA and protein from the same instrument. RNA readouts rely on in situ hybridization probes linked to indexing oligonucleotides, while protein readouts substitute antibodies carrying the same photocleavable tag chemistry.

A feasibility study in Barrett's esophagus applied both RNA and protein assays on the platform, profiling immune-related transcripts and proteins separately across epithelial and stromal compartments of the same fixed tissue samples. This approach lets researchers validate a protein finding against the transcript that encodes it while preserving the original ROI coordinates.

That capability is particularly useful in translational studies, where a candidate biomarker identified at the protein level often needs transcript-level confirmation from the exact same tissue region before it moves toward clinical validation.

Protein panels available for the platform range from small, curated marker sets focused on a single pathway to broader immuno-oncology panels covering dozens to several hundred targets at once. Because protein and RNA assays reference the same ROI coordinates, a researcher can run an RNA panel on one section and a protein panel on an adjacent serial section, then align both datasets to the same anatomical structure.

That alignment is especially relevant for biomarker discovery programs, where an early transcript-level signal often needs orthogonal confirmation at the protein level before it is considered a credible candidate for further development.

GeoMx WTA for genome-scale profiling

The WTA assay extended GeoMx from small, disease-focused gene panels to essentially the full protein-coding transcriptome. Peer-reviewed analysis pipelines built around the assay report coverage exceeding 18,000 human genes within each profiled region.

That scale converts GeoMx from a targeted validation tool into a genuine discovery platform, letting researchers ask which genes distinguish two tissue regions without committing to a panel in advance. A 2021 study of synovial sarcoma used the WTA assay to profile tumor and immune compartments separately within the same fixed tissue sections, identifying a malignant cell program linked to low immune infiltration and poorer clinical outcomes.

Because WTA still aggregates signal at the ROI level rather than resolving individual cells, its whole-transcriptome breadth comes at the cost of the subcellular resolution that imaging-based single-molecule platforms offer over a smaller, preselected gene list.

A mouse version of the assay extended the same genome-scale approach beyond oncology and immunology into fields such as neuroscience and developmental biology, where model organism tissue is the primary substrate for spatial profiling work. Because the mouse and human assays share the same underlying chemistry and instrument workflow, findings from a discovery-stage mouse experiment can often carry forward into a human FFPE follow-up study without redesigning the overall approach.

Many labs treat WTA as a first-pass discovery step, nominating a shortlist of candidate genes from an unbiased, whole-region survey before narrowing to a smaller, targeted panel for higher-throughput validation across additional samples.

GeoMx vs single-molecule spatial platforms

GeoMx occupies a specific niche relative to single-molecule imaging platforms such as Xenium, the CosMx Spatial Molecular Imager, and MERSCOPE, which localize individual transcripts to a segmented cell rather than aggregating signal across a hand-drawn region. Weighing that trade-off is part of the broader work of selecting a transcriptomics platform for a given tissue and research question. Commercial single-molecule gene panels typically profile several hundred to several thousand preselected genes at subcellular resolution, a narrower scope than WTA's genome-scale coverage.

An independent benchmarking study that compared single-cell imaging platforms against GeoMx data from tumor samples found that concordance with bulk RNA sequencing varied meaningfully by platform and probe design, a reminder that no single spatial technology is universally superior. Choosing between them depends on the question at hand rather than on any one platform's marketing specifications.

In practice, many spatial biology programs combine both approaches rather than choosing one exclusively. A GeoMx experiment can nominate candidate regions and genes from a whole-transcriptome survey, after which a single-molecule imaging platform resolves those same candidates down to individual cells within the highest-priority samples.

That progression, from broad regional discovery toward targeted single-cell validation, mirrors how many labs already structure experiments in other omics fields. It lets a research program reserve the more expensive single-molecule imaging step for the samples and gene sets that have already shown a signal worth resolving further.

Table 1: A comparison of GeoMx digital spatial profiling and single-molecule imaging platforms.

Feature	GeoMx Digital Spatial Profiler	Single-molecule imaging platforms
Spatial unit	Researcher-selected ROI	Individual segmented cell
Gene coverage	Up to whole transcriptome	Hundreds to several thousand targeted genes
Analyte options	RNA and protein from one instrument	Primarily RNA, with limited protein add-ons
Best suited for	Hypothesis-driven, pathology-anchored questions	Subcellular resolution and neighborhood mapping

For hypothesis-driven, pathology-anchored questions, where the ROI is already defined by histology and whole-transcriptome breadth matters more than single-cell resolution, GeoMx remains a practical first choice. It is especially useful when a project also needs matched protein data from the same tissue section.

A short framework helps clarify whether GeoMx fits a given project before committing tissue and budget to a platform:

1. Confirm that the biological question centers on defined tissue regions, such as a tumor margin or immune infiltrate, rather than requiring whole-slide, single-cell mapping.
2. Determine whether the project needs matched RNA and protein data from the same section, a combination GeoMx supports natively.
3. Decide whether whole-transcriptome discovery or a smaller, targeted panel better fits the research stage.
4. Assess whether ROI-level aggregation is an acceptable trade-off against the subcellular resolution of single-molecule platforms.
5. Confirm that FFPE or fresh-frozen tissue of adequate quality is available, since sample condition affects probe performance.

GeoMx Digital Spatial Profiler earns its place in ROI-driven research

The GeoMx Digital Spatial Profiler fills a specific gap in the spatial biology toolkit. By letting researchers define exactly where a measurement happens, it turns pathology expertise into an upfront experimental design choice rather than a downstream computational one.

That region-based approach will not replace single-molecule imaging for questions that genuinely require subcellular resolution or a full cellular neighborhood map. For hypothesis-driven, ROI-based work built around known tissue structures, though, GeoMx remains a proven and practical way to generate matched RNA and protein data from exactly the regions that matter.

As WTA has matured alongside a growing library of protein panels, the platform has moved from a niche pathology tool into a standard option many core facilities offer alongside single-molecule imaging instruments. Researchers planning a spatial biology project are increasingly likely to weigh a region-based approach against a single-cell one as a routine part of experimental design, rather than defaulting to whichever instrument happens to be locally available.

This content includes text that has been created with the assistance of generative AI and has undergone editorial review before publishing. Technology Networks' AI policy can be found here.