Spatial transcriptomics is a family of methods that measures gene expression while keeping each measurement anchored to its physical location within a tissue section. Where conventional bulk or single-cell RNA sequencing dissolves tissue into a suspension and loses all positional information, spatial approaches record where in the tissue each transcript came from. The field divides into two main technology categories with a fundamental trade-off: sequencing-based platforms capture the whole transcriptome at coarser spatial resolution, while imaging-based platforms deliver single-molecule precision but only for predefined gene panels. Choosing between them is largely a matter of matching that trade-off to the research question.

Two Main Approaches: NGS-based vs. Imaging-based Technologies

France Génomique's technical overview classifies spatial omics methods along four axes: spatial resolution (the minimum size of molecular units profiled), coverage (breadth of tissue covered), scale and throughput, and multiplexing capacity (how many molecular entities are profiled simultaneously). On that map, the field splits into two broad families: targeted, multiplexed probe- or antibody-based approaches, and transcriptome-wide, next-generation sequencing (NGS)-based approaches.

NGS-based platforms work by capturing RNA from an intact tissue section onto a surface carrying positional barcodes, then sequencing the captured molecules. Because every read carries a barcode tying it back to a spot on the slide, the expression profile can be reconstructed as a map. Illumina describes its own sequencing-based spatial technology as enabling "hypothesis-free gene expression profiling while preserving spatial context," and notes that this design addresses the limitations of imaging-based approaches, "which are often constrained by predefined gene panels and limited tissue coverage."

Imaging-based platforms take the opposite route. Rather than sequencing everything and reconstructing position afterward, they detect RNA molecules where they sit, using fluorescently labeled probes and sequential rounds of microscopy. A Nature Communications benchmarking study led by Pengfei Ren and colleagues explains that imaging-based platforms such as NanoString's CosMx 6K and 10x Genomics' Xenium 5K "rely on fluorescently labeled probes and sequential imaging," offering single-molecule precision. The price of that precision is that only genes with designed probes can be seen.