The problem. Fluorescence microscopy is limited by spectral overlap: you can stain only a handful of proteins at once before their colours collide. Tissue biology is driven by many proteins interacting in space, so a readout capped at four or five markers can’t capture a microenvironment. The bottleneck is the reporter, not the microscope.

The idea. Imaging mass cytometry swaps fluorophores for metal-isotope tags — the same chemistry that let mass cytometry (CyTOF) measure ~40 parameters in suspension. A laser ablates the tissue spot by spot; the vaporised material goes to a mass spectrometer that counts each isotope, so every pixel carries a high-dimensional protein profile. The result is subcellular-resolution images of 30-plus proteins simultaneously, with no spectral overlap.

Why it matters. This is the imaging counterpart to the RNA-imaging methods from the spatial day (MERFISH, seqFISH+) — proteins, not transcripts, mapped in situ — and it’s foundational to the STU’s remit, where spatial proteomics sits beside spatial transcriptomics. It also connects to the proteomics thread running through this reading (MaxQuant, mass spec): the same detector principle, now imaging tissue.

Verdict. A landmark that founded highly-multiplexed tissue proteomics; throughput is slow and destructive, and marker panels need antibody validation. Read it as the paper that made “many proteins, in place” possible — the technique CODEX and MIBI would race alongside.