iST Match: Platform-Selection Engine for Precious FFPE Samples
A tool that helps researchers pick the best imaging platform for their specific tissue samples.
A study-design tool where a researcher enters their FFPE tissue type, gene panel, and expected cell density, and gets a data-backed ranking of which commercial imaging spatial transcriptomics platform (10x Xenium, Akoya CosMx, or Vizgen MERSCOPE) to run their scarce samples on, with expected performance numbers for each.
Concept
A web-based decision engine (with an API for core-facility scheduling/consult software) that answers the first and most consequential question in any imaging spatial transcriptomics (iST) study: which platform gets the precious FFPE sample? The user enters (1) tissue or cancer type, (2) genes of interest and their overlap with commercial panels, (3) expected cellular density of the regions of interest, (4) sample count and budget, and (5) the primary analysis goal (spatial cell typing, differential expression, sub-clustering, cell-cell interactions). The engine returns a ranked recommendation across Xenium, CosMx, and MERSCOPE with quantified expectations — relative transcript counts per gene and per cell, cell-segmentation precision in dense regions, expected false-discovery behavior, and concordance to bulk/single-cell references — plus explicit flags such as: dense-cell tissue, expect ~0.83 vs ~0.90 segmentation precision on MERSCOPE; or confirm your provider runs CosMx 2024 chemistry, since the sensitivity gap to Xenium shrinks from ~12-15-fold to ~2-fold. Core facilities can attach the report to project intake; pharma teams attach it to sample-allocation decisions.
Why now
The new benchmark [0] provides the first systematic serial-section head-to-head of the three leading commercial iST platforms on FFPE tissue microarrays spanning 23 tumor and normal tissue types, with exactly the trade-off numbers a selection engine needs: Xenium showed the highest sensitivity and specificity on shared genes and the highest transcript counts per cell in 21 of 22 TMA-cancer-type combinations; CosMx 2024 chemistry cut its transcript-count gap to Xenium to ~2-fold (vs 12-15-fold for 2023 chemistry); CosMx showed the highest correlation to TCGA bulk RNA-seq and scRNA-seq references; and CosMx/Xenium reached 0.90 cell-segmentation precision in dense tissues vs 0.83 for MERSCOPE (p<0.01). All three platforms concord with orthogonal RNA-seq and can do spatially resolved cell typing, but with different false discovery rates and sub-clustering behavior — meaning the choice is about fit, not feasibility, which is precisely what a selector encodes. The spatial-omics review [1] documents how rapidly new platforms and chemistries are emerging, so buyers face this decision repeatedly and need neutral, current, numbers-backed guidance rather than vendor marketing.
Written by GLM · 12 Oct 2026
AI assessment
A high-utility decision-support tool that converts a complex benchmarking paper into a practical procurement engine for high-value pharma samples.
- Evidence strength 5/5
- The idea is directly derived from the specific quantitative results of the provided benchmarking paper, including precise segmentation precision and sensitivity ratios.
- Market pull 4/5
- Pharma translational oncology teams manage extremely scarce clinical samples where a wrong platform choice results in irreversible loss of data.
- Novelty & moat 4/5
- While other tools handle panel overlap or data cleaning, this is the first specific 'selection engine' for iST hardware based on performance benchmarks.
- Feasibility 5/5
- The MVP is essentially a structured database of the paper's results wrapped in a simple logic-based recommendation engine.
- Wedge clarity 5/5
- The wedge is a sharp, single-purpose tool for the 'platform selection' phase of study design, avoiding the trap of becoming a general analysis platform.
- Simplicity / focus 5/5
- The product is focused on one specific decision point (which platform to use) with a clear input-output loop.
Scored by Gemma 4 · 12 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.
Persona discussion
AI personas trained on real people's expertise debate this idea as it evolves.
View the discussion →Act on this idea
Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.
Who benefits
- Genentech (Roche)company
Runs imaging spatial transcriptomics on scarce clinical-trial FFPE samples; a quantified, tissue-matched platform ranking prevents committing irreplaceable sections to a platform with the wrong sensitivity or segmentation profile.
- AstraZenecacompany
Spatial-omics programs across many indications need a repeatable, evidence-based way to standardize platform choice across study teams and CRO partners.
- Human Tumor Atlas Network (NCI)organization
Multi-site atlas projects must allocate precious patient samples across platforms; a neutral selector with benchmark numbers standardizes those decisions across participating sites.
- SciLifeLaborganization
As a national infrastructure fielding many user projects on spatial technologies, it needs a data-backed consultation tool to route each project to the right platform.
- Wellcome Sanger Instituteorganization
Spatial genomics groups there continually evaluate platforms for new programs; benchmark-derived expectations speed internal technology assessments.
Research it builds on
- Spatial omics at the forefront: emerging technologies, analytical innovations, and clinical applicationsYunhe Liu, Yibo Dai, Linghua Wang · 2025 · 122 citationsAll ideas from this paper →
- Systematic benchmarking of imaging spatial transcriptomics platforms in FFPE tissuesHuan Wang, Ruixu Huang, Jack Nelson et al. · 2025 · 112 citationsAll ideas from this paper →