Rigorous Validation, Platform by Platform
Every panel IMCO delivers — whether run on the Lunaphore COMET cycIF platform or the NanoString GeoMx Digital Spatial Profiler — is built on a validated reagent foundation and confirmed by independent wet lab and dry lab checks before data is released.
Antibody-Level Validation
Clone selection, titration, and epitope-stability testing establish that every antibody in a panel performs before it's ever run against tissue at scale.
See the processWet Lab Validation
mIHC confirmation staining, registered to the same tissue region, gives a technician/scientist-reviewed qualitative check of marker localization and specificity.
See the processDry Lab / Analytical Validation
Per-cell marker intensities from cycIF and mIHC are correlated statistically, confirming the image-analysis pipeline reproduces expected marker relationships.
See the processAntibody-Level Validation
Multiplex spatial assays are highly sensitive to reagent quality — a single underperforming antibody can compromise an entire panel. Before any clone is placed into a cycIF panel, it's screened, titrated, and tracked so that panel design starts from data, not guesswork.
Literature-Backed Clone Choice
Clones are prioritized by published spatial/IF performance (BenchSci, Human Protein Atlas) and prior IHC/IF track record in the FFPE or fresh-frozen tissue of interest — never on vendor datasheet claims alone.
Full-Assay Compatibility Checks
Species reactivity, tissue type, and fixation compatibility are confirmed for the cyclic multiplex context — a clone that performs in singleplex won't always hold up across a full cycIF panel.
Broad-Range Titration
Each clone is tested across a broad serial dilution range in both fluorescence channels, screened for specificity and signal-to-noise, and checked against positive/negative tissue controls spanning normal and disease tissue.
Cycle Stability Testing
Signal is compared across early and late cycles (e.g., cycle 1 vs. cycle 20) to catch cycle-dependent signal loss early and guide where each marker is placed in the panel.
Organization & tracking behind every clone
- Antibody inventory tracked by catalog number, lot, concentration, and conjugate
- Validation status recorded per clone, tissue type, and fixation condition
- Titration results, known challenges, and recommended conditions logged as performance notes
- Centralized tracking prevents repeated mistakes and redundant re-validation across projects
mIHC Confirmation Staining for Qualitative Confirmation
Once a panel is assembled, each marker is checked against a traditional multiplex IHC (mIHC) stain — a qualitative, technician/scientist-reviewed comparison confirming the marker appears in the right cell type and subcellular compartment, on the very same tissue region evaluated by cycIF.
Sequential CycIF → mIHC Staining
CycIF staining is completed first, and mIHC staining for two markers from the panel follows on the same slide — an order chosen to minimize tissue degradation from mIHC processing.
Pathologist-Guided ROI Selection
Images are reviewed in Visiopharm, where 2–3 regions of interest are selected under pathologist guidance — each corresponding to a single core on the TMA and focused on the highest concentration of positive cells.
Image Registration
Visiopharm's TissueAlign module aligns the mIHC image to the Lunaphore cycIF image for each ROI, so both stains are compared on exactly the same cells.
Qualitative Technical Review
A trained technician or scientist visually compares expression pattern and subcellular localization between cycIF and mIHC on the registered ROIs — a qualitative check on every marker.
What the wet lab review confirms
- Marker localization matches expectation — nuclear, membranous, or cytoplasmic as appropriate
- Sanity-check co-positivity and negativity across the expected marker set on the same cells
- Registration confirms cycIF and mIHC are being compared on the identical tissue region, not adjacent sections
- Technician/scientist sign-off provides a qualitative check alongside the quantitative correlation below
Orthogonal Confirmation Using Pearson Correlation
Alongside the qualitative pathology review, the same registered ROIs are used to quantitatively test concordance between cycIF and mIHC — scoring each cell's marker intensity on both platforms and summarizing the relationship with a Pearson correlation coefficient (R).
Nuclear Segmentation & Intensity
Following registration, nuclei are segmented from the DAPI channel, and average marker intensity per cell is computed and recorded for both cycIF and mIHC.
Intensity Normalization
Average marker intensities are scaled from 0 to 1 across all ROIs on a slide, accounting for dynamic-range differences between the cycIF and mIHC assays.
Pearson Correlation Plot
Scatter plots of average marker intensity are generated, and the Pearson correlation coefficient (R) is calculated per ROI, then averaged to a single value per slide.
Threshold & Escalation
An R value above 0.7 signifies a successful correlation. Below 0.7, the scatter plot is reviewed for cell-count sufficiency, which may prompt staining of additional tissue.
Representative per-marker correlation plots from a validation run — actual R values are generated per marker, per slide.
Panel-wide view — Pearson R between cycIF and mIHC across markers and tissues, tracked individually rather than as one pass/fail number.
What the dry lab review confirms
- Per-ROI Pearson R averages above 0.7 per slide, meeting the pass threshold
- Intensity normalization accounts for dynamic-range differences between cycIF and mIHC
- Low correlation triggers a scatter-plot review for cell-count sufficiency before any panel-level conclusion
- Results are documented per marker, per slide, alongside the technician/scientist's qualitative sign-off
Antibody & Panel Validation
Every clone is screened, titrated, and cross-platform confirmed before it's placed into a finalized protein or RNA panel.
See the processWet Lab Validation
Morphology-guided ROI/AOI segmentation is reviewed against tissue architecture before a single oligo is collected.
See the processDry Lab / Analytical Validation
Sequencing QC, filtering, and Q3 normalization precede a cross-platform correlation check against RPPA and RNAseq.
See the processAntibody & Panel Validation
Before a clone or probe set is placed into a GeoMx panel, it passes through a staged validation pipeline — from literature-backed clone selection through a full reference-cohort run — so panel composition is backed by data at every step.
Literature & Vendor Screening
Candidate clones are identified from the literature and vendor data, then validated by conventional IHC/IF on cell lines, tissue, drug-treated, and siRNA-knockdown controls using the unconjugated antibody.
DNA-Barcode Labeling & Re-Confirmation
A carrier-free antibody order is DNA-barcode labeled in-house, and IHC/IF is repeated with the barcode-conjugated antibody to confirm labeling hasn't altered staining performance.
Standard TMA Control
The labeled antibody is run in a full DSP panel mix against a standard tissue microarray control spanning multiple cell lines and treatment conditions, evaluating dynamic range and signal-to-noise.
Reference Cohort Run
The finalized clone is run across a reference patient cohort (e.g., breast, ovarian, sarcoma) and cross-platform compared against RNAseq, RPPA, and IHC/IF before the panel is locked.
What antibody-level validation confirms
- Unconjugated and barcode-conjugated antibody give concordant IHC/IF staining
- Standard TMA control shows expected dynamic range and signal-to-noise across cell lines and treatments
- Reference-cohort correlation with RNAseq/RPPA/IHC meets the panel's acceptance criteria before finalization
- Clone, lot, and validation status are tracked centrally to prevent redundant re-validation
Morphology-Guided ROI/AOI Segmentation
Regions and areas of interest are drawn from fluorescent morphology markers rather than guessed at — each ROI is segmented into biologically meaningful compartments (e.g., tumor vs. peritumoral stroma) before any oligo is collected for counting.
Morphology Marker Panel
Fluorescent morphology markers (e.g., DNA, PanCK, CD45) are applied to visualize tumor, immune, and stromal compartments prior to ROI selection.
Compartment-Level AOI Masking
Each ROI is segmented into areas of interest by compartment — for example tumor and peritumoral (~20 µm surround) regions — using an erosion-based masking function.
Qualitative Technical Review
A trained technician or scientist confirms that segmentation boundaries track the expected tissue architecture before AOI collection proceeds.
Representative ROI with AOI segmentation mask — individual cells within the region are outlined ahead of oligo collection.
What the wet lab review confirms
- AOI boundaries track the intended tissue compartment (e.g., tumor vs. peritumoral stroma)
- Morphology marker staining is specific, with minimal cross-compartment bleed
- Segmentation masks are reviewed against the underlying fluorescence image before collection
- Technician/scientist sign-off is logged before AOI collection proceeds
QC, Normalization & Cross-Platform Correlation
Every run is held to a defined set of QC thresholds before analysis, normalized to remove technical variation, and — for new or custom targets — checked against an orthogonal platform to confirm the analytical pipeline holds up.
Run-Level QC Thresholds
Each AOI must clear defined default thresholds: ≥1,000 raw reads, ≥80% aligned reads, and ≥50% sequencing saturation (below 50% flags a sample for re-sequencing).
Segment & Control QC
No Template Controls must stay below 1,000 reads, and each AOI needs ≥100 nuclei (≥400 recommended for clinical samples) over ≥16,000 µm² of surface area.
Q3 Normalization
Low-expression targets are filtered out (requiring ≥1,000 genes above the limit of quantification for pathway analysis), and remaining counts are Upper-Quartile (Q3) and negative-control normalized.
Orthogonal Correlation
Normalized signal for key targets is correlated against RPPA and/or RNAseq on the same cell lines, with correlation strength documented per target in the validation record.
Sequencing saturation vs. read depth across AOIs — the dashed line marks the 50% saturation QC threshold.
What the dry lab review confirms
- Every AOI clears read count, alignment, and saturation thresholds before analysis
- Nuclei count and surface area meet minimums for reliable low-expression target detection
- Q3 and negative normalization remove technical variation before biological comparison
- New or custom targets are cross-platform confirmed before being reported with confidence
Cross-Platform Correlation in Practice
Representative correlation of DSP protein signal against orthogonal RPPA data across a cell-line panel — each point is one cell line, with Pearson's r reported per target. Strength of correlation varies by target, which is why concordance is assessed and documented individually rather than as a single panel-wide pass/fail.
Illustrative cross-platform correlation set — actual acceptance thresholds and cell-line panels are documented per study.