Multiplex Panels
Per-analyte calibration and bounded unknown quantification for multiplex panels
A multiplex panel measures multiple analytes in one experiment. Gradiance fits each analyte independently with the selected linear or quadratic model and back-calculates unknown signals only inside that analyte's calibrated range.
What it measures
The concentration of each analyte in each sample, derived from a per-analyte standard curve. The same plate can carry cytokines, chemokines, and growth factors that span orders of magnitude in concentration - every one gets its own fit.
Data you upload
- A tidy table with
analyte,sample,kind, andsignalcolumns. Setkindtostandardorunknown. - A
concentrationvalue for every standard. Unknown rows may include a positivedilution_factor; otherwise the factor is one. State the concentration unit explicitly.
What you get back
- The selected linear or quadratic calibration for every analyte, with fit coefficients and R².
- Back-calculated concentrations for each sample, with values below the lowest standard or above the highest marked as out-of-range rather than silently extrapolated.
- A row for every unknown, including its analyte, fitted concentration, dilution factor, model, and fit R².
Method note
Choose linear only when signal is adequately linear over the reported range. Choose quadratic for bounded curvature; the quadratic response must remain monotonic over the calibration interval so its inverse is unique. The method does not fit 4PL or 5PL curves. Signals outside the observed calibration response are marked out of range and are never extrapolated.
Related
For a single-analyte standard curve, see Standard Curves & Quantification. For spot-counting immune readouts, see ELISpot. For flow cytometry population summaries, see Flow Summary.