Before a multi-colour flow panel can be read, the spillover between detectors has to be measured and removed. This tool computes the spillover percentages for a two-fluorophore pair from single-stain control medians, the first step in building a compensation matrix.
Why compensation matters
Fluorescent dyes are imprecise light sources. A dye designed to be detected in the FITC channel also emits photons that land in the PE channel. If you do not correct for this spillover, a FITC-bright population appears falsely positive in the PE detector — a catastrophic artefact when you are trying to identify rare subsets or establish meaningful gates. Compensation is the mathematical correction that subtracts this “leakage” and restores independent channel readings.
For a two-colour panel (for example, FITC and PE), you need to know two off-diagonal spillover values: how much FITC spills into the PE detector, and how much PE spills into the FITC detector. This tool computes both from your single-stain control MFI readings.
How spillover is measured
Run one single-stain control per dye and record the median fluorescence (MFI) of the positive and negative populations in each detector. The spillover of fluorophore i into detector j is its net leak signal divided by its own net signal:
spillover%(i → j) = 100 × (MFI_pos_j − MFI_neg_j) / (MFI_pos_i − MFI_neg_i)Subtracting the negative population MFI from each detector reading removes autofluorescence and instrument background in every detector. The diagonal — a dye into its own detector — is 100% by definition. The two off-diagonal values form the spillover matrix the cytometer software inverts to compensate your experimental data.
Worked example
Suppose you run a FITC single-stain control and measure:
| Detector | Positive MFI | Negative MFI | Net |
|---|---|---|---|
| FITC (own) | 20,200 | 400 | 19,800 |
| PE (other) | 2,650 | 400 | 2,250 |
FITC spillover into PE = 100 × 2,250 / 19,800 ≈ 11.4%
You then run a PE single-stain control and measure PE spillover into FITC in the same way. Enter all four positive/negative MFI values into this tool and both spillover percentages are calculated instantly.
Spillover is not the same as spreading error
The single most misunderstood point in compensation: a correctly compensated plot can still be uninterpretable. Compensation removes the median spillover exactly, but it cannot remove the spread that spillover leaves behind. When a bright FITC population leaks into the PE detector, the leaked signal carries photon-counting (Poisson) noise, so after compensation the PE-negative population is not a tight blob — it is smeared upward into a wedge. This spreading error is why a dim PE marker can become ungateable next to a bright FITC one even with perfect compensation.
Practical consequences:
- Over-compensation is a myth people chase. If a population’s median sits correctly on the axis but the distribution fans out, the answer is not more compensation (which would push the median negative) — it is a better dye choice or a brighter fluorophore for the weak marker.
- Put dim markers on bright, well-separated dyes. Assign your lowest-density antigen to a fluorophore with little spillover from your bright channels, not to the one that happens to be free.
- Tandem dyes degrade. PE-Cy7 and APC-Cy7 tandems break down with light and fixation, shifting spillover between lots and even between the start and end of a long acquisition — re-run compensation controls from the same lot.
Critical rules for reliable compensation
Match brightness. Controls must be as bright as or brighter than the actual samples. Compensation calculated from a dim control will over-compensate a bright experimental population.
Use the same voltages. Run single-stain controls and experimental tubes at identical detector voltages. Voltage changes shift the MFI values and invalidate previously calculated spillover.
Prefer beads over cells when cells are scarce. Compensation beads (sold for FITC, PE, APC, etc.) bind antibodies at a very high, consistent density and produce clear positive populations even when cells are rare or have high autofluorescence.
Background-subtract always. Autofluorescence and non-specific binding raise the baseline signal in every channel. The formula’s negative-MFI subtraction handles this — never skip entering the unstained negative values.
High spillover is a panel design problem. Values in the 10–30% range are common between adjacent dyes like FITC and PE and are manageable. Spillover above 50% means spectral overlap so severe that spreading error is large; the best fix is a different dye pairing, not heavier compensation.
Sources and references
- Roederer M. Spectral compensation for flow cytometry: visualization artifacts, limitations, and caveats. Cytometry (2001) — the foundational treatment of compensation and spreading error
- Maecker HT, Trotter J. Flow cytometry controls, instrument setup, and the determination of positivity. Cytometry A (2006) — single-stain controls and background subtraction
- BD Biosciences — Compensation in flow cytometry (technical guide) — practical bead-vs-cell and voltage-matching guidance
Maintained by the Gera Tools editorial team. Spillover = net leak MFI ÷ net source MFI from single-stain controls; controls must match sample brightness and use identical voltages. For research/lab use, not diagnostic interpretation. Last reviewed 2026-07-02.