Flow Cytometry and FACS: Compensation, Controls, and the Sorting Trade-off

Sample preparation · Compensation and spectral unmixing · FMO controls · Surface, intracellular, phospho-flow · Sorting purity vs yield

Published 2026 · Last updated 2026-09-27 · Hailu Zhou

Flow cytometry measures optical properties of single cells at tens of thousands per second, and fluorescence-activated cell sorting extends that to physically isolating the cells it measures. The technique is powerful precisely because it is quantitative per cell, but every number it produces depends on instrument settings, compensation, and controls that are invisible in the final figure. A gate that looks obvious in a plot can be a filtering artefact, and a sorted population that looks pure can be a doublet. Here is what to check before trusting either.

Sample preparation: single cells or nothing

The cytometer can only analyse single particles, so the preparation must produce a true single-cell suspension. Adherent cells are detached with trypsin or, more gently, with EDTA or a non-enzymatic dissociation buffer — enzymatic cleavage of surface antigens is a common cause of vanishing signal. Solid tissues require mechanical dissociation plus enzymatic digestion (collagenase, dispase) tuned to preserve the epitope of interest; over-digestion destroys surface markers, under-digestion leaves clumps.

Compensation: the arithmetic of spectral spillover

Every fluorophore emits across a broad spectrum, so signal from one dye appears in detectors intended for another. Compensation subtracts the measured spillover. It is computed from single-colour controls run on the same instrument with the same detector settings as the experiment. If the instrument settings change, the compensation matrix is invalid.

Two honest facts. First, compensation is a linear matrix subtraction, so it works well when signal is linear but becomes unreliable at very bright or very dim extremes; a partially compensated double-positive population can look like two single-positive populations. Second, compensation controls must be as bright as or brighter than the sample, and they should be single-stained cells or beads matching the sample's autofluorescence, not antibody alone.

Spectral flow cytometry changes the problem rather than solving it. Instead of a few bandpass detectors, it disperses the full emission spectrum across many detectors and unmixes spectrally. This resolves dyes that are inseparable on a conventional cytometer and can distinguish autofluorescence from dim fluorophores, but it demands reference spectra for every dye and an accurate autofluorescence reference, and unmixing errors propagate into all channels.

Controls: FMO, isotype, and the gate that needs evidence

The gate boundary is the highest-stakes decision in a flow experiment, and the correct control is the fluorescence-minus-one (FMO). An FMO sample contains all antibodies except the one whose channel is being gated; it therefore reproduces the exact spillover background in that channel. Gating on unstained cells or on isotype controls systematically over- or under-estimates the real background, because isotype controls differ in antibody load and fixation behaviour from the test antibody.

The standard control set:

As panels grow, FMO controls become laborious but not optional; for eight or more colours, an automated or bead-based alternative such as the isotype-stained cells or the "fluorescence-minus-one with all other channels at full load" approach is a reasonable substitute if documented.

Surface, intracellular, and phospho-flow

Surface staining is performed on live cells at 4 °C to prevent internalisation and capping. Intracellular staining requires fixation and permeabilisation, which increases nonspecific binding and requires a fully optimised blocking step. Phospho-flow detects phosphorylation states and is the most demanding application: cells must be fixed within seconds of stimulation (pre-warmed paraformaldehyde directly into the culture), because phosphatase activity erases the signal during handling. Controls for phospho-flow must be stimulated and unstimulated cells processed identically, and the readout is usually a ratio of median fluorescence intensity rather than a gate.

For all intracellular assays, the fixation step itself can alter light scatter and autofluorescence, so compensation and gating should be re-verified after fixing, not carried over from the live-cell panel.

Sorting: purity versus yield, and the trade-off you cannot escape

FACS sorting has three performance metrics that trade against each other: purity, yield (recovery of desired cells), and throughput. Maximising purity means discarding events near the gate boundary and running at a lower event rate to reduce coincident-abort events; maximising yield means accepting looser gates and accepting contamination. The machine's "yield mask" and "purity mask" settings make this choice explicit.

The most common sorting failure is a shift in the population during the run — cells settle, the sample warms, or the laser drifts — so gates set at the start no longer match the sample at the end. Monitor scatter and fluorescence over time, and re-acquire a small sample at the end to check that the population has not moved.

High-dimensional analysis: from gates to embeddings

With 10–50 parameters, manual gating becomes untenable, and dimensionality reduction (t-SNE, UMAP) plus clustering (FlowSOM, PhenoGraph, CITRUS) is now standard. These methods find structure that manual gating misses, but they are also easy to misuse. Claims of a new subset should be verified by returning to the raw channel values and confirming that the cluster is defined by a coherent marker combination, not by an artefact of the embedding parameters. The dimensionality-reduction parameters (perplexity, number of neighbours) change the picture, so they must be reported and the result shown to be robust to reasonable changes. Batch effects from staining or acquisition day are a major source of false clusters; include a common reference sample in every batch and, where possible, use a curated panel with validated antibody clones and a fixed instrument setup across the whole study.

References

  1. Herzenberg LA, Parks D, Sahaf B, et al. The history and future of the fluorescence activated cell sorter and flow cytometry: a view from Stanford. Clin Chem. 2002;48(10):1819–1827.
  2. Maecker HT, Frey T, Nomura LE, Trotter J. Selecting fluorochrome conjugates for maximum sensitivity. Cytometry A. 2004;62(2):169–173.
  3. Perfetto SP, Chattopadhyay PK, Roederer M. Seventeen-colour flow cytometry: unravelling the immune system. Nat Rev Immunol. 2004;4(8):648–655.
  4. Roederer M. Compensation is not dependent on signal intensity in logarithmic space. Cytometry A. 2001;43(2):1–4.
  5. Nolan JP, Condello D. Spectral flow cytometry. Curr Protoc Cytom. 2013;Chapter 1:Unit1.27.
  6. Krutzik PO, Nolan GP. Intracellular phospho-protein staining techniques for flow cytometry: monitoring single cell signaling events. Cytometry A. 2003;55(2):61–70.
  7. Van Gassen S, Callebaut B, Van Helden MJ, et al. FlowSOM: using self-organizing maps for visualization and interpretation of cytometry data. Cytometry A. 2015;87(7):636–645.
  8. Levine JH, Simonds EF, Bendall SC, et al. Data-driven phenotypic dissection of AML reveals progenitor-like cells that correlate with prognosis. Cell. 2015;162(1):184–197.
  9. Aghaeepour N, Finak G, FlowCAP Consortium, et al. Critical assessment of automated flow cytometry data analysis techniques. Nat Methods. 2013;10(3):228–238.
  10. Cossarizza A, Chang HD, Radbruch A, et al. Guidelines for the use of flow cytometry and cell sorting in immunological studies. Eur J Immunol. 2017;47(10):1584–1797.