Flow Cytometry Beginner Guide | Why Your Interpretation Is Still Wrong---Even When You "Understand" the Plots
In the previous article, we discussed a basic reading sequence for flow cytometry plots:
start with population gating, then examine marker expression, and finally look at percentages.
However, many researchers encounter a more frustrating issue:
you follow the correct steps, you can "read" the plots, yet your conclusions still don't seem right.
In most cases, the problem is not a lack of understanding---but rather small, easily overlooked details.
In this article, we won't introduce new concepts.
Instead, we focus on one thing: the most common pitfalls for beginners---and what they actually look like in your data.
Pitfall 1: What You Think Are "Cells" May Include a Lot of Debris
A typical first step is to look at the FSC/SSC plot and assume the population looks clean enough to proceed.
But there's a critical assumption behind that step:
are all those events really single, viable cells?
In reality, three types of unwanted events are commonly mixed in:
- Cellular debris
- Dead cells
- Doublets (two cells stuck together)
These can appear deceptively similar to real cells, but they have recognizable patterns:
- Debris: mostly located in the lower-left (low FSC/SSC), but often with a "tail" extending into the main population
- Doublets: higher FSC signals, typically appearing as a "larger cell" population on the right side
- Dead cells: more dispersed, often lacking a clearly defined cluster
If you move directly from FSC/SSC to downstream analysis, you risk analyzing a heterogeneous and unclean population.
A more reliable workflow is:
FSC/SSC → singlet gating → live cell gating → fluorescence analysis
In simple terms:
- Singlet gating removes doublets (commonly via FSC-A vs FSC-H, where true singlets fall along a diagonal)
- Live cell gating excludes dead cells (using viability dyes such as PI or 7-AAD)
Skipping these steps means that many of your downstream "populations" and "percentages" may be fundamentally misleading.
Pitfall 2: Dead Cells Can Create Convincing False Positives
This is one of the most common---and impactful---issues for beginners.
When cells die, membrane integrity is compromised, leading to:
- Increased non-specific antibody entry
- Higher retention of fluorescent signals
- Overall elevation of background fluorescence
On the plot, this typically appears as:
- A rightward shift of the negative population
- Blurred separation between negative and positive populations
- The emergence of a "dim positive" tail
Top Figure (no viability dye):
- Negative population shifts to the right
- Background appears broadly elevated
- Population boundaries are poorly defined
Bottom Figure (after PI exclusion):
- Negative population returns to baseline
- Background is significantly reduced
- Clear separation between populations
This pattern is often misinterpreted as:
"Maybe the marker is just lowly expressed."
But a more likely explanation is:
dead cells are being counted as weak positives.
A practical rule of thumb: if the entire plot looks "slightly bright," suspect dead cells before assigning biological meaning.
Pitfall 3: Defining Positivity Without Proper Controls
As discussed previously, "positive" does not mean "looks brighter"---it means exceeds a defined threshold.
In practice, however, a common mistake is to draw a gate simply because a subset of cells appears slightly brighter than others.
Without proper controls (FMO, isotype, or unstained), it becomes difficult to distinguish:
- Background noise
- Low-level expression
- True positive signal
Typical visual patterns include:
- A continuous distribution (smear or tail) rather than two distinct populations
- A general rightward shift without a clear boundary
If you force a gate under these conditions, you may obtain a percentage that looks reasonable---but lacks reproducibility.
A more robust interpretation guideline:
If there is no clear bimodal distribution,
the "positive population" should be considered tentative, not definitive.
When controls are unavailable, treat results as qualitative trends, rather than quantitative conclusions.
Pitfall 4: Interpreting Percentages Without Context
Percentages are often the most eye-catching numbers in flow cytometry:
"28%", "12%", "65%".
But every percentage is calculated within a specific gate.
For example:
- 60% T cells within the lymphocyte gate
- 60% T cells within total events
These are fundamentally different interpretations.
Ignoring the gating hierarchy can lead to:
- Mistaking subset proportions for total population frequencies
- Invalid comparisons between samples with inconsistent gating strategies
A common issue in plots:
- Focusing only on the percentage value
- Overlooking the parent gate definition
- Ignoring whether upstream gating is consistent across samples
A simple but powerful habit: "Whenever you see a percentage, ask: 'Percentage of what?'"
Pitfall 5: "Double-Positive" May Just Be Signal Spillover
In multicolor flow cytometry, it is common to observe cells that appear positive in two channels.
A natural assumption is:
these cells co-express both markers.
Top Figure (no compensation):
- Large number of events distributed along a diagonal
- Apparent "double-positive" population
Bottom Figure (proper compensation):
- Diagonal pattern disappears
- True single-positive populations become distinct
- Any real double-positive cells form a discrete cluster, if present
However, a more frequent explanation is spectral overlap (spillover) and improper compensation.
In essence:
- Fluorescence from one channel is partially detected in another
- Single-positive cells appear artificially double-positive
Typical visual features include:
- A diagonal distribution from lower-left to upper-right
- A broad population shifting along both axes
- No clearly separated double-positive cluster
A practical rule: If the "double-positive" population forms a continuous diagonal or diffuse region, suspect compensation artifacts before biological interpretation.
Final Thoughts
Looking across these pitfalls, a clear pattern emerges:
Most errors do not come from "misreading plots,"
but from:
- Inadequate data cleaning
- Poorly defined thresholds
A more reliable approach is to ask three questions before interpreting results:
- Are these truly single, viable cells?
- How is "positivity" defined?
- What is the reference population for this percentage?
Clarifying these points will eliminate most conclusions that seem reasonable---but are actually incorrect.
