Skip to content

Your first analysis

This walks a single sample from file to exported numbers. It assumes you have installed the app and have either a trial or a license — viewing works without one, but gating does not.

Click Open File in the toolbar (or press Ctrl+O), or drag an FCS file onto the window. You can open several at once.

The file appears in the file list on the left, and the app picks a starting pair of axes. Large files are fine — a million events opens in seconds.

A freshly opened 979,290-event file showing an FSC-A by SSC-A scatter with no gates

979,290 events, compensation applied, no gates yet. The footer reads 50,000 / 979,290 events — the plot draws a subsample for speed, while statistics are computed over every event.

Click either axis label to choose a parameter. Next to the parameter you choose a scale — and for fluorescence channels the right answer is almost never linear.

The app suggests a scale per channel. If a population is smeared against the axis or piled up at zero, the scale is wrong before the gate is wrong. See Scales and transforms.

If the FCS file carries a spillover matrix from your acquisition software, the file is already compensated with it — the file list shows Comp beside it, and the file bar reads Comp: ON. If you have single-stain controls, the app can compute a matrix from them instead.

A file without a matrix opens uncompensated, which is fine for a scatter-only first look. See Compensation.

Pick a gate tool from the gating toolbar — rectangle is the easiest to start with — and drag it around the population you want.

The gate appears in the gate list with its statistics. Draw the hygiene gates first, in the order a cytometrist expects: scatter to exclude debris, then singlets, then live cells.

Double-click a gate to drill into it. The plot now shows only the events inside that gate, and any gate you draw is a child of it. Escape navigates back up.

This is what makes the statistics mean what you expect: a child population’s frequency is a percentage of its parent, and the breadcrumb above the plot shows you which parent you are in.

6. Let the AI propose the rest — optional

Section titled “6. Let the AI propose the rest — optional”

Two different options, documented separately because they behave differently:

  • AI-1 identifies a fixed list of immune populations on your machine, and declines the ones it cannot identify rather than guessing.
  • The AI copilot takes a plain-language request (“gate CD4 T cell memory subsets”) and proposes a full strategy, step by step, for you to approve.

Neither applies anything without your click.

The statistics table lists each population with its count and frequency, plus per-channel statistics. It updates as you move gates — adjust a boundary and the numbers follow.

See Statistics for what each column means and, importantly, what each percentage is a percentage of.

  • Statistics as CSV for your own analysis.
  • Events as CSV, for all events or for one gate.
  • Figures as SVG, PDF or PNG.

See Exporting.

Ctrl+S saves the gate tree, each file’s compensation and your file names — not a copy of your data, a reference to it. Reopening restores the analysis. Saving works with or without a license.

If the FCS files have moved since you saved, the app lists them as missing and lets you locate each one, rather than failing. See Workspaces.

  • The interface — what every panel is for
  • Gating — all five gate types, editing, and automated gating
  • Batch analysis — apply this strategy to a whole experiment