FACS–scRNA Integration
Turn your FlowJo export into publication-ready statistics: which populations and markers differ between conditions, how they relate to clinical scores, and whether flow and single-cell RNA from the same animals tell the same story.
Runs in your browser · no upload · works from FlowJo table exportsLoading this browser tool…
How to use it
1. Export from FlowJo. In the Table Editor add the gates and statistics you need (Freq. of Parent, Count, Median or Geometric Mean of a marker) and export as CSV with samples as rows. Include a Count for each tested gate and its parent so frequencies can be modelled from event counts. 2. Add a sample sheet with sample (FCS name), condition and optionally animal_id, batch and numeric variables such as a clinical score. 3. Review the detected statistic types, reference condition, pairing and covariates, add ratios such as Th17/Treg, and optionally load scRNA pseudobulk to compare populations and markers with mRNA. 4. Read the results: plain-language findings, volcano plot, per-statistic dot plots, correlation, PCA and heatmap, and the scRNA comparison. Export statistics as CSV, plots as SVG/PNG and a methods paragraph.
What it computes
Frequencies are compared on the logit scale with a quasi-binomial generalized linear model on event counts (like R’s glm(family = quasibinomial)), which weights each sample by its parent count and estimates the extra variation between animals; without counts it uses an empirical-logit linear model. MFIs are compared as log2 fold changes of geometric means (or on the asinh scale for negative values). Paired designs use animal as a blocking factor and covariates such as batch or sex enter the model. P values are Benjamini–Hochberg adjusted within each statistic type. Optional correlations use Spearman with a condition-adjusted partial correlation. scRNA integration maps gates to annotated cell types and compares proportions, marker protein with mRNA, and condition effects side by side.
Limits
The tool analyses gated summaries; gating, compensation and population definitions remain yours and determine the results. MFIs are only comparable within one acquisition batch or after normalization. Event counts depend on acquisition and are not tested unless you choose to. Flow proportions and droplet scRNA proportions differ systematically, and intracellular cytokine staining after restimulation is not expected to match ex vivo mRNA. Statistics support, but do not prove, biological differences.
Keep your work
Nothing is stored or uploaded; reloading or “Delete all loaded data” clears the analysis. Export CSV, SVG/PNG, JSON and the methods text before closing the page.
Privacy and local-data controls · Credits and licenses · Interpretation limits