BayesFlow: latent modeling of flow cytometry cell populations

Abstract

BackgroundFlow cytometry is a widespread single-cell measurement technology with a multitude of clinical and research applications. Interpretation of flow cytometry data is hard; the instrumentation is delicate and can not render absolute measurements, hence samples can only be interpreted in relation to each other while at the same time comparisons are confounded by inter-sample variation. Despite this, most automated flow cytometry data analysis methods either treat samples individually or ignore the variation by for example pooling the data. A key requirement for models that include multiple samples is the ability to visualize and assess inferred variation, since what could be technical variation in one setting would be different phenotypes in another.ResultsWe introduce BayesFlow, a pipeline for latent modeling of flow cytometry cell populations built upon a Bayesian hierarchical model. The model systematizes variation in location as well as shape. Expert knowledge can be incorporated through informative priors and the results can be supervised through compact and comprehensive visualizations.BayesFlow is applied to two synthetic and two real flow cytometry data sets. For the first real data set, taken from the FlowCAP I challenge, BayesFlow does not only give a gating which would place it among the top performers in FlowCAP I for this dataset, it also gives a more consistent treatment of different samples than either manual gating or other automated gating methods. The second real data set contains replicated flow cytometry measurements of samples from healthy individuals. BayesFlow gives here cell populations with clear expression patterns and small technical intra-donor variation as compared to biological inter-donor variation.ConclusionsModeling latent relations between samples through BayesFlow enables a systematic analysis of inter-sample variation. As opposed to other joint gating methods, effort is put at ensuring that the obtained partition of the data corresponds to actual cell populations, and the result is therefore directly biologically interpretable. BayesFlow is freely available at GitHub.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-015-0862-z) contains supplementary material, which is available to authorized users.

Robust calibration of hierarchical population models for heterogeneous cell populations

“…A challenge in the analysis of single-cell data is that the observed cell population is often skewed (Pyne et al, 2009). Therefore, distributions which account for skewness are often employed in the analysis of, e.g., flow cytometry data (Johnsson et al, 2016). The multivariate skew normal distribution has distribution parameters ϕ = (µ, Σ, δ), with location µ ∈ R ny , covariance matrix Σ ∈ R ny×ny and skew parameter δ ∈ R ny .…”

Ultrafast clustering of single-cell ow cytometry data using FlowGrid

“…Flock [8] maintains a high accuracy and reasonable runtime. After the challenge, several algorithms have been built for flow cytometry data analysis such as FlowPeaks [3], FlowSOM [10] and BayesFlow [6].…”

“…BayesFlow uses a Bayesian hierarchical model to identify different cell populations in one or many samples. The key benefit of this method is its ability to incorporate prior knowledge, and captures the variability in shapes and locations of populations between the samples [6]. However, BayesFlow tends to be computational expensive as Markov Chain Monte Carlo sampling requires a large number of iterations.…”

What is a “unimodal” cell population? Using statistical tests as criteria for unimodality in automated gating and quality control

“…In contrast, in all criteria hitherto used for unimodality in flow cytometry data analysis-except (7), which uses the dip test-the generating probability density is estimated explicitly. The method favored by most, used in SWIFT (2), curvHDR (3), flowDensity (4), gaussNorm and fdaNorm (5), and flowMeans (6), is to use a kernel density estimator with a bandwidth selected by some rule of thumb.…”

“…This allows for a large variety of cell population shapes-in particular no assumptions about Gaussianity or other distributional assumptions are made. Counting density peaks has further been used in normalization of flow cytometry data (5), in preparatory steps for automated gating (6), and in quality control of gated populations (7). The criteria for unimodality used for these different purposes are typically based on density estimates, which are essentially smoothed histograms.…”