Modeling of inter‐sample variation in flow cytometric data with the joint clustering and matching procedure

Abstract

We present an algorithm for modeling flow cytometry data in the presence of large inter-sample variation. Large-scale cytometry datasets often exhibit some within-class variation due to technical effects such as instrumental differences and variations in data acquisition, as well as subtle biological heterogeneity within the class of samples. Failure to account for such variations in the model may lead to inaccurate matching of populations across a batch of samples and poor performance in classification of unlabeled samples. In this paper, we describe the Joint Clustering and Matching (JCM) procedure for simultaneous segmentation and alignment of cell populations across multiple samples. Under the JCM framework, a multivariate mixture distribution is used to model the distribution of the expressions of a fixed set of markers for each cell in a sample such that the components in the mixture model may correspond to the various populations of cells, which have similar expressions of markers (that is, clusters), in the composition of the sample. For each class of samples, an overall class template is formed by the adoption of random-effects terms to model the inter-sample variation within a class. The construction of a parametric template for each class allows for direct quantification of the differences between the template and each sample, and also between each pair of samples, both within or between classes. The classification of a new unclassified sample is then undertaken by assigning the unclassified sample to the class that minimizes the distance between its fitted mixture density and each class density as provided by the class templates. For illustration, we use a symmetric form of the Kullback-Leibler divergence as a distance measure between two densities, but other distance measures can also be applied. We show and demonstrate on four real datasets how the JCM procedure can be used to carry out the tasks of automated clustering and alignment of cell populations, and supervised classification of samples. V C 2015International Society for Advancement of Cytometry Key terms flow cytometry; classification; class template; inter-sample variation; clustering; matching; skew mixture models; EM algorithm; JCM FLOW cytometry (FCM) is a powerful tool in clinical diagnosis of health disorders, in particular, immunological diseases. It offers rapid high-throughput measurements of multiple characteristics on every cell in a sample, enabling biologists to study a variety of biological processes at the cellular level. A critical component in the pipeline of FCM data analysis is the identification of cell populations from the multidimensional FCM dataset, currently performed manually by visually separating regions or gates of interest on a series of sequential bivariate projections of the data, a process known as gating. However, this approach has many problems and limitations, including variability between different analysts, non-standardization and nonreproducibility of results, an unrealistic assumption that bi...

Immunophenotype Discovery, Hierarchical Organization, and Template-Based Classification of Flow Cytometry Samples

“…This division and ordering of work in FC data analysis is simply our view to tackle subproblems independently and develop algorithms to automated data analysis. Other researchers have divided these steps into smaller substeps (28), merged multiples steps into one (23), or ordered these steps differently based on the need of a particular experiment. In the rest of this section, we briefly discuss these steps.…”

“…By modeling the inter-sample variation within a class with random-effects terms, they construct a parametric template for each class of samples. These templates are used to classify new samples with high accuracy (23), demonstrating the effectiveness of template-based classifiers in flow cytometry.…”

“…It is more efficient since one compares a sample with a few templates only, rather than with all other samples; it is more robust since a template describes the features common to cell populations in several samples, while ignoring noise and small sample-specific variations. Previous work (3, 15, 19, 23) acknowledged the advantage of this collective approach and developed software to automate this process. In recent work, Lee et al (23) proposed a joint clustering and matching (JCM) algorithm for simultaneous segmentation and alignment of cell populations across multiple samples.…”

“…Previous work (3, 15, 19, 23) acknowledged the advantage of this collective approach and developed software to automate this process. In recent work, Lee et al (23) proposed a joint clustering and matching (JCM) algorithm for simultaneous segmentation and alignment of cell populations across multiple samples. By modeling the inter-sample variation within a class with random-effects terms, they construct a parametric template for each class of samples.…”

High-speed automatic characterization of rare events in flow cytometric data

“…Advances in mass cytometry have promised the ability to determine 50-100 features per cell [2,3]. To address such increasingly multi-parametric and multiplexed immunoprofiling of each cell, studies have demonstrated the critical need for systematic and automated multivariate analysis and visualization suitable for high-dimensional data [4][5][6]. As the number of potential combinations of markers continues to grow exponentially (with the number of markers), a thorough search for rare events in high-dimensional marker-space clearly gets difficult with the more subjective and painstaking approach of traditional manual gating [7].…”