On multiparameter data analysis in flow cytometry

Increasing numbers of parameters that are accessible to simultaneous measurement in flow cytometric instruments, combined with the extremely large sample sizes common in flow cytometry, make it necessary to examine methods of multivariate statistics for their applicability to problems of visualization and quantitative analysis of flow cytometric data. This article describes some approaches to dimensionality reduction that appear well suited for data sets obtained by flow cytometry.     

Optimizing transformations for automated,high throughput analysis of flow cytometry data

Background  

In a high throughput setting, effective flow cytometry data analysis depends heavily on proper data preprocessing. While usual preprocessing steps of quality assessment, outlier removal, normalization, and gating have received considerable scrutiny from the community, the influence of data transformation on the output of high throughput analysis has been largely overlooked. Flow cytometry measurements can vary over several orders of magnitude, cell populations can have variances that depend on their mean fluorescence intensities, and may exhibit heavily-skewed distributions. Consequently, the choice of data transformation can influence the output of automated gating. An appropriate data transformation aids in data visualization and gating of cell populations across the range of data. Experience shows that the choice of transformation is data specific. Our goal here is to compare the performance of different transformations applied to flow cytometry data in the context of automated gating in a high throughput, fully automated setting. We examine the most common transformations used in flow cytometry, including the generalized hyperbolic arcsine, biexponential, linlog, and generalized Box-Cox, all within the BioConductor flowCore framework that is widely used in high throughput, automated flow cytometry data analysis. All of these transformations have adjustable parameters whose effects upon the data are non-intuitive for most users. By making some modelling assumptions about the transformed data, we develop maximum likelihood criteria to optimize parameter choice for these different transformations.     

A novel statistical analysis and interpretation of flow cytometry data

A recently developed class of models incorporating the cyton model of population generation structure into a conservation-based model of intracellular label dynamics is reviewed. Statistical aspects of the data collection process are quantified and incorporated into a parameter estimation scheme. This scheme is then applied to experimental data for PHA-stimulated CD4+T and CD8+T cells collected from two healthy donors. This novel mathematical and statistical framework is shown to form the basis for accurate, meaningful analysis of cellular behaviour for a population of cells labelled with the dye carboxyfluorescein succinimidyl ester and stimulated to divide.     

Multi-user system for analysis of data from flow cytometry.

A new program is described for the analysis of DNA histograms from flow cytometry. The fundamental model representing the cell population is similar to one described previously. It assumes the population is grouped into compartments, each consisting of cells having approximately the same DNA content. After staining the cells with an appropriate fluorochrome, the fluorescence distribution of cells within each compartment is assumed to be Gaussian. In the present algorithm, the parameters of the model can either be computed directly by the program from the data, or can be specified as input by the user. When synchronous cell populations lacking distinct G1 and G2/M phases are analyzed, the parameter values must first be obtained using an appropriate control. Percentages of cells in the various compartments are computed using a gradient search method described by Bevington.     

Modified histogram subtraction technique for analysis of flow cytometry data

Analysis of flow cytometry histogram data by the subjective selection of an integration window can be a tedious and time-consuming task and is often inaccurate. A new method for automated calculation of the percent positive from immunofluorescence histograms is presented. This new method is a modification of the currently used method of channel-by-channel histogram subtraction. Its accuracy is compared to that of the channel-by-channel histogram subtraction method and to another currently used automated method, which selects an integration window by finding the channels that contain the most fluorescent 2% of a control histogram. The new histogram subtraction method is objective, easy to use, and is more accurate than other currently used automated analysis methods. PASCAL source code is given for each method of analysis.     

On-line three-parameter data uptake, analysis, and display device for flow cytometry and other applications

Multiparameter flow cytometric measurements are of growing interest in the study of complex features of biological cells. With state of the art instrumentation, three-parameter (3-P) data handling is relatively complicated and time consuming and the display methods are not satisfactory. As an alternative, an interactive 3-P analyzing module, Cytomic 123 is described, which displays 3-P fields during and immediately after data uptake in the form of a cubic array of 32,768 channels. The fields can be randomly rotated by hardware and software. The event frequencies in the field are primarily visualized by brightness modulation of the display dots. Additionally, the display of the field may be confined to user selected ranges of event frequencies, which may also be superposed to mixed frequency displays. A set of preprogrammed functions is available for the following tasks: (a) uptake of 3-P histograms combined with on-line control of the transducer pulses, (b) automatic uptake of a series of 2-P time correlated histograms in the cube, (c) generation and numerical evaluation of sections and projections of cube histograms, (d) interactive generation and evaluation of spatial subfields for integration, or as sorting matrix by successive erosion of section planes, or reprojection of projection windows, and (e) isometric display of sections and projections and exchange of data sets with other Cytomic modules or other data systems, especially the Cytomic 12 module, whose 2-P capabilities can be used. The module is built with low cost Z80 microprocessor eurocards. A standard oscilloscope serves as a display unit.     

DNA analysis by flow cytometry

Accurate quantification of DNA from cells of several species is possible with flow cytometry. When one species is used as a reference, cytometric readings from two or more different species can be compared to obtain relative percent DNA or DNA indices. Differences in DNA from the male and female of the same species also can be measured. The method allows rapid screening of chromosomal abnormalities among large clinical populations, and evaluation of errors of sex determination such as XY sex reversal.     

Nonparametric flow cytometry analysis.

A nonparametric statistical test for the analysis of flow cytometry derived histograms is presented. The method involves smoothing and translocation of data, area normalization, channel by channel determination of the mean and S.D., and use of Bayes' theorem for unknown histogram classification. With this statistical method, different sets of histograms from numerous biological systems can be compared.     

Introduction to flow cytometry data file standard

The Data File Standards Committee of the Society for Analytical Cytology presents a Standard to be used for the storage of data associated with flow cytometric measurements. The Standard specifies a format that provides for the inclusion of all information necessary to fully describe: 1) the instrument used for the measurement; 2) the sample measured; 3) the data obtained; and 4) the results of analysis of the data. The Committee and the Society for Analytical Cytology point out that the use of this Standard by all those individuals and companies that generate or use data taken with flow cytometers or generate methods of analysis for the data will encourage the sharing of such data and methods of analysis.     

Nonparametric discriminant analysis of phytoplankton species using data from analytical flow cytometry

BACKGROUND: Analytical flow cytometry (AFC) provides rapid and accurate measurement of particles from heterogeneous populations. AFC has been used to classify and identify phytoplankton species, but most methods of discriminant analysis of resulting data have depended on normality assumptions and outcomes have been disappointing. METHODS AND RESULTS: In this study, we consider nonparametric methods based on density estimation. In addition to the familiar kernel method, methods based on wavelets are also implemented. Full five-dimensional wavelet estimation proves to be computationally prohibitive with current workstation power, so we employ projection pursuit for reduction of dimensionality. AFC typically produces very large samples, so we also investigate data simplification through binning. Further modifications to the discrimination strategy are suggested by specific features of phytoplankton data, namely, a hierarchical group structure, the possible presence of many groups, and the likelihood of encountering an aberrant group in a test sample. CONCLUSIONS: We apply all the resultant procedures to appropriate subsets of a very large data set, demonstrate their efficacy, and compare their error rates with those of more conventional methods. We further show that incorporation of the specific features of phytoplankton data into the analysis leads to improved results and provides a general framework for analysis of such data.     

Using flowViz to visualize flow cytometry data

Summary: Automated analysis of flow cytometry (FCM) data isessential for it to become successful as a high throughput technology.We believe that the principles of Trellis graphics can be adaptedto provide useful visualizations that can aid such automation.In this article, we describe the R/Bioconductor package flowVizthat implements such visualizations. Availability: flowViz is available as an R package from theBioconductor project: http://bioconductor.org Contact: dsarkar{at}fhcrc.org Associate Editor: Olga Troyanskaya     

Recent advances in microbial community analysis from machine learning of multiparametric flow cytometry data

1. Download : Download high-res image (105KB)
2. Download : Download full-size image

     

Modeling flow cytometry data for cancer vaccine immune monitoring

Flow cytometry (FCM) is widely used in cancer research for diagnosis, detection of minimal residual disease, as well as immune monitoring and profiling following immunotherapy. In all these applications, the challenge is to detect extremely rare cell subsets while avoiding spurious positive events. To achieve this objective, it helps to be able to analyze FCM data using multiple markers simultaneously, since the additional information provided often helps to minimize the number of false positive and false negative events, hence increasing both sensitivity and specificity. However, with manual gating, at most two markers can be examined in a single dot plot, and a sequential strategy is often used. As the sequential strategy discards events that fall outside preceding gates at each stage, the effectiveness of the strategy is difficult to evaluate without laborious and painstaking back-gating. Model-based analysis is a promising computational technique that works using information from all marker dimensions simultaneously, and offers an alternative approach to flow analysis that can usefully complement manual gating in the design of optimal gating strategies. Results from model-based analysis will be illustrated with examples from FCM assays commonly used in cancer immunotherapy laboratories.     

Chromosome analysis by high illumination flow cytometry

Fluorescence measurements from metaphase chromosomes of the Chinese hamster, stained with propidium iodide excited at high illumination irradiance, completely resolve each chromosome type. The measurements are performed in a specially designed flow cytometer that achieves high irradiance (4 MW/cm2) by using high power laser output (2 W at 488 nm) focused to small spot size (1% irradiance variation over 2 microns). The coefficient of variation of each chromosome peak is near 1.5%. Saturation of the fluorescence transition and photobleaching, two consequences of high irradiance, are shown to occur. Even with a nonlinear dependence of fluorescence upon illumination irradiance, fluorescence retains a proportional response to chromosome type; each chromosome peak maintains a consistent ratio to the others at every irradiance. No perturbation of fluorescence by the optical or geometrical properties of the chromosomes is evident. The advantages of high irradiance illumination are an increase in fluorescence sufficient to reduce the statistical error in photoelectron number to a low level and reduced influence of laser power fluctuations and variable chromosome flow trajectories on the precision. These benefits improve the resolution of chromosome analysis by flow cytometry, particularly the resolution of smaller chromosomes.     

Machine learning analysis of microbial flow cytometry data from nanoparticles,antibiotics and carbon sources perturbed anaerobic microbiomes

Background

Flow cytometry, with its high throughput nature, combined with the ability to measure an increasing number of cell parameters at once can surpass the throughput of prevalent genomic and metagenomic approaches in the study of microbiomes. Novel computational approaches to analyze flow cytometry data will result in greater insights and actionability as compared to traditional tools used in the analysis of microbiomes. This paper is a demonstration of the fruitfulness of machine learning in analyzing microbial flow cytometry data generated in anaerobic microbiome perturbation experiments.

Results

Autoencoders were found to be powerful in detecting anomalies in flow cytometry data from nanoparticles and carbon sources perturbed anaerobic microbiomes but was marginal in predicting perturbations due to antibiotics. A comparison between different algorithms based on predictive capabilities suggested that gradient boosting (GB) and deep learning, i.e. feed forward artificial neural network with three hidden layers (DL) were marginally better under tested conditions at predicting overall community structure while distributed random forests (DRF) worked better for predicting the most important putative microbial group(s) in the anaerobic digesters viz. methanogens, and it can be optimized with better parameter tuning. Predictive classification patterns with DL (feed forward artificial neural network with three hidden layers) were found to be comparable to previously demonstrated multivariate analysis. The potential applications of this approach have been demonstrated for monitoring the syntrophic resilience of the anaerobic microbiomes perturbed by synthetic nanoparticles as well as antibiotics.

Conclusion

Machine learning can benefit the microbial flow cytometry research community by providing rapid screening and characterization tools to discover patterns in the dynamic response of microbiomes to several stimuli.

     

Chromosome banding analysis by slit-scan flow cytometry

We have investigated the use of fluorescence banding patterns for the resolution of metaphase chromosomes by slit-scan flow cytometry. Fluorescence scans of R-banded chromosomes have been obtained for the entire human karyotype. Metaphase chromosomes were R-banded in suspension by staining with chromomycin A3 after hypotonic treatment in Ohnuki's buffer. Specific fluorescent landmark bands were detected for human chromosomes 1-12. Scans obtained for chromosomes 13-22 did not contain sufficient information for classification. Characteristic fluorescence patterns for human chromosomes 1 and 3 provided the clearest evidence for the detection of R-bands by slit-scan flow cytometry. Specific patterns were detected for human chromosomes 9-12 in which the number and placement of the fluorescent bands served as classifiers.     

Microscope-based multiparameter laser scanning cytometer yielding data comparable to flow cytometry data

We describe a computer-controlled 10 microns spot size laser scanning cytometer for making multiple wavelength fluorescence and scatter measurements of unconstrained cells on a surface such as a microscope slide. Designated areas of slides placed on a microscope stage are automatically scanned, and cells which generate above-threshold scatter or fluorescence values are found and individually processed to determine a list of measurement parameters. For each fluorescence or scatter measurement parameter, this list contains the integrated and peak values and bit pattern images of a scan window centered on the cell. The measurement time, the position of the cell on the slide, and two segmentation indices are also included in the list. Measurement time, cell position, and properties derived from the bit patterns are used interchangeably with integrated or peak measurement values as coordinates of multiproperty displays. Cells may be selected for counting, data display in various forms, or visual observation based on their meeting complex criteria among a chain of two property screens. Cells with selected properties may be viewed during an experiment or retrospectively. A designated specimen field may be repeatedly remeasured to perform kinetic cell studies. An argon ion and a HeNe- based laser instrument have been constructed and software has been written and evaluated with the specific goal of increasing the precision of propidium iodide-stained cellular DNA measurements. Some of the capabilities of the instrument and its current performance are described.