Manual Gating Research Articles

Improved Detection of Minimal Residual Disease By Flow Cytometry in AML By Combining Manual Gating and Visne Clustering

Background: In acute myeloid leukemia (AML), detection of minimal residual disease (MRD) by flow cytometry is an adverse prognostic factor besides pre-treatment risk classifications, including cytogenetic and molecular aberrations. High dimensional multiparameter flow cytometry (MPFC) offers improved sensitivity and specificity, however manual analysis is increasingly challenging. In this study, we explore the value of the recently proposed viSNE algorithm to quantify MRD levels in patients with AML achieving complete remission (CR) after intensive induction chemotherapy.Methods: Bone marrow samples from patients with AML (excluding patients with acute promyelocytic leukemia) were analyzed by 8-10 MPFC using a NAVIOS flow cytometer (Beckman Coulter, Brea, CA, USA). Only patients achieving a CR or CR with incomplete blood count recovery (CRi) post-induction were included in this analysis. Manual gating of MRD flow data was performed as described previously (Köhnke et al., Leukemia 2014) using a cutoff for MRD positivity of 0.1%. The viSNE algorithm was performed as described previously (Amir et al., Nat. Biotech. 2013) and MRD positivity was defined as the presence of a distinct cluster of >100 cells which consisted of >90% patient cells. Kaplan-Meier estimator and log-rank test as well as Cox's proportional hazards regression model were used to analyze survival data.Results: Post-induction flow cytometry and clinical data of 38 patients with AML achieving CR (n=34) or CRi (n=4) were available for analysis (median age 53 years; de-novo AML n=32, tAML n=1, sAML n=5). Most patients belonged to the intermediate cytogenetic risk group (MRC favorable n=5, intermediate n=22, adverse n=11).19/38 patients were MRD positive post-induction by manual gating. 12/19 patients deemed MRD positive relapsed, whereas 3/19 patients deemed MRD negative relapsed. Therefore, MRD positivity by manual gating correlated with reduced relapse free survival (median RFS for MRD positive patients: 7.5 months vs. median not reached for MRD negative patients, log-rank test p=0.017). For overall survival (OS), no significant impact of MRD positivity could be detected so far (p=0.3), however follow-up was short (median follow-up 9.3 months). MRD positivity by manual gating remained an independent risk factor for RFS (HR 4.8, p=0.021) when compared to genetic risk and age.MRD positivity by viSNE clustering was seen in 19/38 patients. 10/19 patients deemed MRD positive by viSNE relapsed, while 5/19 deemed MRD negative by viSNE experienced a relapse. This resulted in a trend towards shorter RFS for MRD positivity by viSNE (median RFS 9.9 months vs. 19.0 months for MRD negative patients, p=0.185). Among the patients deemed MRD positive by viSNE who did not relapse, i.e. false-positive patients, follow-up was very short (<3.5 months) in 4/9 cases and final judgment whether these patients are truly negative should be withheld. However, differentiation between healthy monocytes and potential MRD cells by viSNE seems to be especially challenging and warrants optimization.Nevertheless, within the group of patients deemed MRD negative by manual gating, viSNE was able to detect all patients who experienced a subsequent relapse. To maximize sensitivity for MRD detection, we therefore combined the results of both methods and defined MRD positivity as positivity by manual gating and/or viSNE clustering. Using this combined strategy, 28/38 patients were defined as MRD positive and 10/38 as MRD negative. Within the MRD positive group, 15/28 patients relapsed, whereas none of the patients in the MRD negative group relapsed (Figure 1, median RFS for MRD positive patients: 7.7 months, log-rank test p=0.028). Of note, within the MRD negative group only 2 patients (both with intermediate genetic risk) went on to receive a allogeneic stem cell transplantation whereas 8 remained in remission with chemotherapy alone.Conclusion: In summary, viSNE clustering used in combination with manual gating improves the sensitivity of MRD detection by flow cytometry. Importantly, viSNE is able to detect MRD positivity within the manual MRD negative patient population. This is the first report of MRD detection in AML using this method. However, improvements in the algorithm as well as further studies are needed to validate the prognostic value of viSNE clustering in AML. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Mapping cell populations in flow cytometry data for cross-sample comparison using the Friedman-Rafsky test statistic as a distance measure.

Flow cytometry (FCM) is a fluorescence‐based single‐cell experimental technology that is routinely applied in biomedical research for identifying cellular biomarkers of normal physiological responses and abnormal disease states. While many computational methods have been developed that focus on identifying cell populations in individual FCM samples, very few have addressed how the identified cell populations can be matched across samples for comparative analysis. This article presents FlowMap‐FR, a novel method for cell population mapping across FCM samples. FlowMap‐FR is based on the Friedman–Rafsky nonparametric test statistic (FR statistic), which quantifies the equivalence of multivariate distributions. As applied to FCM data by FlowMap‐FR, the FR statistic objectively quantifies the similarity between cell populations based on the shapes, sizes, and positions of fluorescence data distributions in the multidimensional feature space. To test and evaluate the performance of FlowMap‐FR, we simulated the kinds of biological and technical sample variations that are commonly observed in FCM data. The results show that FlowMap‐FR is able to effectively identify equivalent cell populations between samples under scenarios of proportion differences and modest position shifts. As a statistical test, FlowMap‐FR can be used to determine whether the expression of a cellular marker is statistically different between two cell populations, suggesting candidates for new cellular phenotypes by providing an objective statistical measure. In addition, FlowMap‐FR can indicate situations in which inappropriate splitting or merging of cell populations has occurred during gating procedures. We compared the FR statistic with the symmetric version of Kullback–Leibler divergence measure used in a previous population matching method with both simulated and real data. The FR statistic outperforms the symmetric version of KL‐distance in distinguishing equivalent from nonequivalent cell populations. FlowMap‐FR was also employed as a distance metric to match cell populations delineated by manual gating across 30 FCM samples from a benchmark FlowCAP data set. An F‐measure of 0.88 was obtained, indicating high precision and recall of the FR‐based population matching results. FlowMap‐FR has been implemented as a standalone R/Bioconductor package so that it can be easily incorporated into current FCM data analytical workflows. © 2015 International Society for Advancement of Cytometry

Duplicate analysis method: a cheaper alternative to commercial IQC materials in limited resource settings for monitoring CD4 testing.

BackgroundIndia has a large number of HIV infected patients being followed up at anti-retroviral therapy (ART) centers. The patients are regularly offered CD4 count estimation for deciding their eligibility for ART initiation as well as for monitoring response to ART, making CD4 count estimation a very critical test. Hence, quality control of CD4 testing is utmost important for ultimate success of ART program. As the commercial controls are very expensive, internal quality control (IQC), at present, is being done by duplicate analysis method using previous day samples in most of the laboratories. Hence the study was undertaken to review performance of duplicate analysis method for monitoring daily IQC.MethodsQuality control (QC) data from 11 Indian laboratories using duplicate analysis and/or commercial controls for IQC of CD4 testing was collected for reviewing information on QC parameters such as precision, accuracy and trend monitoring. Precision was determined by r2 values and mean % variation for duplicate analysis and coefficient of variation (% CV) for commercial controls. Accuracy was monitored by rate of QC failures for both the types of control and trend monitoring was done by plotting LJ charts for commercial controls and by plotting daily % variation for duplicate analysis.ResultsThe laboratories using duplicate analysis for IQC showed good precision with mean % variation ranging from 0.5 to 7.2. There was good match between r2 values and % CV of the laboratories performing both the types of QC methods. Rates of QC failures were 2.3 for duplicate analysis and 3 per laboratory-year for IMMUNO-TROL controls. Daily trend monitoring showed fluctuation of daily counts around mean in LJ charts and of percent variation around 0% in duplicate analysis method. Commercially available controls showed limitations such as altered specimen quality leading to difficulties in manual gating and issues with the establishment of laboratory range.ConclusionDuplicate analysis can serve as a cheaper alternative to commercially available controls for IQC of CD4 testing especially when supplemented with other QC measures for controlling variations caused by reagent, equipment, staff and environment in addition to the successful participation in External Quality Assurance programme.

Algorithmic Tools for Mining High-Dimensional Cytometry Data.

The advent of mass cytometry has led to an unprecedented increase in the number of analytes measured in individual cells, thereby increasing the complexity and information content of cytometric data. Although this technology is ideally suited to the detailed examination of the immune system, the applicability of the different methods for analyzing such complex data is less clear. Conventional data analysis by manual gating of cells in biaxial dot plots is often subjective, time consuming, and neglectful of much of the information contained in a highly dimensional cytometric dataset. Algorithmic data mining has the promise to eliminate these concerns, and several such tools have been applied recently to mass cytometry data. We review computational data mining tools that have been used to analyze mass cytometry data, outline their differences, and comment on their strengths and limitations. This review will help immunologists to identify suitable algorithmic tools for their particular projects.

An Introduction to Automated Flow Cytometry Gating Tools and Their Implementation.

Current flow cytometry (FCM) reagents and instrumentation allow for the measurement of an unprecedented number of parameters for any given cell within a homogenous or heterogeneous population. While this provides a great deal of power for hypothesis testing, it also generates a vast amount of data, which is typically analyzed manually through a processing called “gating.” For large experiments, such as high-content screens, in which many parameters are measured, the time required for manual analysis as well as the technical variability inherent to manual gating can increase dramatically, even becoming prohibitive depending on the clinical or research goal. In the following article, we aim to provide the reader an overview of automated FCM analysis as well as an example of the implementation of FLOw Clustering without K, a tool that we consider accessible to researchers of all levels of computational expertise. In most cases, computational assistance methods are more reproducible and much faster than manual gating, and for some, also allow for the discovery of cellular populations that might not be expected or evident to the researcher. We urge any researcher who is planning or has previously performed large FCM experiments to consider implementing computational assistance into their analysis pipeline.

A time to amaze, a time to settle down, and a time to discover.

A time to amaze, a time to settle down, and a time to discover.

Detection of minimal residual disease in B lymphoblastic leukemia using viSNE.

Minimal residual disease (MRD) following treatment is a robust prognostic marker in B lymphoblastic leukemia. However, the detection of MRD by flow cytometric immunophenotyping is technically challenging, and an automated method to detect MRD is therefore desirable. viSNE, a recently developed computational tool based on the t-Distributed Stochastic Neighbor Embedding (t-SNE) algorithm, has been shown to be capable of detecting synthetic "MRD-like" populations of leukemic cells created in vitro, but whether viSNE can facilitate the immunophenotypic detection of MRD in clinical samples has not been evaluated. We applied viSNE retrospectively to 8-color flow cytometric immunophenotyping data from normal bone marrow samples, and samples from B lymphoblastic leukemia patients with or without suspected MRD on the basis of conventional manual gating. In each of 14 bone marrow specimens containing MRD or suspected MRD, viSNE identified a putative MRD population; an abnormal composite immunophenotype was confirmed for the putative MRD in each case. MRD populations were not identified by viSNE in control bone marrow samples from patients with increased normal B-cell precursors, or in post-treatment samples from B lymphoblastic leukemia patients who did not have detectable MRD by manual gating. viSNE shows promise as an automated method to facilitate immunophenotypic MRD detection in patients treated for B lymphoblastic leukemia.

AUTOMATED GATING OF PORTABLE CYTOMETER DATA BASED ON SKEW t MIXTURE MODELS

A major component of flow cytometry (FCM) data analysis involves gating, which is the process of identifying homogeneous groups of cells. With the rapid development of the portable flow cytometer, manual gating techniques have been unable to meet the demand for accurate and rapid analysis of samples. To provide a practical application for portable devices, we propose a flexible, statistical model-based clustering approach for identifying cell populations in FCM data. This approach, which mimics the manual gating process, employs a finite mixture model with a density function of skew t distribution and estimates parameters via an expectation maximization algorithm. Data analysis from an experiment on a patient’s peripheral blood samples have proven that the proposed methodology yields better results in terms of robustness against outliers than current state-of-the-art automated gating methods, has more flexibility in clustering symmetric data and leads to lower misclassification rates (misclassification rates of skew t method is 0.06442) when handling highly asymmetric data. The method we proposed will improve data analysis of portable flow cytometers, especially when the users have no professional training.

ImmunoClust--An automated analysis pipeline for the identification of immunophenotypic signatures in high-dimensional cytometric datasets.

Multiparametric fluorescence and mass cytometry offers new perspectives to disclose and to monitor the high diversity of cell populations in the peripheral blood for biomarker research. While high-end cytometric devices are currently available to detect theoretically up to 120 individual parameters at the single cell level, software tools are needed to analyze these complex datasets automatically in acceptable time and without operator bias or knowledge. We developed an automated analysis pipeline, immunoClust, for uncompensated fluorescence and mass cytometry data, which consists of two parts. First, cell events of each sample are grouped into individual clusters. Subsequently, a classification algorithm assorts these cell event clusters into populations comparable between different samples. The clustering of cell events is designed for datasets with large event counts in high dimensions as a global unsupervised method, sensitive to identify rare cell types even when next to large populations. Both parts use model-based clustering with an iterative expectation maximization algorithm and the integrated classification likelihood to obtain the clusters. A detailed description of both algorithms is presented. Testing and validation was performed using 1) blood cell samples of defined composition that were depleted of particular cell subsets by magnetic cell sorting, 2) datasets of the FlowCAP III challenges to identify populations of rare cell types and 3) high-dimensional fluorescence and mass-cytometry datasets for comparison with conventional manual gating procedures. In conclusion, the immunoClust-algorithm is a promising tool to standardize and automate the analysis of high-dimensional cytometric datasets. As a prerequisite for interpretation of such data, it will support our efforts in developing immunological biomarkers for chronic inflammatory disorders and therapy recommendations in personalized medicine. immunoClust is implemented as an R-package and is provided as source code from www.bioconductor.org.

Computational prediction of manually gated rare cells in flow cytometry data.

Rare cell identification is an interesting and challenging question in flow cytometry data analysis. In the literature, manual gating is a popular approach to distill flow cytometry data and drill down to the rare cells of interest, based on prior knowledge of measured protein markers and visual inspection of the data. Several computational algorithms have been proposed for rare cell identification. To compare existing algorithms and promote new developments, FlowCAP-III put forward one computational challenge that focused on this question. The challenge provided flow cytometry data for 202 training samples and two manually gated rare cell types for each training sample, roughly 0.02 and 0.04% of the cells, respectively. In addition, flow cytometry data for 203 testing samples were provided, and participants were invited to computationally identify the rare cells in the testing samples. Accuracy of the identification results was evaluated by comparing to manual gating of the testing samples. We participated in the challenge, and developed a method that combined the Hellinger divergence, a downsampling trick and the ensemble SVM. Our method achieved the highest accuracy in the challenge.

Data analysis as a source of variability of the HLA-peptide multimer assay: from manual gating to automated recognition of cell clusters.

Multiparameter flow cytometry is an indispensable method for assessing antigen-specific T cells in basic research and cancer immunotherapy. Proficiency panels have shown that cell sample processing, test protocols and data analysis may all contribute to the variability of the results obtained by laboratories performing ex vivo T cell immune monitoring. In particular, analysis currently relies on a manual, step-by-step strategy employing serial gating decisions based on visual inspection of one- or two-dimensional plots. It is therefore operator dependent and subjective. In the context of continuing efforts to support inter-laboratory T cell assay harmonization, the CIMT Immunoguiding Program organized its third proficiency panel dedicated to the detection of antigen-specific CD8(+) T cells by HLA-peptide multimer staining. We first assessed the contribution of manual data analysis to the variability of reported T cell frequencies within a group of laboratories staining and analyzing the same cell samples with their own reagents and protocols. The results show that data analysis is a source of variation in the multimer assay outcome. To evaluate whether an automated analysis approach can reduce variability of proficiency panel data, we used a hierarchical statistical mixture model to identify cell clusters. Challenges for automated analysis were the need to process non-standardized data sets from multiple centers, and the fact that the antigen-specific cell frequencies were very low in most samples. We show that this automated method can circumvent difficulties inherent to manual gating strategies and is broadly applicable for experiments performed with heterogeneous protocols and reagents.

Monocytes in Icelandic horses: comparison between automatic gating by the ADVIA 2120i, a manual gating method and microscopic differentiation

Inspection of the ADVIA 2120i PEROX cytograms from healthy Icelandic horses revealed a mismatch of the automatic gating (A) for monocytes, resulting in a lower percentage. The raw data of 40 Icelandic horses were transformed to FCS files. Data were imported to WinMDI 2.9 (Purdue Cytometry Laboratories) and gated manually (M). In addition, a manual 200-cell differential (D) was performed. ANOVA was used to test for significant differences. RefVal 2.1 (Toulouse Veterinary School) was used to calculate the reference values by different methods. Significance level was set at P < 0.05. Monocytes were significantly higher by M compared to A and D, neutrophils were significantly lower for A compared to M, but not for A compared to D. Data were used to calculate the RI as follows: neutrophils (A) 37.7–65.6 % (1.7–4.6 × 109/L); neutrophils (M): 33.8–61.8 % (1.6–4.3 × 109/L); lymphocytes (A): 26.0–52.7 % (1.1–4.6 × 109/L), lymphocytes (M): 28.1–55.5 % (1.2–4.0 × 109/L); monocytes (A): 1.5–4.6 % (0.1–0.4 × 109/L); monocytes (M): 3.4–6.9 % (0.2–0.5 × 109/L). The ADVIA 2120i built-in export function and the use of WinMDI allow a very precise recalculation of cell populations. Icelandic horses reportedly have the lowest monocyte number of all horse breeds. This may be possibly caused by higher MPO activity and/or an inaccuracy of the algorithm used.

Automated flow cytometric analysis across large numbers of samples and cell types

Multi-parametric flow cytometry is a key technology for characterization of immune cell phenotypes. However, robust high-dimensional post-analytic strategies for automated data analysis in large numbers of donors are still lacking. Here, we report a computational pipeline, called FlowGM, which minimizes operator input, is insensitive to compensation settings, and can be adapted to different analytic panels. A Gaussian Mixture Model (GMM)-based approach was utilized for initial clustering, with the number of clusters determined using Bayesian Information Criterion. Meta-clustering in a reference donor permitted automated identification of 24 cell types across four panels. Cluster labels were integrated into FCS files, thus permitting comparisons to manual gating. Cell numbers and coefficient of variation (CV) were similar between FlowGM and conventional gating for lymphocyte populations, but notably FlowGM provided improved discrimination of “hard-to-gate” monocyte and dendritic cell (DC) subsets. FlowGM thus provides rapid high-dimensional analysis of cell phenotypes and is amenable to cohort studies.

The Ratio Controller for Regulation of Turnout Flow Rate

AbstractSetting and adjusting flow rates through turnout gates is a common task for canal operators and irrigators. When downstream flow measurement is available, the flow source is a relatively stable canal, and the turnout is controlled by a sluice gate for which flow rate is approximately proportional to gate opening, a simple ratio calculation can be used to estimate the change of gate position needed to reach a desired flow rate. This method can be used to estimate the next necessary manual gate adjustment or it can be incorporated into an automated ratio controller. The control method is analysed to demonstrate its stability and example applications are modelled to compare its performance to that of a typical proportional controller. The method generally performs as well as the proportional controller, with no need for empirical tuning. The one tuning parameter required by the method is the selection of a control cycle time. Cycle time is related to the settling time required for the canal reach between the turnout gate and the measurement device to attain equilibrium following a flow change. A design parameter for estimating a stable control cycle time is proposed. Copyright © 2015 John Wiley &amp; Sons, Ltd.

Profiling Myelodysplastic Syndromes By Mass Cytometry Demonstrates Distinct Immunophenotypic Aberrancies in Stem and Progenitor Populations

Introduction: Flow cytometry is commonly used to characterize bone marrow (BM) cells of patients with myelodysplastic syndrome (MDS). However the diagnostic utility of this technique has been limited. To address this, we utilized 31-parameter single cell mass cytometry (MCM) to comprehensively analyze primary MDS BM samples.Methods: Expression levels of 31 surface markers, including most previously reported aberrant markers in MDS, were measured on 30 whole BM samples from 10 patients with higher-risk MDS (HR-MDS; IPSS = Int2/High/RAEB-T), 10 with lower-risk MDS (LR-MDS; IPSS = Low/Int1), and 3 patients with non-clonal cytopenias. In addition, 5 BM samples from normal donors were simultaneously analyzed as internal reference comparisons. All samples were barcoded, such that 20 samples (MDS and healthy) could be combined into a single tube for simultaneous antibody staining and analysis. Aberrant marker expression was defined as a median expression level falling outside 4 times the absolute variance of the normal samples in each gated population. Further analysis compared manual gating with unsupervised clustering (spanning tree progression analysis of density-normalized events [SPADE]).Results: MCM analysis generated 31-parameter single-cell data that defined 28 major immunophenotypic populations for each sample. This enabled detection of an aberrant expression of 25/31 markers in at least one population, encompassing essentially every previously reported surface marker aberrancy in MDS. Additionally, 3 previously unrecognized aberrant expression patterns were identified by both manual gating and SPADE: increased CD321 (64% of samples) and CD99 (36% of samples); and decreased CD47 (14% of samples).We focused further analyses on the stem and progenitor cell compartment (HSC, MPP, CMP), in which 20 of the 22 MDS samples exhibited at least one aberrancy (average 2.7) in one of these 3 populations (RAEB-T samples exhibited an average of 4). By contrast, no aberrancies were detected within these populations in the 3 samples from patients with non-clonal cytopenias. In addition to the identification of aberrant expression patterns within the subdivided stem and progenitor cell populations (HSPC) of individual samples, analysis of the HSPC population (CD34+CD38low) as a whole, revealed significant increases (~2-fold) in median expression of CD117 (p=0.003) and HLA-DR (p=0.028) for MDS samples compared to normal. Differences in CD117 and HLA-DR could also be appreciated as aberrant expression patterns (outside 4-fold the variance of normal) in 12/22 and 13/22 samples, respectively. Comparison of marker expression within the HSPCs between patients with HR-MDS and LR-MDS also revealed significant differences. HR-MDS HSPCs were characterized by a ~2-fold increase in CD99 compared to LR-MDS (p=0.0018) and a ~3-fold decrease in CD45 compared to LR-MDS (p=8.8x10-5). Differences in CD99 and CD45 could also be appreciated as aberrant expression patterns in 7/12 and 6/12 of the HR-MDS samples, respectively.Finally, the distribution of cell frequencies across the immunophenotypic populations (by SPADE analysis or manual gating) was used to perform a hierarchical clustering of all samples. This clustered patients into groups with different clinical risk. The most significant single distinguishing feature between clinical risk groups was the increased frequency (>40-fold) of HSPCs in HR-MDS compared to LR-MDS (p=9x10-7) or normal (p=6.3x10-6). Furthermore, this high-parameter analysis detected a >12-fold increase in the HSPC frequency in 2 patients with IPSS Int-2 disease with blast frequencies of <5% (following therapy).Conclusions: This first application of MCM for the analysis of MDS detected all major established aberrant expression patterns in MDS, as well as novel aberrant expression patterns of CD321, CD99, and CD47. Importantly, using high-parameter single-cell analysis and internal normal reference samples, we detected numerous deviations from the immunophenotypic boundaries of normal hematopoiesis in every analyzed MDS sample. Clustering of the cell frequency distribution across the immunophenotypic populations also defined groups of patients with differing clinical risk. These results demonstrate that high-parameter diagnosticcytometry methods can greatly enhance the diagnostic utility of immunophenotypic analysis in MDS. [Display omitted] DisclosuresBehbehani:Fluidigm: Consultancy. Finck:Fluidigm: Consultancy. Nolan:Fluidigm, Inc: Consultancy, Equity Ownership.

Unsupervised Discovery of Early Markers of Erythroid Maturation in Human Donor Marrow

Antigenic expression during erythroid maturation is not well understood. Despite the large number of surface antigens currently described for leukocytes, only a few (e.g. CD36, CD45, CD71, CD117, and CD235a) have been well characterized on erythroid cells. Here, we apply novel bioinformatic tools to select antigens with high potential for identifying successive stages of erythroid maturation in an unsupervised and unbiased manner from a high dimensional dataset. In brief, flow cytometry was performed on 3 normal donor bone marrows using Becton-Dickinson lyoplates containing a total of 275 unique antibodies and 8 gating reagents (CD15, CD19, CD34, CD38, CD45, CD71, CD117, and CD123). Each of the 275 independent data files from the same donor was aligned using the gating reagents and a weighted nearest neighbor algorithm in order to synthesize a flow data set with 283 antigens and 50,000 events. We applied a modified SPADE algorithm to generate a maturational path for cells of the erythroid lineage starting with CD34+/CD38- progenitors through the mature erythrocyte stage. A non-parametric Kruskal–Wallis test was used to identify and rank antigens that show differential expression along the erythroid maturational sequence.To discern antigens having the greatest discrimination for early vs. late erythroid maturation, we identified 15 antigens common to the top 25 most differentially expressed antigens from each donor sample. As expected, our method correctly identified the three erythroid gating reagents (CD45, CD71 and CD117) as well as previously a previously described erythroid associated antigen (CD36). A few of the identified antigens have been described are less common but also known to be differentially expressed on erythroid cells (e.g. integrin members CD29, CD44, and CD49d), however several additional novel antigens were also identified with strong differential expression including CD46, CD58, CD81, CD98, CD99, CD164, CD220 and CD321 (Figure 1). Of note, CD49d, CD98 and CD164 are of particular interest as they appear to be gradually lost during maturation from the early normoblast through the mature erythrocyte stages. This is in contrast with the other antigens that show high expression on pronormoblasts—with rapid decline during the early normoblast stage to the level of erythrocytes. We examined selected antigens from the synthetic data of all 3 samples using manual gating, and interestingly CD99 retained expression among the erythroid cells to late stage normoblasts in 2 of the 3 samples, suggesting some element of phenotypic variability among normal individuals for this antigen.Additionally, several antigens appear useful in distinguishing erythroid lineage cells from non-erythroid precursors including beta-2 microglobulin, CD50, and HLA-ABC which all showed ubiquitous expression in non-erythroid precursors with low to absent expression among cells of the erythroid lineage.In summary, using a novel unbiased method for large-scale antigen discovery, we have identified multiple novel antigens that are differentially down regulated with progressive erythroid maturation and appear useful in further delineating erythroid progenitor maturation. Additional work is underway to correlate these antigens with our morphologic understanding of erythroid maturation. Further work in characterizing these changes in myeloid stem cell disorders is on going and may be of diagnostic utility in the diagnosis of myelodysplastic syndromes. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Revealing the Immune Population Hierarchy Using Novel Interactive Visualization Tools, a Comparison of Methods: Sunburst, viSNE, and Spade on Mass Cytometry Data

The development of new technologies for high-parameter data has resulted in a critical bottleneck: identification of immune subsets is restricted to expert-based analysis, focusing on post-acquisition characterization of cell populations. Identification of cell subsets in flow cytometry has primarily focused on manual analysis, despite the fact that computational tools have proven useful for high-parameter and cross-sample comparisons. Sharing well-annotated data improves transparency and facilitates vital reproduction of results by external groups. Adoption of these new tools for immune subset discovery requires thorough collaborative investigation and validation of identified cell populations. To this end, in this study we compare the ease of discovery of immune subsets by comparing analysis through the use of three visualization tools: the sunburst hierarchy, the SPADE tree, and dimensionality reduction using viSNE. The sunburst hierarchy is a visual and interactive representation of traditional manual gating, whereas the SPADE tree is a semi-automated clustering and visualization tool for identification of cell subsets. viSNE allows interaction with high parameter data in the context of two-dimensional space where gating can be accomplished. In this study, we demonstrate the ability to automatically elucidate many immune subsets using Cytobank via an iterative analytic approach, combining computational tools (viSNE and SPADE) to recapitulate manually derived cell subsets. DisclosuresChen:Cytobank, Inc: Employment, Equity Ownership. Kotecha:Cytobank, Inc: Employment, Equity Ownership.

ISAC's classification results file format.

Identifying homogenous sets of cell populations in flow cytometry is an important process for sorting and selecting populations of interests for further data acquisition and analysis. Many computational methods are now available to automate this process, with several algorithms partitioning cells based on high-dimensional separation versus the traditional pairwise two-dimensional visualization approach of manual gating. ISAC's classification results file format was developed to exchange the results of both manual gating and algorithmic classification approaches in a standardized way based on per event based classifications, including the potential for soft classifications expressed as the probability of an event being a member of a class. © 2014 International Society for Advancement of Cytometry.

Unfold High-Dimensional Clouds for Exhaustive Gating of Flow Cytometry Data

Flow cytometry is able to measure the expressions of multiple proteins simultaneously at the single-cell level. A flow cytometry experiment on one biological sample provides measurements of several protein markers on or inside a large number of individual cells in that sample. Analysis of such data often aims to identify subpopulations of cells with distinct phenotypes. Currently, the most widely used analytical approach in the flow cytometry community is manual gating on a sequence of nested biaxial plots, which is highly subjective, labor intensive, and not exhaustive. To address those issues, a number of methods have been developed to automate the gating analysis by clustering algorithms. However, completely removing the subjectivity can be quite challenging. This paper describes an alternative approach. Instead of automating the analysis, we develop novel visualizations to facilitate manual gating. The proposed method views single-cell data of one biological sample as a high-dimensional point cloud of cells, derives the skeleton of the cloud, and unfolds the skeleton to generate 2D visualizations. We demonstrate the utility of the proposed visualization using real data, and provide quantitative comparison to visualizations generated from principal component analysis and multidimensional scaling.

FlowDensity: reproducing manual gating of flow cytometry data by automated density-based cell population identification.

flowDensity facilitates reproducible, high-throughput analysis of flow cytometry data by automating a predefined manual gating approach. The algorithm is based on a sequential bivariate gating approach that generates a set of predefined cell populations. It chooses the best cut-off for individual markers using characteristics of the density distribution. The Supplementary Material is linked to the online version of the manuscript. R source code freely available through BioConductor (http://master.bioconductor.org/packages/devel/bioc/html/flowDensity.html.). Data available from FlowRepository.org (dataset FR-FCM-ZZBW). rbrinkman@bccrc.ca Supplementary data are available at Bioinformatics online.