Automated analysis of multi-dimensional B-cell flow cytometry data (44.23)

Abstract

Abstract Understanding B cell diversity and its functional correlates requires the combined study of large numbers of surface and intracellular molecules. Advances in multi-parameter flow cytometry (FCM) allow for the independent detection of many fluorochromes on individual cells. However, traditional gating strategies based on single or two-color displays are inadequate for separating cell populations defined in multi-dimensional space. We have developed a novel program, FLOCK, which uses density-based clustering to algorithmically identify cell subsets in multi-dimensional space in an unbiased fashion, reducing operator variability. FLOCK has been used to accurately identify and quantify 17 distinct B cell subsets in human peripheral blood, including novel plasmablast subsets after tetanus vaccination. It is being used to monitor B cell subsets in different autoimmune conditions including systemic lupus erythematosus patients treated with BAFF inhibitors. FLOCK has been assessed by the FlowCAP consortium (Critical Assessment of Population Identification Methods), and achieved encouraging results in a recent FlowCAP competition compared with other automated methods. FLOCK has been implemented in the publically available Immunology Database and Analysis Portal - ImmPort (www.immport.org) for open use by the immunology research community. Core modules of FLOCK will also be integrated into GenePattern of the Broad Institute as a component of the FCM analysis suite.