Seminar at OUS, Radiumhospitalet, 1. September 2016. 13:00-15:15 CyTOF opening seminar featuring Jonathan Irish

¹ Norsk Hydro's Fund Lecturer

Abstracts

Human Systems Immunology Enabled by Mass Cytometry
Axel Olin

By simultaneously analyzing the many components of human immune systems, a greater understanding of immune system regulation in health and disease is within reach. Using Mass cytometry the amount of information that can be retrieved from a limited blood sample has increased substantially. By combining Mass cytometry with novel computational tools for data visualization and analysis, we are investigating human immune system variation among healthy individuals and estimating the role of environmental factors in shaping an individual's immune system, particularly early in life.

Follicular lymphoma analyzed by mass cytometry
Cara E. Wogsland

Follicular lymphoma (FL) is a malignancy of B cells of the immune system. Mass cytometry combined with similarity mapping and unsupervised clustering (viSNE and SPADE) was used to immunophenotype malignant B cells, non-malignant B cells, and T cells in FL tumors. This revealed that diversity in cellular phenotype between tumors was greater for the malignant B cells than for non-malignant B cells and T cells. Furthermore, FL B cells bore little phenotypic similarity to germinal center B cells which is thought to be the healthy counterpart of FL B cells. A suppression of the germinal center reaction in the tumoral non-malignant B cells was observed and may help explain why FL malignant B cells can persist as tumors for so long without generating tumor-clearing immune responses.

 
Learning and communicating cell identity using quantitative single cell measurements
Jonathan M. Irish

While advances in single cell biology have enabled measurement of 35 or more features on millions of single cells, learning cell identity still relies on human experts ¹.  Quantitative cytometry workflows have developed diverse approaches to grouping cells into populations and visualizing results in graphs that arrange populations based on phenotype.  These strategies work well when feature variance is low and cells match established types, but computational analysis of single cell data is now routinely revealing novel cells with non-canonical phenotypes ².  This is especially the case for diseases where abnormal expression profiles and signaling responses distinguish clinically significant cell subsets ³.  A quantitative system for objectively labeling and communicating cell type is now needed.  Marker Enrichment Modeling (MEM) addresses this need to quantitatively describe cell subsets by quantifying feature enrichment and reporting a text label that can be used by machines and humans to learn cell type (cytotype).  For a given subset of cells within a sample, MEM compares both feature magnitude and feature variance for the subset with those of a reference cell population, such as stem cells or other cells in the samples.  In datasets including healthy human blood, bone marrow, and tonsil, murine tissues, and leukemia patient blood, MEM labels reported the key proteins used by experts to distinguish rare and novel cell subsets.  The MEM approach also generated quantitative labels for cohorts of patients characterized by gene expression profiling and by phospho-specific flow cytometry.  Quantitative MEM labels can be used to learn cytotype by comparing newly observed cell populations to existing reference data.  Based solely on comparison of computationally generated MEM text labels, an algorithm correctly identified 59 cell populations from 7 separate studies of 3 human tissues created using different measurement technology, mass spectrometry and fluorescence cytometry.  MEM label reporting thus provides a new way to objectively and quantitatively communicate cell type characteristics in studies of complex, dynamic tissue microenvironments, such as stem cell niches, immune responses, and human tumors.

1               Diggins, K. E., Ferrell, P. B. & Irish, J. M. Methods for discovery and characterization of cell subsets in high dimensional mass cytometry data. Methods, doi:10.1016/j.ymeth.2015.05.008 (2015).

2               Irish, J. M. Beyond the age of cellular discovery. Nature immunology 15, 1095-1097, doi:10.1038/ni.3034 (2014).

3               Greenplate, A. R., Johnson, D. B., Ferrell, P. B. & Irish, J. M. Systems immune monitoring in cancer therapy. European journal of cancer 61, 77-84, doi:10.1016/j.ejca.2016.03.085 (2016).