PROBABILITY STATE MODELING: A NEW PARADIGM FOR CYTOMETRIC ANALYSIS - Flow Cytometry in Drug Discovery and Development - page 271

14

PROBABILITY STATE MODELING: A
NEW PARADIGM FOR CYTOMETRIC
ANALYSIS

C. B RUCE B AGWELL

14.1 INTRODUCTION

Cytometry is likely to be the technology of choice for observing and understanding
cellular processes in the foreseeable future because of its ability to relate complex sets
of measurements to a single cell. With the advent of new cytometers capable of
measuring numerous fluorescence wavelengths and the ever-increasing number of
antibodies and fluorochromes, it is becoming increasingly difficult to deal with the
dimensionality of cytometry data. Traditional analysis techniques that are based on
examining one measured parameter versus another run headlong into what is
commonly referred to as the dimensionality barrier. In some cases, hundreds of
histograms are necessary to assess the status of a single specimen.
This chapter describes a new paradigm called probability state modeling (PSM)
that is capable of visualizing and analyzing the rich multiparameter information
embedded in cytometric listmode files. Model-derived parametric plots merge
intensity, percentage, and variance information into easy-to-understand graphical
formats that represent virtually every bit of information contained in these files. These
models incorporate expert metainformation that helps elucidate the hidden meanings
in the data even to those not trained in cytometry. Model-based analyses are likely to
be the method of choice in bringing automation to cytometry.