This Site umd.edu

Graduate Student Seminar - 10/12/2012

"Cluster Analysis of Protein Pattern Sets Using Minkowski Functionals"

by Josh Parker

Friday, October 12, 2012 -- 12:00 p.m.
Large Conference Room, 1207 Energy Research Facility

Advisor:  Associate Professor Wolfgang Losert

Point patterns arise in many different areas of physical and applied research, often resulting in sets of patterns that may or may not be fundamentally different. We introduce a numerical taxonomy procedure for clustering point pattern sets using their approximated Minkowski functionals. We demonstrate that this procedure is robust in distinguishing different spatial processes, even when the number of points in the patterns are small, vary wildly from pattern to pattern, or when the patterns are drawn from very similar processes. We then place this routine in a quantitative biology context by analyzing two point pattern sets of fluorescently labeled intercellular proteins, LAT and TAC, that have been acquired from experiments with immune cells. We demonstrate that our technique provides a biologically meaningful result; namely, that there is a remarkable amount of cell-to-cell consistency in the patterns of TAC proteins. Overall, we find that this routine is a robust method for distinguishing point pattern sets and provides meaningful insight regarding the homogeneity of a spatial process.

For additional information about the IREAP Graduate Student Seminars, contact David Meichle.

Return to the Graduate Student Seminars schedule.