Title: "Analyzing High Dimensional Flow Cytometry Data: Combinatorial and Statistical Algorithms"
Speaker: Ariful Azad, Department of Computer Science, Purdue University

Place: Mechanical Engineering (ME) 1130
Date: March 19, 2013; Tuesday
Time: 4:30pm

Abstract:
Flow cytometry is a nearly 50-year old technology for studying properties of single cells via scattering and fluorescence induced by lasers with applications in immunology and diagnosis of diseases. In recent years, flow cytometry has become multispectral (thirty or more signals can be detected simultaneously), and high throughput (millions of cells can be analyzed per minute at the single cell level). However, for analyzing the high dimensional, large-scale data generated by the new experimental methodologies, new algorithms from computer science, mathematics, and statistics are needed. We will describe a few of the computational problems in this context.

We describe FlowMatch, an algorithm for registering different cell types from patient samples using matchings in graphs and hierarchical template construction algorithms from multiple sequence alignment. Variance stabilization is also a critical issue in computing homogeneous cell populations. These cell types are then followed across multiple time points and experimental conditions. We report results from flow cytometry data generated from leukemia, Multiple Sclerosis, and phosphorylation shifts in T-cells.



Associated Reading:
Azad,A., Langguth, J., Fang,Y., Qi, A. and Pothen, A. 2010. Identifying Rare Cell Populations in Comparative Flow Cytometry, Workshop on Algorithms in Bioinformatics (WABI) 2010, Lecture Notes in Bioinformatics, 6293, pp. 162-175.

Azad,A., Pyne,S. and Pothen, A. 2012. Matching phosphorylation response patterns of antigen-receptor-stimulated T cells via flow cytometry, BMC Bioinformatics, 13(Suppl 2): S10.



Click here www.stat.purdue.edu/~doerge/BIOINFORM.D/FALL12/sem.html for a full schedule of BIOINFORMATICS SEMINARS, past and present.