Abstract
Flow cytometry is a standard platform for studying intracellular and extracellular protein expression of different cell populations in a tissue sample. Using specific antibody profiles, surface protein expression may be found as different for certain cell populations in samples that belong to different classes such as disease and normal or to cohorts with different genotypes. Analysis of such statistically significant differential expression can yield important biomarkers. Here we describe a computational tool DVisE to identify and localize precisely the cell subpopulations with statistically significant differential expression across different cohorts and classes. We analyzed HLA-DQ surface expression in Lymphoblastic cell lines using 266 out of 270 samples from the HapMap project. The cohorts were subdivided into 3 genotypic classes according to an allelic variant within upstream of the HLA-DQ gene. With the help of the present tool we were able to identify a significantly distinctive cytomic signature that is well preserved among genotypes in all the populations. Because of its novel ability to locate distinct areas where immune cells differentially express proteins, DVisE can play a very useful role in our study of the immune system. Indeed the tool could be extended to multiple different applications in bioinformatics and pattern recognition such as data visualization and discriminant analysis.