Wade Rogers has been interested in pattern discovery for about 20 years. It began with 'TupleWare', an algorithm for discovering patterns in gene and protein sequence data, first at DuPont and later at Bristol-Myers Squibb. Later, he and colleagues expanded to include patterns in arbitrary categorical data. For the past 10 years he has focused almost exclusively on the challenges of analyzing flow cytometry data at the University of Pennsylvania.
Single cell analyses are becoming major players in life science markets. However, the increasing dimensionality, complexity and volume of data produced by flow cytometry pose a serious challenge that must be overcome in order to realize the full clinical potential of the method. High dimensional cell-based measurements, coupled with sophisticated computational analysis (collectively termed “cytomics”) may yield phenotypic or functional patterns that can provide informative biomarkers for discovery and clinical use. We envision an industrial-scale computational framework that will support the development and deployment of a data analysis pipeline that (a) is robust, hardened, and fully automated, eliminating analysis subjectivity and facilitating regulatory filing, (b), can be deployed in a centralized application-as-service business model for prospective, on-demand analysis of laboratory data and (c) enables retrospective datamining of cytomic data for analysis and discovery of new biomarkers.