Joint Clustering and Matching of Multivariate Samples Across Objects (2012-2014)

We will develop a novel and widely applicable mixture model-based framework for the simultaneous clustering of multivariate samples observed on objects in a class and the matching of the object clusters. This statistical approach obviates the post-clustering need to match the object clusters since the matching is done during the fitting of the overall mixture model, which can be used as template for the class distribution. It thus provides a basis for discriminating between different classes in addition to the identification of anomalous events within a sample and a class. Key applications include image analysis and the automated analysis of data in flow cytometry which is one of the fundamental research tools for the life scientist.