Clustering of protein expression data: a benchmark of statistical and neural approaches

摘要: Clustering issues are fundamental to exploratory analysis of bioinformatics data. This process may follow algorithms that reproducible but make assumptions about, for instance, the ability estimate global structure by successful local agglomeration or alternatively, they use pattern recognition methods sensitive initial conditions. paper reviews two clustering methodologies and highlights differences result from changes in data representation, applied a protein expression set breast cancer (n = 1,076). The approach model-free probabilistic competitive neural network. results compared with existing studies same set, preferred solutions profiled clinical interpretation.