The authors respond to N. W. Bailey and T. Ly's commentary that draws attention to data mining as a specific approach to proteomic data analysis. They say that proteomic signatures can be acquired via an independent choice of differentially expressed proteins based on statistical criteria or can be extracted from complicated proteomic sets of data using machine learning algorithms.