Computational modeling of the human serum proteome response to colon resection surgery

Principal component analysis (PCA) is much used in exploring time-course biological data sets, but does not distinguish variation between time and subjects. This study proposes a new integrated approach by combining analysis of variance (ANOVA) and three component modeling methods. The former was used to separate the between- and within-subject variation, and the latter represent modeling strategies on a scale moving from commonality to individuality. The proposed approach was applied to a surface-enhanced laser desorption and ionization time of flight mass spectrometry (SELDI-TOF-MS) data set of a serum protein expression time course before and after colon resection. Two common biological processes are identified and individual differences among patients were also detected, and the biological relevance of both is discussed.