Monitoring Athletes' Physiological Responses to Endurance Training with Genomic-wide Expression Data

A system of fixed effect regression modeling for genome-wide expression data from DNA microarray hybridization is described that uses statistical methods in longitudinal or matched case-control data analysis to monitor athletes' physiological responses to endurance training. We first identify significantly differential expressed genes with endurance training-induced muscle contraction using fixed effect regression modeling, which effectively adapts the confounding effects arising from the interaction between genes. Next, we map key genes onto acknowledged KEGG pathways to attain a linkage between key molecules and biochemical pathways with endurance training-induced muscle contraction in a cause-effect format. To demonstrate this approach, we have used fixed effect logistic regression modeling to study a gene expression model relating to endurance training-induced vastus lateralis muscle contraction. We have found the development of carbohydrate, lipid and energy metabolisms, respectively, the transcriptional regulations of endurance training-induced vastus lateralis muscle contraction status, and the presence of the deleterious effects of oxygen from the metabolic reduction of the reactive oxygen species. The approach described here can supply general tools to monitor athletes' physiological responses to endurance training on the genomic scale.