To examine whether gene-expression profiling of circulating leukocytes is different between patients with sepsis as compared to controls. The signature genes identified will then be used to develop a prediction model to assist the diagnosis of sepsis. Keywords: Prediction study
Overall design
Differentially expressed genes are identified from a training set of 44 patients. The predictive accuracy of these genes is then assessed in an independent validation set of 50 patients.
Training set: GSM134895 - GSM135279 Validation set: GSM135655 - GSM135704