FIG. 1.

Diagram of computations performed. This figure summarizes the computations performed using as input the four genes from ColSBE (CHI3L1, LDHA, LGALS1, and IGFBP3) (top panel A) and the two genes selected from the PCA approach (MGMT and LDHA) (bottom panel B). Sets of genes are represented by a squared-box, datasets by a rounded-edges box, and computations described inside empty circles. The crossed empty circles indicate that only one of the following items are performed at the same time. The gray-colored text provides the figures and/or files containing the output information. “Standard” is an abbreviation for standardization, “Tr. mean-sd” stands for mean and standard deviation from the training set, “Quant” means quantiles normalization, “Norm” denotes normalization, “Cox res” corresponds to Cox residuals and “expr values” means expression values.