Skip to main content

Advertisement

Figure 1 | BMC Bioinformatics

Figure 1

From: Lack of sufficiently strong informative features limits the potential of gene expression analysis as predictive tool for many clinical classification problems

Figure 1

Spiked probe set recovery rates as function of informative samples size and signature strength. Both, the number of spiked probes (genes) and the number of features included in the predictor model were set to 10. The solid lines indicate the average recovery rates and the dots represent results from the 20 individual iterations. Results from the MAQC-II data set (n = 233) are shown. The "c" value which is the log2 fold-change takes on the values 0.5, 1.0, 1.2 and 1.5.

Back to article page