Power of SSMD test and validation of significant L- and S-SSMD gene sets.

<p>(<b>A</b>) The change of power as a function of sample size. (<b>B</b>) The change of power as a function of the size of a gene set. (<b>C</b>) Validation of significant L- and S-SSMD gene sets using different expression data. Original: Geuvadis LCL expression data normalized using PEER (i.e., data used for the main results); Rep1: first set of replication of Geuvadis LCL expression data without PEER normalization; Rep2: second set of replication of Geuvadis LCL expression data without PEER normalization; Whole blood: GTEx whole blood expression data; and Muscle: GTEx muscle expression data. The boxplot shows the frequency of observed SSMD is greater than the control SSMD of 1,000 random replicates.