I am running Paradigm on a matrix of mRNA expression. I read
Vaske, Charles J., et al. "Inference of patient-specific pathway activities from multi-dimensional cancer genomics data using PARADIGM." Bioinformatics 26.12 (2010): i237-i245.
and also the help section in the web application. However it is not defined which format is best-suited for mRNA expression. Using log$_2$-ratios instead of log$_2$ data format gave me different results.