Top ranked terms in SLEPR analysis results for a well-studied prostate cancer dataset.

A well studied prostate cancer dataset [42], [43] was analyzed with SLEPR method. SLEPR method used the 25 tumor samples as Inclusion/Target class compared to 9 nonmalignant tissues as Exclusion/Background class for GSEA annotated terms, by two-sided MADe method for selection of both highly and lowly expressed genes as sample-level differentiated genes. Only top ranked functional terms were shown in the table (the chromosomal location-based annotation terms were taken out for simplicity). 1000 permutations were performed.