We therefore developed a Connectivity Score (Figure 2D) that provides three measures of confidence: 1) a nominal p-value derived by comparing the similarity between the query and reference signature, using the Kolmogorov-Smirnov enrichment statistic (Subramanian et al., 2005), to a null distribution of random queries; 2) a false discovery rate (FDR) that adjusts the p-value to account for multiple hypothesis testing; and 3) Tau (τ), which compares an observed enrichment score to all others in the database (see STAR Methods).
