How does TargetDiscovery take into account multiple testing?
TargetDiscovery applies the Benjamini-Hochberg method for controlling the False Discovery Rate (FDR). Under mild assumptions and validity of the p-values, this will guarantee that the expected proportion of false positives among all target variables that have an FDR-adjusted p-value smaller than some significance level alpha is smaller than alpha. In TargetDiscovery we declare a reported effect as statistically significant if the FDR-adjusted p-value is smaller than or equal to alpha=0.05.
