Threshold determination of strictly standardized mean difference in RNA interference high throughput screening assays

RNA interference (RNAi) is a powerful genetic analysis tool for the effficient silencing of target genes and RNAi high throughput screening (HTS) is becoming an important tool in drug discovery and development. The application of RNAi HTS relies on good analytic methods for quality control and hit selection in RNAi HTS experiments. Recently, strictly standardized mean difference (SSMD) has been proposed for measuring the magnitude of difference between two populations. For the use of SSMD in either QC or hit selection, we should determine the threshold of SSMD. To do so, we first present and prove a theorem that establishes the link between SSMD and d⁺ -probability. Based on the original meaning of SSMD and its probability interpretation, we then propose the cutoff criteria for quality control and hit selection respectively and discuss how to apply these thresholds in RNAi HTS experiments. The link between SSMD and d⁺-probability and the SSMD thresholds constructed in this paper provide a basis for the use of SSMD in both quality control and hit selection in RNAi HTS experiment.