**Acknowledgements**

the null hypothesis, which is the mean of a test sample being equal to the mean of the negative reference group, is accepted or not. Paired *t*-test (first pairing of the test sample and reference value within each plate, then calculating *t* statistic on the paired values) is often preferred to avoid the distortion of results due to inter-plate variability, whereas unpaired *t*-test is used for global comparison of the sample repli‐ cates with all reference values in the experiment (Zhang 2011a). The *p* value calculat‐ ed from *t* statistic is then used to determine the significance of the sample activity compared to the reference. An alternative to standard *t*-test, namely randomized var‐ iance model (RVM) *t*-test (Wright and Simon 2003), was found to be more advanta‐ geous for screens with few replicates to detect relatively less strong "hits" (Malo et

**•** SSMD: While *t*-test is a useful method to calculate the significance of the sample activ‐ ity by incorporating its variability across replicates, it lacks the ability to rank the samples by their effect sizes. As an alternative to *t*-test, SSMD-based "hit" selection method for replicates was proposed to enable the calculation of RNAi effects as previ‐ ously illustrated in (Zhang 2011a). While SSMD-based method is more robust with small sample sizes as opposed to *t*-test (Zhang 2008a), at least 4 replicates is recom‐ mended in confirmatory screens to identify samples with moderate or higher effects

**•** Various other *p* value calculation methods (e.g., redundant siRNA activity, or RSA) (Ko‐ nig et al. 2007) and rank products method (Breitling et al. 2004)) are available, which can

HT screening is a comprehensive process to discover new drug targets using siRNA and drug candidates from small molecule libraries. Statistical evaluation of the assay perform‐ ance is a very critical step in HT screening data analysis. A number of data analysis methods have been developed to correct for plate-to-plate assay variability and systematic errors, and assess assay quality. Statistical analysis is also pivotal in the "hit" selection process from pri‐ mary screens and in the evaluation during confirmatory screens. While some of these meth‐ ods may be intuitively applied using spreadsheet programs (e.g., Microsoft Excel), others may require the development of computer programs using more advanced programming environments (e.g., R, Perl, C++, Java, MATLAB). Besides commercially available compre‐ hensive analysis tools, there are also numerous open-access software packages designed for HT screening data management and analysis for scientist with little or no programming knowledge. A short compilation of freely available analysis tools is listed in Table 3. The growing number of statistical methods will accelerate the discovery of drug candidates with

be adapted to detect "hits" in RNAi screens (Birmingham et al. 2009).

al. 2010).

218 Drug Discovery

(Zhang and Heyse 2009).

**6. Conclusion**

higher confidence.

This work was supported by the American Lebanese Syrian Associated Charities (ALSAC), St. Jude Children's Research Hospital, and National Cancer Institute grant P30CA027165.
