Fig. 3

Gene-specific phenotypes (GSPs) estimated by gespeR are highly reproducible between different RNAi libraries across all pathogens. Mutual concordance is evaluated between phenotypes for the same genes. *Significantly better than second best method (Wilcoxon rank sum test, p < 0.05). a gespeR GSPs for four Qiagen genome-wide sub-libraries are significantly more reproducible than RSPs and estimates from haystack and RSA. b gespeR GSPs exhibit significantly stronger concordance than in silico pooled RSPs (ISPs) between single and pooled siRNA libraries from different vendors (Qiagen single-siRNA versus Dharmacon pooled)