efficient FDR correction methods) and to (2) meta-analyze multiple QTL studies together. Second, the beta approximation behaves well enough with only 100 permutations to rapidly assess the impact on the analysis of important parameters such as cis-window size and covariates like the number of PEER factors (Stegle et al., 2012). And finally, FastQTL is fast (∼16× faster than Matrix eQTL for the same number of permutations) due to an efficient implementation of linear regressions, optimized C ++ code, efficient permutation schemes and rapid data retrieval from indexed files, while remaining user and cluster friendly. To summarize, FastQTL provides better adjusted P-values than the best method so far, Matrix eQTL with a direct permutation scheme, while being significantly faster. In addition, FastQTL also provides a modular base onto which new functionalities are being implemented, such as fine mapping of causal variants, conditional analysis to discover multiple independent QTLs per phenotype and interaction analysis to discover sex or disease specific QTLs.
