The success of GWAS meta-analyses has shown that there is considerable benefit in being able to derive association tests on the basis of summary statistics. With these advantages in mind, He et al. [7] developed a statistical test to match the pattern of gene expression with a GWAS dataset. This approach, coded in the software Sherlock, can accommodate p-values as input. However, their hypothesis of interest differs from the question of colocalisation, with the focus of the method being on genome-wide convergence of signals, assuming an abundance of trans eQTLs. In particular, SNPs that are not associated with gene expression do not contribute to the test statistic. Such variants can provide strong evidence against colocalisation if they are strongly associated with the GWAS outcome.
