Hypothesis free MR
MR phenome-wide association studies (MR-pheWAS) are a hypothesis-free approach that use MR to search for the causal effects of a specified trait. This approach tests the association of a genetic score denoting the trait of interest, with a potentially large set of phenotypes, to identify associations for detailed follow-up. MR-pheWAS are non-trivial because epidemiological cohorts usually contain a diverse range of phenotypes such that they require highly heterogeneous models. For this reason, the first MR-pheWAS restricted the phenome scan to continuous phenotypes only, such that it was possible to automate the tests of association across these phenotypes in a consistent way .
Recently, an open source software package called PHESANT (PHEnome Scan ANalysis Tool) has been developed to perform phenome scans, including MR-pheWAS, in UK Biobank . PHESANT tests the association of a trait of interest with all continuous, integer and categorical variables in UK Biobank, and uses a novel rule-based algorithm to determine how to appropriately test each trait. UK Biobank contains a diverse range of phenotypes and a large sample size, providing an opportunity to identify novel, potentially causal associations with MR-pheWAS. A preliminary MR-pheWAS that searched for the causal effects of BMI using PHESANT, performed over 12,000 tests of association . This analysis identified 86 phenotypes that were associated with the BMI genetic score at a stringent Bonferroni corrected P value threshold.
 Millard LAC, Davies NM, Timpson NJ, Tilling K, Flach PA, Davey Smith G. MR- PheWAS: hypothesis prioritization among potential causal effects of body mass index on many outcomes, using Mendelian randomization. Sci Rep. Nature Publishing Group; 2015 Jan 16;5:16645.
 Millard LAC, Davies NM, Gaunt TR, Davey Smith G, Tilling K. PHESANT: a tool for performing automated phenome scans in UK Biobank. bioRxiv (2017)