Brian Boutwell, David Hinds, the 23andMe Research Team, Jorim Tielbeek, Ken Ong, Felix Day, John Perry
Review posted on 22nd February 2017
It's not clear to me that MSRA replicated the finding since there wasn't an irritability scale and the personality characteristics that were associated with the SNP were quite variable in nature. "For MSRA, there was no single personality trait that matched self-reported irritability in the 23andMe dataset; however, the allele previously associated with increased irritability was significantly associated (Figure 3) with reduced energy and enthusiasm (“activity”, P=7.2x10-4), less compliance (P=4.8x10-4), more depressive feelings (P=6x10-7), more neuroticism (P=1.1x10-7), less creative thinking (“ideas”, P=0.02), more introversion (P=8x10-3), more anxiousness (P=6.9x10-5), less openness to new ideas (P=9.8x10-3) and more adverse feelings to taking risks (P=0.01)." Most of these characteristics seem the opposite of increased irritability (reduced energy). I'm not a statistical geneticist - what do you consider a significant p value in this analysis given the wide range of p value scores provided? You say how they were computed but not what you consider significant.