Re: "credible mendelian randomization studies: approaches for evaluating the instrumental variable assumptions".