Taking each of LDL‐c, HDL‐c, and additionally triglycerides as the exposure variable in turn, we consider two analysis strategies. First, we take all genetic variants associated with the exposure at a genome‐wide level of significance (taken as P<10−8). Second, we restrict to genetic variants for which the P‐value for association with the target exposure (say, HDL‐c) is less than the P‐values for association with the nontarget exposures (say, LDL‐c and triglycerides). We perform each of the methods presented in the manuscript (IVW, simple median, weighted median, penalized weighted median, MR‐Egger regression). The data are presented as scatter plots in Figure 4 and Supporting Information Figure A1, and as funnel plots in Supporting Information Figure A2.
