Replies to post #142753 on Biotech Values

[genisi]

He pointed out that the genes they looked at in this study were all selected based also on the fact that they had no impact on LDL levels. Given the synthesis of LDL and HDL are related, this choice might be significant and might be a possible explanation for this result.

One requirement in a mendelian randomisation analysis is that the genetic instrument (SNP) must associate with the outcome (MI) only through effects on the intermediate phenotype (HDL cholesterol). So the selected SNPs were exclusively associated with plasma HDL-C and not having pleiotropic effects on purpose - because the hypothesis tested via a mendelian randomisation analysis was that the association of a plasma HDL-C with MI is causal. I find the study data very compelling, still given that no study is perfect and HDL is complex, it is possible and makes sense that subtypes of HDL molecules and/or other interactions/factors (not plasma level) play a role and are yet to be determined.