I'm not going to repost any of the insane antisemitic conspiracy bullshit that grok is spewing today, but it highlights how absolutely essentially is that we not let LLMs become a form of epistemic grounding for our society.

Totally agree with this Kaur. I have also been playing with the data (thanks for sharing!). I found that nearly all univariate MR associations with CAD were highly diminished or disappeared in a multivariate MR including APOB. IMO this should be the baseline analysis for this kind of data.

Thanks for indulging my curiosity. Do you believe this report that PDGFD lof causes PAH? genomemedicine.biomedcentral.com/articles/10.... If so, this and KDR lof association gives decent indication of what aspects of imatinib might be beneficial.

Interesting work. Imatinib hits a lot of things. Could hitting just one of the pathways be safer/more effective? Anything from genetics point to VEGF or PDGF pathways?

I guess the reality is that if we want to analyse 100k+ samples things get tricky unless you use LLMs or something similar, like in regenie. For locus discovery I don't see it as a big issue, but when you're using these results to create PRS or in aggregate I agree there must be some sort of bias

I'm guessing it's the gene that encodes a nicotine acetylcholine receptor

Thanks, certainly agree with that, but the mechanism for raised T2D risk is thought to be impaired insulin sensitivity and compromised β cell function, rather than through elevated BMI. So it's unexpected to see that here, and may be due to non-specificity of the instruments. Either way, nice work!

Interesting paper, but when deriving instruments this way can you be sure the associations are acting exclusively through HMGCR? The relationship between HMGCR (and statins) and BMI is not clear cut so some of the effect could be through an alternative gene.

5 cats? Perhaps a picture of Leiden would have been more fitting!

F5. Too easy!

Not sure if "I disagree with this person so they deserve violence" is quite the strategy we need

The best data for the pathway is from GIPR (the other target in Lilly's zepbound), where rare deleterious missense variants lower BMI (pubmed.ncbi.nlm.nih.gov/29273807/). PCSK1 KO also causes obesity and is responsible for processing proglucagon into GLP1 (www.thelancet.com/journals/lan....

Totally agree and very easy to disprove. E.g Someone who is homozygous KO for PCSK9 will get very marginal (if any) LDL improvement (and therefore CAD risk reduction) from being homoygous KO for APOB. In individuals there are clear limits to additivity!

For history’s sake, record of a late stage capitalism life arc

Makes sense, thanks

Is it possible this is confounded by assortative mating?

Crikey shit

Once again, the 🧬 predicted the trial results. PCSK9 inhibition to prevent abdominal aortic aneurysms #CardioSky #GWAS