Mendelian Randomization: How it Works, and What it Reveals about Vitamin D

[luisabonet]

Fantastic explanation! Thank you so much, I finaly understood it.

[qigangdeng]

very well explained. Very clear! Thanks for your sharing.

[sandeepb]

Super explained so nice. Very nicely lecture delivery. You rock.

[belevitt]

It would be awesome if you would link the article you are talking about. I'd love to see this one in particular.

[abritrn]

Can Mendelian randomization account for epigenetics? A person may have a gene for low vitamin D levels but is that gene being expressed?

[cynanomite]

Can the assumption that genes don't influence behavior (e.g., to compensate for some genetically-derived endocrine deficiency) be simply accepted in all cases? If not, the lack of outcome variation across genotypes may not tell the whole story..

[yashir]

damnn, you just explain it so well. thank you.

[diabeteslecture]

2:28 thoroughly confusing. can you make it clear what you want to say pl

[methodofinstruction]

How do you know the genes being observed are the only ones implicated in the outcome variable? If you can’t answer this question, Mendelian randomization is not enough to prove causation…

[johnhogue]

Let me make some corrections. Mendelian Randomization DOES NOT establish cause and effect. It is still correlative work. Yes, the gene may exist for a person’s entire life, but other cofounders still exist. Genetic expression varies depending on environment. The use of Mendelian Randomization to establish causality is as ridiculous as stating that the release on Nicolas Cage movies causes drownings.
Btw, the efforts to try to establish causality in human observational studies are not only “not perfect”, but have literally no capacity to establish cause and effect. Correlation is not causation. It doesn’t even imply causation. No amount of mathematical wizardry will change that.