Heritability estimate

[olivia...@gmail.com]

Hi,

I have just started using LD score regression to estimate chip heritability. I was just wondering how you would convert the heritability estimate from observed scale to a liability scale by hand. I know you use the sample and population prevalence, but I can't work out how to use these to make the conversion. I tried using the prevalence/observed heritability estimate to replicate the liability scale estimate from https://github.com/bulik/ldsc/wiki/Heritability-and-Genetic-Correlation but couldn't do it.

I am sorry if this is really simple question and/or the information is available elsewhere.

Kind regards,
Olivia

[Raymond Walters]

Hi Olivia,

The equation for this conversion (along with it’s derivation) appears in Lee et al, 2011, AJHG, specifically Equation 23. 

Plugging in observed scale h2 = .5907, K=.01, P=.5, and z=dnorm(qnorm(.01))=.02665 should give you 0.326 on the liability scale for SCZ as appears in the tutorial.

Cheers,

Raymond

--
You received this message because you are subscribed to the Google Groups "ldsc_users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ldsc_users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ldsc_users/c8131b5b-6880-4139-956f-1799222639a9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[olivia...@gmail.com]

Hi Raymond,

Thank you for the quick response! I understand now, thank you!

Kind regards,
Olivia