Some question about n

[123569]

In the two sample files 1_oa2016_hm3samp_neur.txt and 1_oa2016_hm3samp_swb.txt you gave me, I don't know exactly how to get the N value, and I still don't fully understand it after reading relevant literature. Can you give me an example of where the N values in these two files come from? This is the N value is the population sample size or the SNP sample size?

[huilisabrina]

Hi @123569 ,

Please find the following quote from the wiki page here:

We demonstrate the use of mtag with GWAS results on neuroticism and subjective well-being from Okbay et. al. (2016). We use a subset of the original GWAS results from a random subsample of SNPs found in Hapmap3. The neuroticism and subjective well-being data summary statistics can be found here and here.

Hope this answers your question!

Best, Hui

[som-soremekun]

I'm stuck too @huilisabrina. So can we say that N is the total sample size?

[paturley]

Hi,

I'm a bit confused by your question. In MTAG, N should correspond to the effective sample size. For quantitative traits analyzed with ordinary least squares, it would be total sample size.

On Wed, Jun 3, 2020 at 8:47 PM Somyjay [email protected] wrote:

I'm stuck too @huilisabrina https://github.com/huilisabrina. So can we say that N is the total sample size?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/omeed-maghzian/mtag/issues/60#issuecomment-638534273, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5MFAFUACXCMHSAWLYDRU3VJTANCNFSM4GM2AI3A .

[FEIFEI502]

Hi,

Can I ask if it is case-control study using logistic regression, how to calculate the effective sample size? Is that simply to sum the number of case and control?

Thanks in advance, Feifei

[paturley]

I believe the standard formula is 4N_casesN_controls/(N_cases + N_controls). This paper might be useful to you. https://www.nature.com/articles/ejhg2016150/

On Wed, Oct 7, 2020 at 8:39 PM FEIFEI502 [email protected] wrote:

Hi,

Can I ask if it is case-control study using logistic regression, how to calculate the effective sample size? Is that simply to sum the number of case and control?

Thanks in advance, Feifei

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/60#issuecomment-705265261, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5JM4KRS4TDHYH2VF4LSJUC3XANCNFSM4GM2AI3A .

[FEIFEI502]

Thanks for your answer, that is really helpful.