N for each of cases and controls or a total N?

[Apprentice]

Hi,

I am working with a case-control study data, which has a high ratio of controls to cases (e.g. 5:1). 

munge_sumstats.py has --N or --N-cas and --N-con options to specify sample sizes in a case-control study.

For a tutorial of LDSC, --N option, which is calculated as N of cases + N of controls, was used as below:

https://github.com/bulik/ldsc/wiki/Heritability-and-Genetic-Correlation

I think that --N-cas and --N-con, which mean N of cases and N of controls, should be used in a case-control study.

Which option should I use to estimate h2 and rgc values for a case-control study?