I am trying to determine exactly what method PLINK uses for meta-analysis. I can't find it in the program documentation/FAQ anywhere. Inverse variance? (given the meta command only requires effect size and SE)? ALso, what method its using to generate a P-value?
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
the Plink method used for meta-analyses is indirectly described here:
https://www.cog-genomics.org/plink/1.9/postproc
"weighted-z' requests weighted Z-score-based p-values (as computed by the Abecasis Lab's METAL software) in addition to PLINK's usual inverse variance-based analysis. This requires P and effective sample size fields (default column names are 'P' and 'NMISS'; these can be changed with --meta-analysis-p-field and --meta-analysis-ess-field). Note that, if the numbers of cases and controls are unequal, effective sample size should be 4 / (1/<# of cases> + 1/<# of controls>)."
