Online MetaGAP calculator


Description


MetaGAP is a versatile tool for calculating the statistical power of a meta-analysis of GWAS results and of the polygenic-score R² in a hold-out sample. The tool allows the user to specify the genetic correlation between the studies in a meta-analysis as well as the genetic correlations between the hold-out sample and the meta-analysis studies.

This calculator, accounting for imperfect genetic correlations between studies, provides (i) the power of a meta-analysis of GWAS results from different studies, and (ii) the expected R² of polygenic-score for a hold-out sample with SNP-weights based on the meta-analysis SNP-effect estimates.

Details on the underlying model, assumptions, and derivations, can be found in the De Vlaming et al. (2017).

Notes: (i) h² denotes SNP-based heritability and N sample size, (ii) lists for h² and N need to be comma-separated, (iii) replacing a list by a single number enforces the same value across all studies in the meta-analysis.


Calculator


Please enter the following quantities and click on the SUBMIT button:

Number of studies in the meta-analysis (maximum = 250):

 

C =

N per study in meta-analysis:

 

N={}

h² per study in meta-analysis:

 

h²={}

h² in hold-out sample:

 

h²out =

Genetic correlation between studies in meta-analysis:

 

ρG =

Genetic correlation between hold-out study and meta-analysis studies:

 

ρG,out =

Effective number of independent SNPs:

 

S =

Effective number of independent causal SNPs:

 

M =

 

 

Offline MetaGAP calculator


An offline MetaGAP calculator for MATLAB can be found here. This calculator allows users to specify any positive definite cross-study genetic correlation matrix.