Table 4.
The Optimal Meta-analysis Strategy to Combine Summary Statistics for Imputed and Observed Phenotype Achieves Maximum Power
| rsID | Imputation-Based Fixed-effect Meta-analysis Power | General Fixed-Effect Meta-analysis Power |
|---|---|---|
| rs673548 | 83.56% | 82.30% |
| rs10096633 | 62.14% | 45% |
| rs3923037 | 63.65% | 60.86% |
| rs6728178 | 80.96% | 80.00% |
| rs6754295 | 75.49% | 74.31% |
| rs676210 | 82.01% | 80.85% |
Imputation-based fixed-effect meta-analysis uses the optimal weights that are shown in Equation 17. General fixed-effect meta-analysis uses the typical fixed-effect meta-analysis weights where the weight for each study is the square root of the number of samples in the study.