Table 4.
Power of the population-based association approach (GEE test) for a SNP with MAF = 0.1
| Nominal Type I Error | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| SNP QTL Heritability | Effect Size (SD)* | Model | 0.05 | 0.01 | 0.001 | 10-4 | 10-5 | 10-6 | 10-7 | 10-8 |
| 100% phenotype available (n = 1345) | ||||||||||
| 1% | 0.24 | GEE | 0.977 | 0.919 | 0.78 | 0.60 | 0.43 | 0.27 | 0.17 | 0.10 |
| 2% | 0.33 | GEE | 1 | 1.00 | 0.99 | 0.97 | 0.91 | 0.84 | 0.72 | 0.59 |
| 3% | 0.41 | GEE | 1 | 1 | 1 | 0.998 | 0.992 | 0.985 | 0.957 | 0.918 |
| 4% | 0.47 | GEE | 1 | 1 | 1 | 1 | 1 | 1 | 0.996 | 0.991 |
| 5% | 0.53 | GEE | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| ~80% subjects have phenotype(~20% missing at random): Sample size = 1076 | ||||||||||
| 1% | 0.24 | GEE | 0.934 | 0.836 | 0.631 | 0.427 | 0.261 | 0.149 | 0.074 | 0.041 |
| 2% | 0.33 | GEE | 1.00 | 0.99 | 0.97 | 0.93 | 0.82 | 0.66 | 0.51 | 0.37 |
| 3% | 0.41 | GEE | 1 | 1 | 0.998 | 0.993 | 0.976 | 0.938 | 0.85 | 0.753 |
| 4% | 0.47 | GEE | 0.998 | 0.998 | 0.998 | 0.998 | 0.991 | 0.984 | 0.971 | 0.949 |
| 5% | 0.53 | GEE | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0.99 |
| ~60% subjects have phenotype(~40% missing at random): Sample size = 783 | ||||||||||
| 1% | 0.24 | GEE | 0.879 | 0.737 | 0.48 | 0.281 | 0.137 | 0.055 | 0.023 | 0.008 |
| 2% | 0.33 | GEE | 0.992 | 0.96 | 0.88 | 0.744 | 0.591 | 0.433 | 0.293 | 0.186 |
| 3% | 0.41 | GEE | 1 | 1 | 0.982 | 0.945 | 0.874 | 0.754 | 0.602 | 0.467 |
| 4% | 0.47 | GEE | 1 | 1 | 1 | 0.998 | 0.981 | 0.95 | 0.884 | 0.797 |
| 5% | 0.53 | GEE | 1 | 1 | 0.999 | 0.997 | 0.996 | 0.982 | 0.964 | 0.931 |
*SD: Standard Deviation. The effect size is expressed in unit of standard deviation of the phenotype.