TABLE 4.
Observed 95th percentile likelihood ratios under the hypothesis of no QTL segregation, test statistic, and power to detect QTL
| Nonoccurrence of selection
|
Selection
|
|||||
|---|---|---|---|---|---|---|
| Formula | Threshold | Test statistic | Power (%) | Threshold | Test statistic | Power (%) |
| Setting 1: 300 individuals, 50 parents | ||||||
| Formula 1 | 4.04 | 5.66 (3.99) | 60 | 4.12 | 5.44 (3.93) | 45 |
| Formula 2a | 3.97 | 6.08 (4.57) | 63 | 3.88 | 5.78 (4.59) | 47 |
| Formula 2b | 4.06 | 8.16 (5.62) | 78 | 3.91 | 8.88 (5.86) | 85 |
| Setting 2: 500 individuals, 100 parents | ||||||
 | ||||||
| Formula 1 | 4.08 | 4.62 (3.52) | 47 | 3.86 | 4.41 (3.94) | 44 |
| Formula 2b | 3.96 | 5.23 (3.97) | 58 | 3.44 | 7.35 (4.63) | 70 |
 | ||||||
| Formula 1 | 4.08 | 7.75 (5.06) | 71 | 3.66 | 6.35 (4.14) | 67 |
| Formula 2b | 3.96 | 9.76 (6.71) | 80 | 3.44 | 11.65 (6.22) | 94 |
 | ||||||
| Formula 1 | 4.08 | 22.03 (10.9) | 100 | — | — | — |
| Formula 2b | 3.96 | 23.69 (10.8) | 100 | — | — | — |
See Table 1 for a description of settings 1 and 2. Threshold represents the empirical chromosomewise threshold calculated for 1000 replicates. Test statistic is the mean and standard deviation of the maximum of the LR test for the 100 replicates. Power is the percentage of replicates with maximum LR exceeding the empirical threshold. —, simulations are not performed under these conditions.