TABLE 8.
Estimating ρ in the presence of biased amplification
| X | NT | pe = 0.01 | pe = 0.05 | pe = 0.1 | pe = 0.15 |
|---|---|---|---|---|---|
| 16× | 3 | 21.7 (7.8) | 21.9 (7.8) | 21.8 (7.6) | 21.7 (7.5) |
| 5 | 21.9 (7.6) | 21.9 (7.7) | 21.7 (7.5) | 21.6 (7.5) | |
| 8× | 3 | 21.8 (7.6) | 22.4 (8.2) | 20.9 (7.6) | 21.2 (7.3) |
| 5 | 21.7 (7.7) | 22.1 (7.6) | 22.1 (7.8) | 21.1 (7.2) | |
| 4× | 3 | 21.5 (7.7) | 21.6 (7.5) | 22.0 (7.8) | 21.1 (7.7) |
| 5 | 21.0 (7.4) | 20.5 (7.1) | 21.0 (7.5) | 20.5 (7.2) | |
| 2× | 3 | 19.5 (6.7) | 20.1 (6.9) | 19.6 (6.6) | 19.5 (6.5) |
| 5 | 13.4 (7.8) | 12.7 (8.22) | 12.5 (8.28) | 12.1 (8.8) |
Here we set the generating values, θ0 = 100 and ρ0 = 20. NT is the number of reads required to use homozygous genotype calls in the inference and pe is the probability that a minor allele is not sequenced. We report mean (and RMSE) of the estimates over 1000 simulations.