Table 5.
Mean and standard deviation (SD) of concordance rate and allelic R 2 calculated for different algorithms, panel densities and scenarios for both imputation to 50K and 777K SNP panels
| CR | Allelic R 2 | ||||
|---|---|---|---|---|---|
| No. | Mean | SD | Mean | SD | |
| Imputation to the 50K SNP panel | |||||
| Algorithm | |||||
| Beagle | 10 | 0.927 | 0.042 | 0.890 | 0.067 |
| Fimpute | 20 | 0.943 | 0.038 | 0.912 | 0.061 |
| Panel | |||||
| 3K | 6 | 0.864 | 0.011 | 0.787 | 0.016 |
| 6K | 6 | 0.946 | 0.008 | 0.919 | 0.011 |
| 8K | 6 | 0.952 | 0.008 | 0.927 | 0.011 |
| 15K | 6 | 0.973 | 0.006 | 0.962 | 0.008 |
| 20K | 6 | 0.953 | 0.008 | 0.929 | 0.011 |
| Scenario | |||||
| NE-P | 5 | 0.943 | 0.041 | 0.913 | 0.065 |
| NE-NP | 10 | 0.935 | 0.041 | 0.901 | 0.066 |
| NNE-P | 5 | 0.943 | 0.042 | 0.912 | 0.067 |
| NNE-NP | 10 | 0.935 | 0.042 | 0.901 | 0.066 |
| Imputation to the 777K SNP panel | |||||
| Algorithm | |||||
| Beagle | 8 | 0.895 | 0.040 | 0.826 | 0.066 |
| Fimpute | 16 | 0.921 | 0.035 | 0.866 | 0.059 |
| Panel | |||||
| 3K1 | 3 | 0.838 | 0.017 | 0.728 | 0.025 |
| 6K1 | 3 | 0.898 | 0.016 | 0.829 | 0.025 |
| 8K1 | 3 | 0.902 | 0.017 | 0.836 | 0.026 |
| 15K1 | 3 | 0.918 | 0.017 | 0.863 | 0.027 |
| 20K1 | 3 | 0.903 | 0.017 | 0.837 | 0.026 |
| 50K | 3 | 0.930 | 0.016 | 0.882 | 0.025 |
| 90iK | 3 | 0.952 | 0.010 | 0.919 | 0.016 |
| 90tK | 3 | 0.955 | 0.009 | 0.925 | 0.014 |
| Scenario | |||||
| NE-P | 8 | 0.9199 | 0.037 | 0.865 | 0.062 |
| NE-NP | 16 | 0.9082 | 0.039 | 0.846 | 0.065 |
| Step | |||||
| One-step | 15 | 0.8064 | 0.884 | 0.674 | 0.147 |
| Two-step | 15 | 0.8920 | 0.032 | 0.819 | 0.053 |
1Means and standard deviation for the two-step analysis.