Table 1.
Data source for the variants and pathogenicity interpretation
| Data name | Data source | Number of pathogenic variants | Number of neutral variants |
|---|---|---|---|
| Training dataset | ClinVar (≥ 1 stars) | 10,370 | 14,954 |
| VKGL (≥ 1 lab support) | 581 | 11,129 | |
| van der Velde et al. [29] | 30,187 | 274,112 | |
| Total * | 40,681 | 293,920 | |
| Benchmark dataset | ClinVar (≥ 2 stars) | 5421 | 20 |
| VKGL (≥ 2 lab support) | 187 | 11 | |
| ExAC | 0 | 5392 | |
| Total | 5421 | 5421 | |
| Benign Benchmark dataset 1 | Niroula et al. [24] | 0 | 60,699 |
| Benign Benchmark dataset 2 | GoNL | 0 | 14,426,914 |
*The total numbers of variants are smaller or equal to the sum of variants from all data sources due to the removal of duplicated variants