Table 2.
Positive predictive value and sensitivity in case of GATK, Platypus, VarScan, LoFreq, FreeBayes, SNVer, SAMtools, VarDict, the combined output of all tools (eight tools), single-appreci8 and appreci8 in training sets 1 and 2
| Approach | Training set 1 |
Training set 2 |
||
|---|---|---|---|---|
| Sens | PPV | Sens | PPV | |
| GATK | 0.92 | 0.85 | 0.82 | 0.71 |
| Platypus | 0.93 | 0.80 | 0.83 | 0.42 |
| VarScan | 0.89 | 0.97 | 0.47 | 0.73 |
| LoFreq | 0.91 | 0.35 | 0.78 | 0.23 |
| FreeBayes | 1.00 | 0.03 | 0.99 | 0.02 |
| SNVer | 0.93 | 0.92 | 0.55 | 0.07 |
| SAMtools | 0.85 | 0.87 | 0.64 | 0.77 |
| VarDict | 0.97 | 0.96 | 0.94 | 0.15 |
| 8 tools | 1.00 | 0.03 | 1.00 | 0.02 |
| single-appreci8 | 0.98 | 0.98 | 0.99 | 0.35 |
| appreci8 | 0.98 | 0.99 | 0.98 | 0.94 |