Table 1 Variant calling sensitivity and specificity for the consensus 3of4 approach ensures high specificity and FDR without a loss of sensitivity
From: A hybrid computational strategy to address WGS variant analysis in >5000 samples
| Â | Consensus 3of4 | Consensus 2of4 | GATK-HC | GATK-UG | GotCloud | SNPTools |
|---|---|---|---|---|---|---|
# SNVs | 72,945,834 | 86,233,412 | 103,439,411 | 104,649,069 | 78,483,824 | 66,290,585 |
Ti/Tv | 2.12 | 2.08 | 2.00 | 2.00 | 2.09 | 1.99 |
% in 1000G | 50.22 % | 43.75 % | 36.35 % | 36.54 % | 46.91 % | 51.17 % |
% in dbSNP | 40.25 % | 35.31 % | 28.88 % | 29.53 % | 37.93 % | 41.91 % |
sensitivity | 63.80 % | 68.98 % | 68.51 % | 69.99 % | 64.17 % | 51.26 % |
specificity | 99.92 % | 99.70 % | 99.30 % | 99.54 % | 99.86 % | 99.13 % |
FDR | 3.34 % | 11.29 % | 22.91 % | 16.16 % | 6.12 % | 33.11 % |