Table 4.
Summary of 12 pipeline comparisons.
| Lyve-SET | kSNP | RealPhy | Snp-Pipeline | SNVPhyl | wgMLST | |
|---|---|---|---|---|---|---|
| Tree sensitivity (Sn)a | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| Tree specificity (Sp)a | 100.0% | 90.2% | 100.0% | 100.0% | 100.0% | 100.0% |
| Average of Sn and Sp | 100.0% | 95.1% | 100.0% | 100.0% | 100.0% | 100.0% |
| Kendall-Colijn (λ = 0)b | – | 1.26E-02 | 7.51E-03 | 9.28E-03 | 9.15E-02 | 1.00E-04 |
| Robinson-Fouldsb | – | 3.16E-69 | 6.79E-40 | 5.39E-74 | 9.61E-49 | 1.55E-147 |
| Mantel | – | 0.60 | 0.77 | 0.77 | 0.79 | 0.74 |
| SNP ratioc,d | – | 0.53, 0.78 | 0.97, 0.84 | 1.61, 1.75 | 0.67, 0.84 | 0.69, 0.72 |
| Goodness-of-fit (R2)d | – | 0.46, 0.42 | 0.7, 0.75 | 0.77, 0.3 | 0.83, 0.68 | 0.75, 0.72 |
| Genome analyzede | 25.9% | 0.1% | 84.8% | 0.3% | 82.1% | 88.2% |
Average percentage from 11 outbreaks. The S. enterica outbreak 1203NYJAP-1 was removed as an outlier because all pipelines except wgMLST produced errors with grouping outbreak vs. non-outbreak isolates. Therefore this dataset was removed from the Sn and Sp calculations as an outlier.
Geometric mean.
Number of SNPs per Lyve-SET SNP, averaged across 12 outbreaks. For wgMLST, this is the number of alleles per Lyve-SET SNP.
The average for 12 outbreaks. First value is for all data points; second value is for distances between only outbreak-associated genomes.
The average for 12 outbreaks. Percentage of the reference genome included for analysis. For wgMLST, the average percentage was calculated by obtaining each GenBank-formatted file with annotated wgMLST loci and calculating the breadth of coverage for all loci.