Table 1.
Positives | n | IUPMStats | IUPMBayes | ||||
---|---|---|---|---|---|---|---|
Estimate | Lower 95% | Upper 95% | Median | Lower 95% | Upper 95% | ||
0 | 3 | 0.17* | 0 | 0.75 | 0.34 | 0.05 | 1.10 |
1 | 16 | 0.29 | 0.04 | 2.06 | 0.61 | 0.15 | 1.68 |
2 | 28 | 0.69 | 0.17 | 2.85 | 1.04 | 0.36 | 2.41 |
3 | 37 | 1.39 | 0.41 | 4.72 | 1.36 | 0.52 | 2.87 |
4 | 16 | ∞ | Undefined | 1.85 | 0.80 | 3.63 |
We summarized the results of each method on 100 simulations of 4 replicate wells with cells each, where the true IUPM was set to 1.0 with five sequence variants of equal frequency (1:1:1:1:1). We obtained results for the IUPMStats method directly from the online calculator. Entries for IUPMBayes were averaged across replicate simulations for each number of positives. *IUPMStats uses a median posterior estimate instead of the maximum likelihood estimate (0) when none of the wells are positive