A) The proportion of mice infected increases as the relative dose increases. The data reveals a sigmoidal pattern that fits well with the model that predicts that infection probability approaches zero at low doses and approaches one at high doses. In the data, infection probability was non zero as low as relative dose –3, but at relative doses –4 (the lowest dose tested) no mice were infected. However, the data at relative dose -3 and the model both indicate that a much larger sample size than the one used (N = 11) would be needed to find at least one infected mice at this dose. These results are consistent with the hypothesis that there exists no safe dose of prions. The modelled infection probability has no free parameters (

) rendering the agreement between theory and data the more convincing. B) Two transformations are applied to the probability of infection data and the results are plotted against relative dose. Plotted as open circles is the transformation

. Using this transformation a straight line would indicate that the data are consistent with the model. Plotted as filled circles is the logit transformation,

, a function that is typically used to transform s-shaped data. Both transformations show similar results – plots that are close to linear from relative doses -2 to 1. Either side of these doses the data are almost linear, but less consistent with the model. Interestingly, at relative dose

the probability of infection is slightly greater than predicted by the model. This is counter to what would be expected if infection probability was governed by a threshold dose. C and D) The probability of infection is plotted against dose in ID50s for low relative doses (C, ≤0.1 ID50s) and higher relative doses (D; ≤10 ID50s). These figures shows how the model assumes a linear relationship between dose and probability of infection at low doses (black lines). The data also supports the assertion that the probability of infection – whilst marginally greater than predicted by the model – is approximately linear at low doses. For comparison, the red line in each of these figures represents a linear relationship which necessarily has 0% probability at no dose and 50% probability of infection at the ID50. This comparison reveals that as dose increases, the relationship becomes increasingly less linear, particular beyond the ID50. E) The mean incubation period is dose-dependent. For relative doses above zero (the ID50), mean incubation period decreases linearly with the relative dose, whereas for relative doses below zero, it is relatively invariant to the relative dose. The model is fitted to the incubation period data using least squares to estimate two parameters. F) The observed variation in incubation periods is markedly greater than the variance predicted by the model. In this panel the circles show the observed variance of the difference from the group mean incubation period for mice from groups with at least two mice (, column 9). The black line represents model predictions of the variation of the incubation period. Since the net growth rate (

*β-μ*) determines the maximum variance (see ), this model prediction assumes that the net growth is equal to that estimated from the mean incubation period data shown in panel E.

## PubMed Commons