Mapping the drivers of within-host pathogen evolution using massive data sets

Palmer, Duncan S.; Turner, Isaac; Fidler, Sarah; Frater, John; Goedhals, Dominique; Goulder, Philip; Huang, Kuan-Hsiang Gary; Oxenius, Annette; Phillips, Rodney; Shapiro, Roger; Vuuren, Cloete van; McLean, Angela R.; McVean, Gil

where λ esc and λ esc , β, c and f HLA are estimates of the population escape and reversion rates, transmission probability, contact rate, and proportion of individuals in the population with the HLA type under investigation. Λ 0 and Λ 1 are the proportion of HLA mismatched hosts with an escape mutant at the investigated site, and proportion of HLA matched hosts with an escape mutant at the investigated site. We note that neither β nor c are required to preserve the ordering of escape and reversion rates, and as such do not affect the associated ROC curves. The ordering is also preserved at the stationary phase [8] . In cases where the resultant escape or reversion rate estimate is negative, we set the escape rate estimate at zero under the assumption that the model does not fit the simulated data.