An exponential distribution is fitted to the individual decay rates, and the best\fit parameter and 95% confidence interval were reported. severityAdmitted to hospital with severe/severe Acrizanib disease moderately, (%)3 (100)15 (68.2)4 (5.1)1 (1.9) Open in a separate window Isotype responses for each antigen were plotted against time (Figure?1). Participant samples were grouped by days/months post\infection [7C62?days (0C2?months, value. is measured in days. This method will likely underestimate the decay rate if the period in which antibodies increase after infection (growth period) for some individuals is long. It is also strongly affected by potential differences in both the size and timing of the peak level of NAbs between individuals. The best\fit coefficients with 95% confidence interval are a?=?66 (56.7, 75.3) and b?=?0.0011 (0.00005, 0.00217). 2. Decay and Growth, which uses a two\part model fitted to the full data set Acrizanib Inhibition={adaysift<cacexp(\b(t\c))iftc This allows an initial growth period to account for the underestimate that results from fitting a decay curve to measurements still in the growth phase. However, strong differences between individuals, particular in the timing of the peak level of NAbs, will affect estimations still. GHRP-6 Acetate This model predicts a faster decay speed than the exponential decay model b slightly?=?0.0016 (0.00053, 0.00274) that starts after 11?days [c?=?11 (5.65, 16.35)]. 3. Individual decay model Of the 189 data points for NAbs, 127 are multiple measurements from 50 individuals, of which 33 (88 measurements) are in the exponential decay stage; that is, inhibition is decreasing with time. The individual decay model incorporates data from these 33 individuals only. An exponential decay curve is separately fitted to each individual. An exponential distribution is fitted to the individual decay rates, and the best\fit parameter and 95% confidence interval were reported. The mean decay rate is faster than predicted by the other models with Acrizanib b?=?0.00476 (95% confidence interval 0.00348, 0.00691). Conflict of interest The authors declare no conflict of interest. Author contributions Alana L Whitcombe: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Visualization; Writing\original draft; Writing\review & editing. Reuben McGregor: Data curation; Formal analysis; Investigation; Methodology; Visualization; Writing\original draft; Writing\review & editing. Alyson Craigie: Conceptualization; Data curation; Investigation; Methodology; Resources. Alex James: Data curation; Investigation; Methodology; Writing\review & editing. Richard Charlewood: Data curation; Investigation; Resources. Natalie Lorenz: Investigation; Methodology. James MJ Dickson: Methodology; Resources. Campbell R Sheen: Funding acquisition; Methodology; Resources. Barbara Koch: Methodology; Resources. Shivani Fox\Lewis: Methodology; Resources. Gary McAuliffe: Investigation; Resources. Sally A Roberts: Investigation; Resources. Susan C Morpeth: Investigation; Resources. Susan Taylor: Investigation; Resources. Rachel H Webb: Funding acquisition; Investigation; Resources. Susan Jack: Investigation; Resources. Arlo Upton: Conceptualization; Investigation; Resources. James Ussher: Conceptualization; Investigation; Writing\review & editing. Nicole J Moreland: Conceptualization; Formal analysis; Funding acquisition; Writing\original draft; Writing\review & editing. Supporting information ? Click here for additional data file.(899K, pdf).