Graph log-likelihood of data

Description

Graph log-likelihood of data

Usage

graph_loglik(
  pop_data,
  curve_params,
  noise_params,
  antigen_isos = pop_data %>% get_biomarker_levels(),
  x = 10^seq(-3, 0, by = 0.1),
  highlight_points = NULL,
  highlight_point_names = "highlight_points",
  log_x = FALSE,
  previous_plot = NULL,
  curve_label = paste(antigen_isos, collapse = " + "),
  ...
)

Arguments

pop_data	a data.frame() with cross-sectional serology data by antibody and age, and additional columns
curve_params	a data.frame() containing MCMC samples of parameters from the Bayesian posterior distribution of a longitudinal decay curve model. The parameter columns must be named: antigen_iso: a character() vector indicating antigen-isotype combinations iter: an integer() vector indicating MCMC sampling iterations y0: baseline antibody level at $t=0$ ($y(t=0)$) y1: antibody peak level (ELISA units) t1: duration of infection alpha: antibody decay rate (1/days for the current longitudinal parameter sets) r: shape factor of antibody decay
noise_params	a data.frame() (or tibble::tibble()) containing the following variables, specifying noise parameters for each antigen isotype: antigen_iso: antigen isotype whose noise parameters are being specified on each row nu: biological noise eps: measurement noise y.low: lower limit of detection for the current antigen isotype y.high: upper limit of detection for the current antigen isotype
antigen_isos	Character vector listing one or more antigen isotypes. Values must match pop_data.
x	sequence of lambda values to graph
highlight_points	a possible highlighted value
highlight_point_names	labels for highlighted points
log_x	should the x-axis be on a logarithmic scale (TRUE) or linear scale (FALSE, default)?
previous_plot	if not NULL, the current data is added to the existing graph
curve_label	if not NULL, add a label for the curve
…	Arguments passed on to log_likelihood verbose logical: if TRUE, print verbose log information to console method how to combine several biomarkers into one likelihood: “composite” (the default) sums the per-biomarker marginal log-likelihoods, giving each biomarker its own latent infection time. This is an independence (composite) likelihood: its point estimate is consistent, but its naive standard error assumes biomarker readings from the same person are independent — see cluster_var in est_seroincidence() for the correction. “joint” integrates once per subject over a shared latent infection time, with the per-biomarker densities multiplied inside the integral (conditional independence given the time since infection). This is the model the methodology article describes. It needs pop_data to identify subjects (see ids_varname()) so that each person’s readings can be paired, a noise model (nu > 0 or eps > 0) for every biomarker, and the same posterior draws (iter) for every biomarker, so that a person’s biomarkers are evaluated under one coherent draw. With a single biomarker the two methods coincide. See issues #637 and #646.

Value

a ggplot2::ggplot()

Examples

library("serocalculator")


library(dplyr)
library(tibble)

# Load cross-sectional data
xs_data <-
  sees_pop_data_pk_100

# Load curve parameters and subset for the purposes of this example
curve <-
  typhoid_curves_nostrat_100 %>%
  filter(antigen_iso %in% c("HlyE_IgA", "HlyE_IgG"))

# Load noise parameters
cond <- tibble(
  antigen_iso = c("HlyE_IgG", "HlyE_IgA"),
  nu = c(0.5, 0.5),                          # Biologic noise (nu)
  eps = c(0, 0),                             # M noise (eps)
  y.low = c(1, 1),                           # Low cutoff (llod)
  y.high = c(5e6, 5e6))                      # High cutoff (y.high)

# Graph the log likelihood
lik_HlyE_IgA <- # nolint: object_name_linter
  graph_loglik(
    pop_data = xs_data,
    curve_params = curve,
    noise_params = cond,
    antigen_isos = "HlyE_IgA",
    log_x = TRUE
)

lik_HlyE_IgA # nolint: object_name_linter