Heterogeneous dose estimation (Mixed Poisson model)

Method based on the paper by Pujol, M. et al. (2016). A New Model for Biological Dose Assessment in Cases of Heterogeneous Exposures to Ionizing Radiation. Radiation Research, 185(2), 151-162. <doi:10.1667/RR14145.1>.

Usage

estimate_hetero_mixed_poisson(
  case_data,
  fit_coeffs,
  fit_var_cov_mat,
  conf_int = 0.95,
  protracted_g_value = 1,
  gamma,
  gamma_error
)

Arguments

case_data

Case data in data frame form.

fit_coeffs

Fitting coefficients matrix.

fit_var_cov_mat

Fitting variance-covariance matrix.

conf_int

Confidence interval, 95% by default.

protracted_g_value

Protracted \(G(x)\) value.

gamma

Survival coefficient of irradiated cells.

gamma_error

Error of the survival coefficient of irradiated cells.

Value

List containing estimated mixing proportions data frame, estimated yields data frame, estimated doses data frame, estimated fraction of irradiated blood data frame, AIC, and conf_int_* used.

Examples

#The fitting RDS result from the fitting module is needed. Alternatively, manual data
#frames that match the structure of the RDS can be used:


fit_coeffs <- data.frame(
  estimate   = c(0.001280319, 0.021038724, 0.063032534),
  std.error  = c(0.0004714055, 0.0051576170, 0.0040073856),
  statistic  = c(2.715961, 4.079156, 15.729091),
  p.value    = c(6.608367e-03, 4.519949e-05, 9.557291e-56),
  row.names =  c("coeff_C", "coeff_alpha", "coeff_beta")
)


fit_var_cov_mat <- data.frame(
  coeff_C      = c(2.222231e-07, -9.949044e-07,  4.379944e-07),
  coeff_alpha  = c(-9.949044e-07, 2.660101e-05, -1.510494e-05),
  coeff_beta   = c(4.379944e-07, -1.510494e-05, 1.605914e-05),
  row.names =  c("coeff_C", "coeff_alpha", "coeff_beta")
)


case_data <- data.frame(
  ID= "example1",
  N = 361,
  X = 100,
  C0 = 302,
  C1 = 28,
  C2 = 22,
  C3 = 8,
  C4 = 1,
  C5 = 0,
  y = 0.277,
  y_err = 0.0368,
  DI = 1.77,
  u = 10.4
)


#FUNCTION ESTIMATE_HETERO_MIXED_POISSON
estimate_hetero_mixed_poisson(
  case_data[1, ],
  fit_coeffs = as.matrix(fit_coeffs),
  fit_var_cov_mat = as.matrix(fit_var_cov_mat),
  conf_int = 0.95,
  protracted_g_value = 1,
  gamma = 1 / 2.7,
  gamma_error = 0
)
#> Warning: no observations informative at iteration 1
#> Warning: glm.fit: algorithm did not converge
#> Error in object$coefficients: $ operator is invalid for atomic vectors