Sample mixtures with random genotypes and random parameters according to priors

Usage

sample_mixtures(
  n,
  contributors,
  freqs,
  linkage_map,
  sampling_parameters,
  model_settings,
  sample_model,
  pedigree,
  results_directory,
  seed,
  number_of_replicates = 1L,
  write_non_contributors = FALSE,
  tag = "simulation",
  silent = FALSE
)

Arguments

n: Integer. Number of samples.
contributors: Character vector with unique names of contributors. Valid names are "U1", "U2", ... for unrelated contributors or the names of pedigree members for related contributors.
freqs: Allele frequencies (see read_allele_freqs)
linkage_map: (optional) A linkage map specifying the recombination fractions between loci. If missing, loci are assumed to be independent. See also sample_many_pedigree_genotypes.
sampling_parameters: List. Passed to the sample_model function.
model_settings: List. Passed to the sample_model function.
sample_model: Function such as sample_log_normal_model.
pedigree: (optionally) ped object. Contributors can be named pedigree members.
results_directory: (optionally) Character with path to directory where results are written to disk.
seed: (optionally) Integer seed value that can be used to get reproducible runs. If results are written to disk, the 'Run details.txt' file will contain a seed that can be used for reproducing the result.
number_of_replicates: Integer. Number of replicate simulations for each sample.
write_non_contributors: Logical. If TRUE, sampled genotypes for non-contributing pedigree members will also be written to disk. Defaults to FALSE.
tag: Character. Used for sub directory name when results_directory is provided.
silent: Logical. If TRUE, then no message will be printed about where the output (if any) was written to disk.

Value

If results_directory is provided, this function has the side effect of writing results to disk.

Return value is a list with simulation results:

call matched call
smash DataFrame with all samples in SMASH format (see SMASH_to_wide_table)
samples Detailed results for each sample
parameter_summary DataFrame with parameters for each sample

Examples

freqs <- read_allele_freqs(system.file("extdata","FBI_extended_Cauc_022024.csv",
                           package = "simDNAmixtures"))
gf <- gf_configuration()

sampling_parameters <- list(min_mu = 50., max_mu = 5e3,
                           min_cv = 0.05, max_cv = 0.35,
                           degradation_shape1 = 10, degradation_shape2 = 1)

mixtures <- sample_mixtures(n = 2, contributors = c("U1", "U2"), freqs = freqs,
                            sampling_parameters = sampling_parameters,
                            model_settings = gf$gamma_settings_no_stutter,
                            sample_model = sample_gamma_model)

# sample a mixture of two siblings taking into account

linkage_map <- data.frame(chromosome = c("12","12"),
                          locus = c("vWA", "D12391"),
                          position = c(16.56662766, 29.48590551))

ped_sibs <- pedtools::nuclearPed(children = c("Sib1", "Sib2"))

sibs_mix <- sample_mixtures(n = 1, contributors = c("Sib1", "Sib2"),
                            freqs = freqs,
                            linkage_map = linkage_map,
                            pedigree = ped_sibs,
                            sampling_parameters = sampling_parameters,
                            model_settings = gf$gamma_settings_no_stutter,
                            sample_model = sample_gamma_model)

# an example using the semi-continuous drop model

drop_model_sampling_parameters <- list(min_dropout_probability. = 0.,
                                       max_dropout_probability. = 0.5)

drop_model_settings <- list(locus_names = gf$autosomal_markers,
                            size_regression = gf$size_regression)

mixtures <- sample_mixtures(n = 2, contributors = c("U1", "U2"), freqs = freqs,
                            sampling_parameters = drop_model_sampling_parameters,
                            model_settings = drop_model_settings,
                            sample_model = sample_drop_model)