Main function to generate transgenerational hologenomic data

Simulates hologenomic data over multiple generations with genetic, microbial, and environmental effects.

Usage

holo_simu(
  h2,
  b2,
  founder_object,
  n_ind = NULL,
  n_clust = NULL,
  n_gen = 5,
  qtn_y = NULL,
  correlation = 0.5,
  otu_g = 0.05,
  lambda = 0.5,
  effect.size = 0.1,
  mix.params = c(0.75, 0.25),
  mix.params.M = c(0.75, 0.25),
  noise.microbiome = 0.1,
  dir = T,
  ao = 25,
  size_rmultinom = 10000,
  selection = F,
  size_selection_F = NULL,
  size_selection_M = NULL,
  selection_type = "GB",
  w.param = c(0.5, 0.5),
  thetaX = NULL,
  env_gen = NULL,
  seed = 1234,
  verbose = T
)

Arguments

h2

direct heritability value, between 0 and 1.

b2

microbiability value, between 0 and 1.

founder_object

output of generate_founder() function.

n_ind

number of individual per generation.

n_clust

vector with taxa assignment, typical output of assign_taxa()

n_gen

number of generation, DEFAULT = 5

qtn_y

number of causal SNPs for the phenotypes.

correlation

Correlation between taxa within the same cluster, value between 0 and 1, DEFAULT = 0.5

otu_g

percentage of taxa under genetic control, between 0 and 1, DEFAULT = 0.05

lambda

proportion of microbiome of offspring coming from vertical transmission, DEFAULT = 0.5

effect.size

Vector giving the size of genetic effect to try.

mix.params

Vector of two numeric values giving the weights for the regularisation of the base population microbiome. mix.params[1] = weight for raw microbiome and mix.params[2] = weight for mean microbiome. DEFAULT = c(0.75,0.25).

mix.params.M

A vector of two numeric values specifying weights between Dirichlet samples and the original mean. DEFAULT = c(0.75,0.25). With mix.params.M[1] the dirichlet microbiome coefficient and mix.params.M[2] the mean microbiome coefficient.

noise.microbiome

sd of microbiome noise, DEFAULT = 0.1

dir

Logical; Mentions if the ambient microbiome is generated via a Dirichlet law or a only the mean_microbiome. DEFAULT = T

ao

A numeric scalar used as the concentration parameter for the Dirichlet distribution.

size_rmultinom

Integer; specifying the total number of object for the multinomial sampling(default: 10000, according to DeruPop.rds dataset).

selection

bool, if selection process needed, DEFAULT = FALSE

size_selection_F

percentage of female to select.

size_selection_M

percentage of male to select.

selection_type

mode of selection to be used, value in ("GB", "B", "G", "diversity", "div.GB"), DEFAULT = "GB"

w.param

in case div.GB selection mode is chosen.

thetaX

Optionnal matrix specifying environmental effects applied to the microbiome, such as antibiotic treatment. This matrix should be of dimension n_taxa x n_individuals, and typically constructed as the product of:

• a vector of taxa-specific effects theta (negative or positive values for all taxa).

• a binary vector X encoding individual exposure (1 for treated, 0 ofr untreated). If NULL (default value), no environmental effect is applied. Example use case: refer to the vignette on Generate figures

env_gen

vector of booleans.

seed

seed value for samplings in the function.

verbose

bool, DEFAULT = T

Value

An important list with several different objects :

• parameters: the set of parameters used for the simulation when calling holo_simu()

• metadata: metadata info such as beta matrix details

• G0 to G5 (5 generations by default): computational characteristics of generations. For each generation, genotypes, microbiomes, phenotypes, pedigree and individuals selected can be reachable.

Required parameters

• h2

• b2

• founder_object

• n_ind

• n_clust

Generation parameters

• n_gen

Genotype-related parameters

• qtn_y

Microbiome-related parameters

• correlation

• otu_g

• lambda

• effect.size

• mix.params

• mix.params.M

• noise.microbiome

• dir

• ao

Selection-related parameters

• selection

• size_selection_F, size_selection_M

• selection_type

Environmental effects

• thetaX

• env_gen

Misc

• seed

• verbose

See also

compute_beta_matrix_cluster(), compute_mean_microbiome(), compute_current_microbiome(), compute_phenotypes()

Examples

if (FALSE) { # \dontrun{
  data("Deru")
  ToyData <- Deru
  taxa_assign_g <- assign_taxa(founder_object = ToyData)
  generations_simu <- holo_simu(h2 = 0.25,
                                 b2 = 0.25,
                                 founder_object = ToyData,
                                 n_clust = taxa_assign_g,
                                 n_ind = 500,
                                 verbose = FALSE,
                                 noise.microbiome = 0.5,
                                 effect.size = 0.3,
                                 lambda = 0.5,
                                 dir = TRUE,
                                 selection = FALSE,
                                 seed = 1234)
} # }