Skip to contents

Compute product of matrixes based on few parameters

compute_beta_g.Rd

This function compute the product between the matrix of OTUs-specific genetic effects and the matrix of genotypes. This is part of gen_effect_calibration(), where individual_level effects of genotypes on taxa are computed.

Usage

compute_beta_g(beta, genotypes, noise, taxa_scale)

Arguments

beta

A matrix of taxa (in rows) across SNPs (in columns) giving the multiplicative effect of genotype on taxa abundances. Typically obtained from compute_beta_matrix_cluster().

genotypes

A matrix of genotypes with SNPs in rows and individuals in columns.

noise

A numeric scalar indicating the standard deviation of the Gaussian noise to add.

taxa_scale

A numeric scalar to scale the noise added to each OTUs abundance.

Value

A matrix with taxa in roxs and individuals in columns, where each element represents the abundance of a given taxa in an individual, influenced by the genotype.

See also

gen_effect_calibration()

Examples

set.seed(123)
n_taxa <- 5
n_snp <- 10
n_ind <- 50
# Simulate a little beta matrix
beta <- matrix(sample(seq(0.1, 1, by = 0.05), size = n_taxa*n_snp, replace = TRUE), nrow = n_taxa, ncol = n_snp)

# Simulate a little genotype matrix
genotypes <- matrix(sample(0:2, size = n_snp*n_ind, replace = TRUE), nrow = n_snp, ncol = n_ind)

# Compute beta_g matrix with noise
beta_g <- RITHMS:::compute_beta_g(beta = beta,
                                  genotypes = genotypes,
                                  noise = 0.05,
                                  taxa_scale = 1)