call_cpp {BeviMed}

R interface to BeviMed c++ MCMC procedure

Description

Allows other functions in the package to call the c++ function passing arguments more succinctly and by name.

Usage

call_cpp(
  samples_per_chain,
  y,
  block_starts,
  block_ends,
  cases,
  counts,
  min_ac,
  tau_shape,
  pi_shape,
  omega_shape,
  temperatures,
  z0_matrix,
  estimate_omega,
  logit_omegas,
  logit_omega_proposal_sds,
  variant_weights,
  estimate_phi,
  log_phis,
  log_phi_mean,
  log_phi_sd,
  log_phi_proposal_sds,
  chain_swaps_per_cycle,
  annealing,
  tandem_variant_updates,
  comphet_variant_block_starts,
  comphet_variant_block_ends,
  comphet_variants,
  return_z_trace,
  return_x_trace,
  burn = 0,
  check = TRUE
)

Arguments

samples_per_chain

Number of samples to draw from each chain.

y

Logical vector of subject affectedness status.

block_starts

Integer vector of k 0-indexed start positions (with respect to cases and counts) for contiguous blocks relating to the k variants.

block_ends

Integer vector of (exclusive) k 0-indexed end positions.

cases

0 based vector of case indices with respect to y.

counts

Vector of variant counts.

min_ac

Integer vector with a length equalling the number of individuals or length 1 (in which case the given value is used for all individuals) giving the minimum number of alleles at pathogenic variant sites each individual requires in order to classify as having a ‘pathogenic allele configuration’. Thus, this parameter encodes the mode of inheritance. For instance, setting this parameter to 1 corresponds to dominant inheritance. If there are differences in ploidy between individuals in the locus, it is necessary to set it on an sample level basis - e.g. to ensure sex is accounted for if the locus lies on the X chromosome.

tau_shape

Beta distribution parameterisation of benign variant configuration rate of affection, q.

pi_shape

Beta distribution parameterisation of pathogenic variant configuration rate of affection, p.

omega_shape

Beta distribution of global rate of pathogenicty of variants in gene given pathogenicity of gene, omega.

temperatures

Numeric vector of temperatures of power posteriors. One chain will be created for each element of the vector at the corresponding temperature.

z0_matrix

Matrix of logicals, where the rows are used as an initial zs for the chains.

estimate_omega

Logical value determining whether to estimate the parameter omega.

logit_omegas

Numeric vector of logit omega values, one value per chain.

logit_omega_proposal_sds

Numeric vector of proposal standard deviations for Metropolis-Hastings sampling of logit omega parameter, one value per chain.

variant_weights

Vector of log-odds off-sets for rates of pathogenicity of individual variants relative to the global rate, omega.

estimate_phi

Logical value determining whether to estimate a scaling factor of variant_weights.

log_phis

Numeric vector of log phi values, one value per chain.

log_phi_mean

Mean for normal prior on scaling factor phi.

log_phi_sd

SD for normal prior on scaling factor phi.

log_phi_proposal_sds

Numeric vector of proposal standard deviations for Metropolis-Hastings sampling of log phi parameter, one value per chain.

chain_swaps_per_cycle

Number of chain swaps to propose per update cycle.

annealing

Logical value determining whether to anneal the chains, e.g. for optimisation.

tandem_variant_updates

Number of tandem variant updates to make per update cycle.

comphet_variant_block_starts

0-indexed start positions for contiguous blocks of variants in comphet_variants.

comphet_variant_block_ends

As comphet_variant_block_starts for (exclusive) stop positions.

comphet_variants

Integer vector giving variant numbers (0-based, i.e. between 0 and k-1). Used to pick pairs of variants for tandem updates from.

return_z_trace

Logical value determining whether to store the z-vectors for each chain, which uses alot of memory, particularly if samples_per_chain, k and length(temperatures) are large.

return_x_trace

Logical value determining whether to store the x variable determined by success samples of z. Potentially uses alot of memory, particularly if samples_per_chain, k and length(temperatures) are large.

burn

Number of samples to drop from the start of the chain.

check

Logical value indicating whether to perform validation on the arguments before calling the c++ function.

Value

Object of class BeviMed_raw, containing the output of the MCMC sampling.