bpr_fdmm: Gibbs sampling algorithm for BPR mixture model

Description Usage Arguments

View source: R/bpr_fdmm.R

Description

bpr_fdmm implements the Gibbs sampling algorithm for performing clustering on DNA methylation profiles, where the observation model is the Binomial distributed Probit Regression function, bpr_likelihood.

Usage

1
2
3
4
bpr_fdmm(x, K = 2, pi_k = NULL, w = NULL, basis = NULL,
  w_0_mean = NULL, w_0_cov = NULL, dir_a = NULL, gibbs_nsim = 5000,
  gibbs_burn_in = 1000, is_parallel = TRUE, no_cores = NULL,
  is_verbose = FALSE)

Arguments

x

A list of elements of length N, where each element is an L x 3 matrix of observations, where 1st column contains the locations. The 2nd and 3rd columns contain the total trials and number of successes at the corresponding locations, repsectively.

K

Integer denoting the number of clusters K.

pi_k

Vector of length K, denoting the mixing proportions.

w

A MxK matrix, where each column contains the basis function coefficients for the corresponding cluster.

basis

A 'basis' object. E.g. see polynomial.object

w_0_mean

The prior mean hyperparameter for w

w_0_cov

The prior covariance hyperparameter for w

dir_a

The Dirichlet concentration parameter, prior over pi_k

gibbs_nsim

Argument giving the number of simulations of the Gibbs sampler.

gibbs_burn_in

Argument giving the burn in period of the Gibbs sampler.

is_parallel

Logical, indicating if code should be run in parallel.

no_cores

Number of cores to be used, default is max_no_cores - 1.

is_verbose

Logical, print results during EM iterations


andreaskapou/mpgex documentation built on Nov. 25, 2017, 8:08 a.m.