fit_msn: Multivariate skew-normal mixture model clustering

In spruce: Spatial Random Effects Clustering of Single Cell Data

Multivariate skew-normal mixture model clustering

Description

Implement Gibbs sampling for MSN model with no spatial random effects

Usage

fit_msn(Y, K, nsim = 2000, burn = 1000, z_init = NULL)

Arguments

Y	An n x g matrix of gene expression values. n is the number of cell spots and g is the number of features.
K	The number of mixture components to fit.
nsim	Number of total MCMC iterations to run.
burn	Number of MCMC iterations to discard as burn in. The number of saved samples is nsim - burn.
z_init	Optional initialized allocation vector. Randomly initialized if NULL.

Value

a list of posterior samples

Examples

 
# parameters
n <- 100
g <- 3 # number of features
K <- 3 # number of clusters (mixture components)
pi <- rep(1/K,K) # cluster membership probability
z <- sample(1:K, size = n, replace = TRUE, prob = pi) # cluster indicators
z <- remap_canonical2(z)
t_true <- truncnorm::rtruncnorm(n,0,Inf,0,1)
t <- t_true

# Cluster Specific Parameters
# cluster specific means
Mu <- list(
  Mu1 = rnorm(g,-5,1),
  Mu2 = rnorm(g,0,1),
  Mu3 = rnorm(g,5,1)
)
# Cluster speficic skewness
Xi <- list(
  Xi1 = rep(2,g),
  Xi2 = rep(0,g),
  Xi3 = rep(-3,g)
)
# cluster specific variance-covariance
S <- matrix(1,nrow = g,ncol = g) # covariance matrix
diag(S) <- 1.5
Sig <- list(
  Sig1 = S,
  Sig2 = S, 
  Sig3 = S
)

Y <- matrix(0, nrow = n, ncol = g)
for(i in 1:n)
{
  Y[i,] <- mvtnorm::rmvnorm(1,mean = Mu[[z[i]]] + t[i]*Xi[[z[i]]],sigma = Sig[[z[i]]])
}

# fit model
fit1 <- fit_msn(Y,3,10,0)

spruce documentation built on March 18, 2022, 7:01 p.m.