smfishHmrf.hmrfem.multi: Performs HMRF for multivariate normal distribution. Accepts R...

In smfishHmrf: Hidden Markov Random Field for Spatial Transcriptomic Data

Description

This function performs HMRF \insertCiteZhu2018smfishHmrf on multivariate normal distributions. Different from other variations, this function accepts R data structures directly as inputs, and only accepts a single value of beta.

This function exists for legacy and compatibility reason. User should use smfishHmrf.hmrfem.multi.it.min function.

Usage

smfishHmrf.hmrfem.multi(
  y,
  neighbors,
  numnei,
  blocks,
  beta = 0.5,
  mu,
  sigma,
  err = 1e-07,
  maxit = 50,
  verbose,
  dampFactor = NULL,
  forceDetectDamp = FALSE,
  tolerance = 1e-60
)

Arguments

y	gene expression matrix
neighbors	adjacency matrix between cells
numnei	a vector containing number of neighbors per cell
blocks	a list of cell colors for deciding the order of cell update
beta	the beta to try (smoothness parameter)
mu	a 2D matrix (i,j) of cluster mean (initialization)
sigma	a 3D matrix (i,j,k) where (i,j) is the covariance of cluster k (initialization)
err	the error that is allowed between successive iterations
maxit	maximum number of iterations
verbose	TRUE or FALSE
dampFactor	the dampening factor
forceDetectDamp	will auto detect a dampening factor instead of using the specified one
tolerance	applicable when forceDetectDamp is set to TRUE

Value

A list of prob, new mu, new sigma, unnormalized prob after iterations finish

More information

Arguments mu and sigma refer to the cluster centroids from running kmeans algorithm. They serve as initialization of HMRF. Users should refer to smfishHmrf.hmrfem.multi.it.min for more information about function parameters and the requirements.

References

\insertRef

Zhu2018smfishHmrf

Examples

    data(seqfishplus)
    s <- seqfishplus
    res<-smfishHmrf.hmrfem.multi(s$y, s$nei, s$numnei, s$blocks, beta=28, 
    mu=s$mu, sigma=s$sigma, err=1e-7, maxit=50, verbose=TRUE, dampFactor=s$damp, 
    tolerance=1e-5)

smfishHmrf documentation built on Jan. 13, 2021, 12:54 p.m.