BOOST.Ising: BOOST-Ising Model for Dichotomised Expression Levels

In estfernan/boost: Bayesian Modeling of Spatial Transcriptomics Data

BOOST-Ising Model for Dichotomised Expression Levels

Description

Fit the BOOST-Ising model to detect whether the gene is spatially variable (SV). The fit is done within a double Metropolis-Hastings (DMH) algorithm. Only one gene must be present and the expression levels must be dichotomised.

Usage

BOOST.Ising(
  bin.expr,
  neighbor.info,
  gene.name = NULL,
  mean.omega0 = 1,
  sigma.omega0 = 2.5,
  mean.theta = 0,
  sigma.theta = 1,
  n.iter = 10000,
  burn.prop = 0.5
)

Arguments

bin.expr	A numeric vector p of length n that denotes the dichotomised gene expression levels. Each entry is one if the gene is highly expressed at spot i and zero otherwise.
neighbor.info	An n-by-K numeric matrix A that denotes the long format of the adjacency matrix. Each entry denotes the neighbor for spot i.
gene.name	A character string that specifies the name of the gene passed. To be used when storing the results. The default value is NULL to keep the gene expression levels unnamed.
mean.omega0	A numeric value that denotes the prior mean of the normally-distributed first-order intensity parameter. The default is a mean of one.
sigma.omega0	A numeric value that denotes the prior standard deviation of the normally-distributed first-order intensity parameter. The default is a standard deviation of 2.5.
mean.theta	A numeric value that denotes the prior mean of the normally-distributed interaction parameter. The default is a mean of 0.
sigma.theta	A numeric value that denotes the prior standard deviation of the normally-distributed interaction parameter. The default is a standard deviation of 1.
n.iter	An integer value to specify the number of iterations for the DMH algorithm. The default is 10,000 iterations.
burn.prop	A numeric value to specify the proportion of iterations to use as warm-up. The default is 0.50 to use half of the iterations for warm-up.

Details

The primary interest lies in the identification of SV genes via making inferences on the interaction parameter between the low and high-expression states. See Jiang et al. (2021) for more information on the model fitting and posterior inference procedures.

Value

BOOST.Ising returns an object of class "BOOST.Ising". The function base::print() i.e., print.BOOST.Ising(), can be used to print a summary of the results.

An object of class "BOOST.Ising" is a list containing the following components:

call	the function call in which all of the specified arguments are specified by their full names.
model	the name of statistical model or technique.
gene.name	the name of gene evaluated.
summary	a summary table that contains a summary of the estimated parameters.
measures	the estimated Bayes factors and corresponding p-values.
time	the execution time of the function.

References

Jiang, X., Li, Q., & Xiao, G. (2021). Bayesian Modeling of Spatial Transcriptomics Data via a Modified Ising Model. arXiv preprint arXiv:2104.13957.

See Also

normalize.st() for normalizing sequence count data; binarize.st() for dichotomising relative expression levels; print.BOOST.Ising for printing a summary of results to console.

Examples

## Not run: 
library(boost)

## load sample dataset
data(mob)

## extract a sample gene, dichotomise expression levels, and get spatial network
g <- binarize.st(mob, "Apoe", cluster.method = "GMC")
A <- get.neighbors(mob.spots, 4, method = "distance")

## fit the model
res <- BOOST.Ising(g, A, gene.name = "Apoe", n.iter = 500)
print(res)

## End(Not run)

estfernan/boost documentation built on June 24, 2022, 12:20 a.m.