DMRMark {DMRMark}R Documentation

Gibbs Sampler to estimate model parameters

Description

Given the M-values and probe distance, this function calls Gibbs Sampler for estimating the parameters of non-homogeneous hidden Markov model.

Usage

DMRMark(mv, L = rep(1, nrow(mv)), starting = NULL, 
		pd = NULL, initHeuristic = TRUE, 
		GSoptions = NULL)

Arguments

mv

The input M-values matrix, NA is not allowed.

L

A vector to specify the distance between each probes in bp. $L < 0$ represents change of chromosome. Default is $L = 1$ for all probes.

starting

A vector to specify the position to initial new chains. We suggest new chains should be initiated at least at starting of new chromosome. When it is null, new chains initiate at beginning and where $L > 100000 or $L < 0$.

pd

A design matrix, which can be generated by 'stats::model.matrix'. If the M-values are totally paired or single paired, just leave it to be NULL.

initHeuristic

If set to TRUE, heuristics will be used for faster computation, which rely on finding good initial value and then using less iterations. . This will mask GS controls parameters of 'GSoptions'. Recommended for getting some quick insight about new study. Default it TRUE.

GSoptions

List of prior parameters and GS control parameters. See MakeGSoptions.

Details

This function is the main functionality of this package. It takes the M-values and probe distance and calls Gibbs Sampler for estimating the parameters of non-homogeneous hidden Markov model. New chains will be initiated at positions specified in 'starting'. Depends on the scale of M-values, this function may take certain time to the GS. In this situation user may first set 'initHeuristic = TRUE' for a quick insight.

Value

The return value depends on 'GSoptions$track'. In default situation ('GSoptions$track = FALSE'), the return value is a list contains:

theta

A vector contains posterior means of non-DMC's control groups.

mu

A 2-by-2 matrix, each row corresponding to the paired posterior mean of DMCs.

sigma12

A vector contains posterior means of variance of non-DMC's control groups.

sigmaN

Single value, the posterior mean of variance of non-DMC's between-group difference.

Sigma34

An Array contains posterior means of DMC's Covariance.

charL

Posterior means of characteristic length.

init

The probabilities of the initial states of all chains. Sum up to 1.

If 'GSoptions$track = TRUE', an additional dimension will be added to each item of the list, and along this dimension user can retrieve the sample from each iterations.

Author(s)

Linghao SHEN <sl013@ie.cuhk.edu.hk>

See Also

See MakeGSoptions for different prior parameters and Gibbs Sampler control parameters. See DMRViterbi for interpreting the estimated parameters.

Examples


	# DMRMark
	# DMR detection performed on chr18 of a small BLCA dataset from TCGA
	data(BLCA)
	
	# Use a small subset
	nprobe <- 500 
	# M-values
	mv <- BLCA$mv[1:nprobe,]
	
	# Distance between probes, L<0 indicates acorssing chromosomes 
	L = BLCA$distance[1:nprobe]
	
	# Initialize new chain when probe distance too long 
	# or across different chromosomes
	newChains <- which((L > 100000) | L < 0)
	# The starting positions of new chains
	starting <- c(1, newChains[-length(newChains)]+1)
	
	# Run DMRMark with default options
	pars <- DMRMark(mv, L, starting)
	pars

[Package DMRMark version 1.1.1 Index]