MPR.genotyping (version 0.8)

globalMPRRefine: refine SNPs by resampling

Description

The function to refine SNPs by resampling and Bayesian inference based on results of globalMPRByMarkers.

Usage

globalMPRRefine(baseData, markers = NULL, alleleA = NULL, numGroup = ncol(baseData), 
	groupSort = FALSE, numPerm = 10, numTry = 3, numBaseStep = 50, 
	numBaseCandidateStep = numBaseStep * 2, numKnownStep = numBaseStep/2, 
	numKnownCandidateStep = numBaseStep * 2, useMedianToFindKnown = TRUE, 
	maxNStep = 3, scoreMin = 0.8, useOnlyKnownToType = FALSE, useBayes = FALSE, 
	errorRate = 5e-04, saveMidData = FALSE, verbose = FALSE, strSTART = "\r", 
	strEND = "", ...)

Arguments

baseData

Your SNP matrix

markers

Your markers data

alleleA

We can use the results of globalMPRByMarkers() to be something like markers

numGroup

The number of group to run in one permutation.

groupSort

Your groups (or RILs) will sort by coverage.(higher coverage will make the result in local region better)

numPerm

The number of permutations

numTry

The maximum number of the times of using one SNP from one group (or RIL).

numBaseStep

The number of SNP to run localMPR().

numBaseCandidateStep

The number of SNP candidate in one step

numKnownStep

The number of makers to run localMPR().

numKnownCandidateStep

The number of makers candidate in one step

useMedianToFindKnown

In one local genomic region (window), we will choose the nearest some makers to this region. Median of the region will be the center.

maxNStep

The parameter of localMPR()

scoreMin

The score is used to control the accuracy of MPR in one local genomic region (window).

useOnlyKnownToType

Using different scoring system

useBayes

"useBayes=TRUE" will make this routine integrate with Bayesian inference

errorRate

the parameter of genotypeCallsBayes() if you set "useBayes=TRUE".

saveMidData

If you want to check the effect of the number of permutations, "saveMidData = TRUE" will help you.

verbose

Report verbose progress.

strSTART

The part of displaying format (verbose)

strEND

The part of displaying format (verbose)

…

Arguments to be passed to other methods.

See Also

globalMPRByMarkers

Examples

# NOT RUN {
data(snpData)
data(markerData)

## select 30 markers randomly
set.seed(123);markers <- sample(names(markerData)[10:50],20)

## select SNP sites which contain the 30 markers
ids <- match(markers,rownames(snpData))
str(myBaseData <- snpData[min(ids):max(ids),])

## global MPR aiding with marker data
allele.MPR <- globalMPRByMarkers(myBaseData,markers=markerData,numTry=3,
			numBaseStep=50,numBaseCandidateStep=100,
            numMarkerStep=10,useMedianToFindKnown=TRUE,
			maxNStep=3,scoreMin=0.8,verbose=TRUE)
			 
## then you need to refine the MPR results
set.seed(123);system.time(all.res <- globalMPRRefine(myBaseData,alleleA=na.omit(
            allele.MPR[,1]),numGroup=238,groupSort=TRUE,numPerm=1,numTry=3,
            numBaseStep=50,numBaseCandidateStep=100,numKnownStep=30,
            numKnownCandidateStep=50,useMedianToFindKnown=TRUE,
            maxNStep=3,scoreMin=0.8,saveMidData=TRUE,verbose=TRUE))
# }