writeEstProfile: Write the estimated profile of copy number data

In paolarancoita/mBPCR: Bayesian Piecewise Constant Regression for DNA copy number estimation

Description

Function to write nicely the results of the copy number profile estimation. The function either writes the tables directly on a tab delimited file or returns the corresponding tables.

Usage

  writeEstProfile(path='', sampleName='', snpName, chr, position, logratio,
                  chrToBeWritten, estPC, estBoundaries=NULL, postProbT=NULL,
                  regrCurve=NULL, regr=NULL)

Arguments

path	path of the folder where the user wants to write the results of the estimation (it must end with '\' in windows, or '//' in linux). If path='', they will be written in the working directory. If path=NULL, the tables will not be written on a file, but only returned by the function.
sampleName	name of the sample. If the name of the sample if provided, it is used to named the files.
snpName	array containing the name of each probe
chr	array containing the name of the chromosome to which each probe belongs. The possible values of the elements of chr are: the integers from 1 to 22, 'X' and 'Y'.
position	array containing the physical position of each probe
logratio	array containing the log2ratio of the raw copy number data
chrToBeWritten	array containing the name of the estimated chromosomes, of which the user wants to write the results. The possible values of the chromosomes are: the integers from 1 to 22, 'X' and 'Y'.
estPC	array containing the estimated copy number profile as a piecewise constant function
estBoundaries	list containing the vectors of the estimated breakpoints, for each of the chromosomes mentioned in chrToBeWritten. If estBoundaries=NULL, then this information is not written.
postProbT	list containing the vectors of the posterior probabilities to be a breakpoint of the estimated breakpoints, for each of the chromosomes mentioned in chrToBeWritten. If postProbT=NULL, then this information is not written in the file containing the estimated breakpoints.
regrCurve	array containing the estimated regression curve. If regrCurve=NULL, then the file containing this information is not written.
regr	choice of the computation of the regression curve. If regr=NULL, then the regression curve was not computed (then the file containing this information is not written), if regr="BRC" the Bayesian Regression Curve with K_2 was computed (BRC with K_2), if regr="BRCAk" the Bayesian Regression Curve Averaging over k was computed (BRCAk).

Details

The function writes or returns at maximum three tables:

-one containing the estimated profile with mBPCR (the columns are: 'SNPname', 'chromosome', 'position', 'rawLog2ratio', 'mBPCRestimate')

-one containing a summary about the estimated profile with mBPCR (the columns are: 'SNPname(start)', 'SNPname(end)', 'chromosome', 'position(start)', 'position(end)', 'nProbes', 'mBPCRestimate' and, eventually, 'breakpointPostProb'). This table is not created if estBoundaries=NULL.

-one containing the estimated profile with a regression curve (the columns are: 'SNPname', 'chromosome', 'position', 'rawLog2ratio' and the name of the regression curve used). This table is not created if regrCurve=NULL.

Examples

##import the 10K data of cell line REC
data(rec10k)
##estimation of chromosome 5
results <- estProfileWithMBPCR(rec10k$SNPname, rec10k$Chromosome, rec10k$PhysicalPosition, 
rec10k$log2ratio, chrToBeAnalyzed=5, maxProbeNumber=2000)
##write the estimated profile of chromosome 5 in a file in  the working directory
writeEstProfile(path='', sampleName='rec10k', rec10k$SNPname, rec10k$Chromosome, 
rec10k$PhysicalPosition, rec10k$log2ratio, chrToBeWritten=5, results$estPC, results$estBoundaries, 
results$postProbT)


#### the same result can be obtained in the following way, by using the function computeMBPCR
#### for the estimation
##estimation of the global parameters
#param <- estGlobParam(rec10k$log2ratio)
##estimation of chromosome 5
#results <- computeMBPCR(rec10k$log2ratio[rec10k$Chromosome == 5], nu=param$nu, 
#rhoSquare=param$rhoSquare, sigmaSquare=param$sigmaSquare)
##write the estimated profile of chromosome 5 in a file in the working directory
#estPC <- array(dim=length(rec10k$SNPname))
#estBoundaries <- list(dim=1)
#postProbT <- list(dim=1)
#estPC[rec10k$Chromosome == 5] <- results$estPC
#estBoundaries[[1]] <- results$estBoundaries
#postProbT[[1]] <- c(results$postProbT[results$estBoundaries[-results$estK]],1)
#writeEstProfile(path='', sampleName='rec10k', rec10k$SNPname, rec10k$Chromosome, 
#rec10k$PhysicalPosition, rec10k$log2ratio, chrToBeWritten=5, estPC, estBoundaries, postProbT)

paolarancoita/mBPCR documentation built on Jan. 12, 2020, 6:24 p.m.