new file mode 100644

@@ -0,0 +1 @@

+Makefile

@@ -1,7 +1,7 @@

 Package: crlmm

 Type: Package

 Title: Genotype Calling (CRLMM) and Copy Number Analysis tool for Affymetrix SNP 5.0 and 6.0 and Illumina arrays.

-Version: 1.13.5

+Version: 1.13.7

 Date: 2010-12-10

 Author: Benilton S Carvalho <Benilton.Carvalho@cancer.org.uk>, Robert Scharpf <rscharpf@jhsph.edu>, Matt Ritchie <mritchie@wehi.edu.au>, Ingo Ruczinski <iruczins@jhsph.edu>, Rafael A Irizarry

 Maintainer: Benilton S Carvalho <Benilton.Carvalho@cancer.org.uk>, Robert Scharpf <rscharpf@jhsph.edu>, Matt Ritchie <mritchie@wehi.EDU.AU>

@@ -14,20 +14,19 @@ Depends: R (>= 2.13.0),

 Imports: affyio (>= 1.19.2),

          ellipse,

          genefilter (>= 1.33.0),

-	 ff (>= 2.2-1),

          mvtnorm,

          preprocessCore (>= 1.13.4),

          splines,

          stats,

          SNPchip,

          utils,

-	 lattice

+	 lattice,

+	 ff

 Suggests: hapmapsnp6,

           genomewidesnp6Crlmm (>= 1.0.4),

           snpStats,

 	  ellipse,

-	  RUnit,

-	  ff (>= 2.2-1)

+	  RUnit

 Collate: AllGenerics.R

 	 AllClasses.R

 	 methods-AssayData.R

@@ -42,9 +41,10 @@ Collate: AllGenerics.R

 	 crlmmGT2.R

          crlmm-illumina.R

 	 snprma-functions.R

+	 cn-functions.R

          utils.R

          zzz.R

 	 test_crlmm_package.R

 LazyLoad: yes

 biocViews: Microarray, Preprocessing, SNP, Bioinformatics, CopyNumberVariants

-Packaged: 2011-04-30 17:10:59 UTC; biocbuild

+Packaged: 2011-04-30 17:10:59 UTC; biocbuild

\ No newline at end of file

@@ -1,4 +1,5 @@

 useDynLib("crlmm", .registration=TRUE)

+

 ##---------------------------------------------------------------------------

 ## Biobase

 ##---------------------------------------------------------------------------

@@ -22,7 +23,7 @@ importFrom(Biobase, assayDataElement, assayDataElementNames,

 ##---------------------------------------------------------------------------

 ## oligoClasses

 ##---------------------------------------------------------------------------

-importClassesFrom(oligoClasses, SnpSuperSet, AlleleSet, CNSet)

+importClassesFrom(oligoClasses, SnpSuperSet, AlleleSet, CNSet)##, ff_or_matrix)

 importMethodsFrom(oligoClasses, allele, calls, "calls<-", confs,

 		  "confs<-", cnConfidence, "cnConfidence<-", isSnp,

 		  chromosome, position, A, B,

@@ -49,9 +50,9 @@ importFrom(stats, coef, cov, dnorm, kmeans, lm, mad, median, quantile, sd, updat

 importFrom(genefilter, rowSds)

 importFrom(mvtnorm, dmvnorm)

 importFrom(ellipse, ellipse)

-importFrom(ff, ffdf, physical.ff, physical.ffdf, ffrowapply)

+##importFrom(ff, ffdf, physical.ff, physical.ffdf, ffrowapply)

-importClassesFrom(oligoClasses, ff_matrix, ffdf)

+##importClassesFrom(oligoClasses, ff_matrix, ffdf)

 ##exportMethods(lines)

 exportMethods(CA, CB)

 export(crlmm,

@@ -70,5 +71,5 @@ export(crlmm,

 export(genotypes, totalCopynumber, rawCopynumber, xyplot)

 exportMethods(A, B, corr, nuA, nuB, phiA, phiB, predictionRegion, posteriorProbability, tau2, Ns, medians, mads,

 	      xyplot, calculateRBaf)

-export(ABpanel, constructInf, preprocessInf, genotypeInf)

+export(ABpanel, constructInf, preprocessInf, genotypeInf, validCEL, celDates)

 exportClasses(PredictionRegion)

@@ -1,4 +1,4 @@

 setOldClass("ellipse")

 setOldClass("ffdf")

-setClassUnion("ff_or_matrix", c("ffdf", "ff_matrix", "matrix"))

+##setClassUnion("ff_or_matrix", c("ffdf", "ff_matrix", "matrix"))

 setClass("PredictionRegion", contains="list")

new file mode 100644

@@ -0,0 +1,117 @@

+copynumber <- function(filenames,

+		       batch,

+		       summaries=c("lrr", "baf", "ca", "cb", "gt", "gtconf"),

+		       write=FALSE,

+		       outdir=".",

+		       onefile=FALSE,

+		       rda=TRUE){

+	fnamelist <- split(filenames, batch)

+	batchlist <- split(batch, batchlist)

+	## for each element in list.  Avoid

+	foreach(i=fnamelist) %do% simpleusage(filenames=fnameslist[[i]], batch=batchlist[[i]])

+}

+

+simpleusage <- function(filenames, batch, ...){

+	object <- genotype(fnamelist, batch, ...)

+	## run genotype on each element in fnamelist

+	object <- genotypeSummary(object)

+	## different than currently implemented

+	##    - shrink to saved batch medians, etc.

+	##    - shinkage across markers (?)

+	object <- shrinkSummary(object)

+	results <- array(NA, nrow(object), ncol(object), length(summaries))

+	if(c("ca", "cb") %in% summaries){

+		## estimateCnParameters

+		##results <- estimatecnParameters(...)

+	} else results <- NULL

+	is.baf <- "baf" %in% summaries || "lrr" %in% summaries

+	if(is.baf){

+		results2 <- calculateRtheta(object) ## returns list

+		## keep as list, or coerce to array

+	}

+	if(write){

+		##write2(, onefile=onefile)

+	}

+	return(results)

+}

+

+

+imputeTheta <- function(ia, ib, theta, S=100){

+##	y <- rbind(y1, y2)

+	y <- theta

+	N <- nrow(y); p <- ncol(y)

+

+	mu0 <- apply(y, 2, mean, na.rm=TRUE)

+	sd0 <- mu0/5

+	L0 <- matrix(.1, p, p); diag(L0) <- 1; L0 <- L0*outer(sd0, sd0)

+	nu0 <- p+2; S0 <- L0

+

+	## starting values

+	Sigma <- S0

+	Y.full <- y

+	O <- 1*(!is.na(y))

+	if(all(O[1, ] == 0)){

+		return(rep(NA, length(ia)))

+	}

+	for(j in seq_len(p)) Y.full[is.na(Y.full[, j]), j] <- mu0[j]

+	## Gibbs sampler

+	##THETA <- SIGMA <- Y.MISS <- NULL

+	## problems:  approx. 90 observations for the means in the other batches

+	##   -- only 3 observations for the mean in the current batch.

+##	THETA <- matrix(NA, iter, p)

+##	SIGMA <- matrix(NA, iter, p^2)

+	Y.MISS <- matrix(NA, S, sum(O[1,]==0))

+	bafs <- matrix(NA, S, length(a))

+	for(s in seq_len(S)){

+		## update lambda

+		ybar <- apply(Y.full, 2, mean)

+		Ln <- solve(solve(L0) + N*solve(Sigma))

+		mun <- Ln%*%(solve(L0)%*%mu0 + N*solve(Sigma)%*%ybar)

+		lambda <- rmvnorm(1, mun, Ln)

+

+		##update sigma

+		Sn <- S0+(t(Y.full)-c(lambda))%*%t(t(Y.full)-c(lambda))

+		Sigma <- solve(rwish(nu0+N, solve(Sn)))

+

+		## update missing data (only care about the first row.)

+		a <- O[1, ] == 1

+		b <- O[1, ] == 0

+		iSa <- solve(Sigma[a,a])

+		beta.j <- Sigma[b,a]%*%iSa

+		Sigma.j <- Sigma[b,b]-Sigma[b,a]%*%iSa%*%Sigma[a,b]

+		lambda.j <- lambda[b]+beta.j%*%(t(Y.full[1,a, drop=FALSE]-lambda[a]))

+		Y.full[1,b] <- rmvnorm(1, lambda.j, Sigma.j)

+		##SIGMA[s, ] <- c(Sigma)

+		Y.MISS[s, ] <- Y.full[1, O[1, ]==0]

+	}

+	na.cols <- which(is.na(y[1, ]))

+	THETA <- matrix(NA, S, 3)

+	THETA[, na.cols] <- Y.MISS

+	if(length(na.cols) < length(ia))

+		THETA[, -na.cols] <- y[1, -na.cols]

+

+	obs.theta <- atan2(ib, ia)*2/pi

+	theta.aa <- matrix(THETA[, 1], S, 3)

+	theta.ab <- matrix(THETA[, 2], S, 3)

+	theta.bb <- matrix(THETA[, 3], S, 3)

+	lessAA <- obs.theta < theta.aa

+	lessAB <- obs.theta < theta.ab

+	lessBB <- obs.theta < theta.bb

+	grAA <- !lessAA ## >= theta.aa

+	grAB <- !lessAB ## >= theta.ab

+	grBB <- !lessBB ## >= theta.bb

+	##not.na <- !is.na(theta.aa)

+	I1 <- grAA & lessAB

+	I2 <- grAB & lessBB

+	##mu <- apply(Y.MISS, 2, mean)

+	##bf <- matrix(NA, S, length(ia))

+	bf <- rep(NA, S*length(ia))

+	I1 <- as.logical(I1); I2 <- as.logical(I2)

+	bf[I1] <- 0.5 * as.numeric(((obs.theta-theta.aa)/(theta.ab-theta.aa)))[I1]

+	bf[I2] <- as.numeric((.5 * (obs.theta - theta.ab) / (theta.bb - theta.ab)))[I2] + 0.5

+	bf[as.logical(lessAA)] <- 0

+	bf[as.logical(grBB)] <- 1

+	bf <- matrix(bf, S, 3, byrow=TRUE)

+	pm.bf <- apply(bf, 2, mean)

+	return(pm.bf)

+}

@@ -53,9 +53,9 @@ getFeatureData <- function(cdfName, copynumber=FALSE){

 	index <- match(rownames(snpProbes), rownames(M)) #only snp probes in M get assigned position

 	M[index, "position"] <- snpProbes[, grep("pos", colnames(snpProbes))]

 	M[index, "chromosome"] <- chromosome2integer(snpProbes[, grep("chr", colnames(snpProbes))])

-	M[index, "isSnp"] <- 1L

-	##index <- which(is.na(M[, "isSnp"]))

 	##M[index, "isSnp"] <- 1L

+	index <- which(is.na(M[, "isSnp"]))

+	M[index, "isSnp"] <- 1L

 	if(copynumber){

 		index <- match(rownames(cnProbes), rownames(M)) #only snp probes in M get assigned position

 		M[index, "position"] <- cnProbes[, grep("pos", colnames(cnProbes))]

@@ -65,9 +65,14 @@ getFeatureData <- function(cdfName, copynumber=FALSE){

 	##A few of the snpProbes do not match -- I think it is chromosome Y.

 	if(any(is.na(M[, "chromosome"])))

 		M[is.na(M[, "chromosome"]), "isSnp"] <- 1L

-	M <- data.frame(M)

-	return(new("AnnotatedDataFrame", data=M))

+	##return(new("AnnotatedDataFrame", data=M))

+	new("GenomeAnnotatedDataFrame",

+	    position=as.integer(M[, "position"]),

+	    chromosome=as.integer(M[, "chromosome"]),

+	    isSnp=as.logical(M[, "isSnp"]),

+	    row.names=featurenames)

 }

+

 getFeatureData.Affy <- getFeatureData

 construct <- function(filenames,

@@ -188,12 +193,18 @@ snprmaAffy <- function(cnSet, filenames,

 	stopifnot(require(pkgname, character.only=TRUE, quietly=!verbose))

 	mixtureParams <- initializeBigMatrix("crlmmMixt-", 4, length(filenames), "double")

 	sampleBatches <- splitIndicesByNode(seq(along=filenames))

-	ocLapply(sampleBatches, rsprocessCEL, filenames=filenames,

-		 fitMixture=TRUE, A=A(cnSet), B=B(cnSet), C=calls(cnSet),

+	ocLapply(sampleBatches,

+		 rsprocessCEL,

+		 filenames=filenames,

+		 fitMixture=TRUE,

+		 A=A(cnSet), B=B(cnSet), C=calls(cnSet),

 		 D=snpCallProbability(cnSet),

 		 SKW=cnSet$SKW, SNR=cnSet$SNR,

-		 mixtureParams=mixtureParams, eps=eps, seed=seed,

-		 mixtureSampleSize=mixtureSampleSize, pkgname=pkgname,

+		 mixtureParams=mixtureParams,

+		 eps=eps,

+		 seed=seed,

+		 mixtureSampleSize=mixtureSampleSize,

+		 pkgname=pkgname,

 		 neededPkgs=c("crlmm", pkgname))

 	if(verbose) message("Cloning A and B matrices to store genotype calls and confidence scores.")

 	return(mixtureParams)

@@ -215,15 +226,14 @@ genotype <- function(filenames,

 		     gender=NULL,

 		     returnParams=TRUE,

 		     badSNP=0.7){

-	stopifnot(require("ff"))

+	stopifnot(isPackageLoaded("ff"))

 	cnSet <- constructAffy(filenames=filenames,

 			       cdfName=cdfName,

 			       batch=batch, verbose=verbose)

 	if(!(is(A(cnSet), "ff") || is(A(cnSet), "ffdf"))){

 		stop("The ff package is required for this function.")

 	}

-

-	mixtureParams <- snprmaAffy(cnSet, filenames=filenames,

+	cnSet@mixtureParams <- snprmaAffy(cnSet, filenames=filenames,

 				    mixtureSampleSize=mixtureSampleSize,

 				    eps=eps,

 				    seed=seed,

@@ -233,7 +243,7 @@ genotype <- function(filenames,

 			verbose=verbose)

 	stopifnot(ok)

 	ok <- genotypeAffy(cnSet=cnSet,

-			   mixtureParams=mixtureParams,

+			   mixtureParams=cnSet@mixtureParams,

 			   SNRMin=SNRMin,

 			   recallMin=recallMin,

 			   recallRegMin=recallRegMin,

@@ -444,7 +444,7 @@ readBPM <- function(bpmFile){

 readGenCallOutput = function(file, path=".", cdfName,

-    colnames=list("SampleID"="Sample ID", "SNPID"="SNP Name", "XRaw"="X Raw", "YRaw"="Y Raw"), 

+    colnames=list("SampleID"="Sample ID", "SNPID"="SNP Name", "XRaw"="X Raw", "YRaw"="Y Raw"),

     type=list("SampleID"="character", "SNPID"="character", "XRaw"="integer", "YRaw"="integer"), verbose=FALSE) {

     if(!identical(names(type), names(colnames)))

@@ -462,9 +462,9 @@ readGenCallOutput = function(file, path=".", cdfName,

 	sepp=s[2]

 	a1=unlist(strsplit(tmp[10][1],s[2]))

     }

-    

+

     if(sum(is.na(match(colnames,a1)))!=0)

-	stop("Cannot find required columns: ", colnames[is.na(match(colnames,a1))], " in ", file, "\nPlease check whether this data was exported.") 

+	stop("Cannot find required columns: ", colnames[is.na(match(colnames,a1))], " in ", file, "\nPlease check whether this data was exported.")

     m1=m=match(a1,colnames)

     m[is.na(m1)==FALSE] =type[m1[!is.na(m1)]]

@@ -472,19 +472,19 @@ readGenCallOutput = function(file, path=".", cdfName,

     names(m) = names(colnames)[m1]

     fc = file(file.path(path, file), open="r")

-	

+

     dat = scan(fc, what=m, skip=10,sep=sepp)

     close(fc)

-	

+

     samples = unique(dat$"SampleID")

     nsamples = length(samples)

     snps = unique(dat$"SNPID")

     nsnps = length(snps)

     if(verbose)

         cat("Check ordering for samples","\n")

-    

+

     X = Y = zeroes = matrix(0, nsnps, nsamples)

-	

+

     for(i in 1:length(samples)) {

         ind = dat$"SampleID"==samples[i]

 	    if(sum(dat$"SNPID"[ind]==snps)==nsnps) {

@@ -502,23 +502,23 @@ readGenCallOutput = function(file, path=".", cdfName,

         }

         gc()

     }

-	

+

     zeroes=(X=="0"|Y=="0")

     colnames(X) = colnames(Y) =  colnames(zeroes) = samples

     rownames(X) = rownames(Y) = snps

-    

-    if(verbose)      

+

+    if(verbose)

         cat("Creating NChannelSet object\n")

-   

+

     XY = new("NChannelSet", X = initializeBigMatrix(name = "X", nr = nrow(X), nc = ncol(X), vmode = "integer", initdata=X),

 			 Y = initializeBigMatrix(name = "Y", nr = nrow(X), nc = ncol(X), vmode = "integer", initdata=Y),

 			 zero = initializeBigMatrix(name = "zero", nr = nrow(X), nc = ncol(X), vmode = "integer", initdata=zeroes),

 			 annotation = cdfName, storage.mode = "environment")

     sampleNames(XY)=colnames(X)

-	

+

     if(verbose)

       cat("Done\n")

-	  

+

     XY

 }

@@ -855,7 +855,7 @@ crlmmIllumina = function(RG, XY, stripNorm=TRUE, useTarget=TRUE,

                   mixtureSampleSize=10^5, eps=0.1, verbose=TRUE,

                   cdfName, sns, recallMin=10, recallRegMin=1000,

                   returnParams=FALSE, badSNP=.7) {

-				  

+

 if(missing(cdfName)) {

    if(!missing(RG))

       cdfName = annotation(RG)

@@ -864,7 +864,7 @@ if(missing(cdfName)) {

 }

 if(!isValidCdfName(cdfName))

    stop("cdfName not valid.  see validCdfNames")

-     

+

 #  if (save.it & (missing(snpFile) | missing(cnFile)))

 #    stop("'snpFile' and/or 'cnFile' is missing and you chose to save.it")

 #  if (load.it & missing(snpFile))

@@ -1213,11 +1213,11 @@ constructInf <- function(sampleSheet=NULL,

 		     call=initializeBigMatrix(name="call", nr, nc),

 		     callProbability=initializeBigMatrix(name="callPr", nr,nc),

 		     annotation=cdfName,

+		     featureData=featureData,

 		     batch=batch)

         if (!is.null(sampleSheet) && !is.null(sampleSheet$Sample_ID)) {

 	   	sampleNames(cnSet) = sampleSheet$Sample_ID

-        }

-        else sampleNames(cnSet) = arrayNames

+        } else  sampleNames(cnSet) = arrayNames

 	if(saveDate){

 		protocolData = getProtocolData.Illumina(grnidats, sep=sep, fileExt=fileExt$green, verbose=verbose)

 	} else{

@@ -1225,7 +1225,7 @@ constructInf <- function(sampleSheet=NULL,

 	}

 	rownames(pData(protocolData)) = sampleNames(cnSet)

 	protocolData(cnSet) = protocolData

-	featureData(cnSet) = featureData

+	##featureData(cnSet) = featureData

 	featureNames(cnSet) = featureNames(featureData)

 	cnSet$gender <- initializeBigVector("gender", ncol(cnSet), vmode="integer")

 	cnSet$SNR = initializeBigVector("crlmmSNR-", ncol(cnSet), "double")

@@ -1248,10 +1248,10 @@ preprocessInf <- function(cnSet,

 		       stripNorm=TRUE,

 		       useTarget=TRUE,

 		       mixtureSampleSize=10^5,

-		       fitMixture=TRUE,

+			  fitMixture=TRUE,

 		       eps=0.1,

 		       verbose=TRUE,

-		       seed=1){

+		       seed=1, cdfName){

 	stopifnot(all(c("gender", "SNR", "SKW") %in% varLabels(cnSet)))

 	open(A(cnSet))

 	open(B(cnSet))

@@ -1304,7 +1304,8 @@ genotypeInf <- function(cnSet, mixtureParams, probs=rep(1/3,3),

 			returnParams=TRUE,

 			badSNP=0.7,

 			gender=NULL,

-			DF=6){

+			DF=6,

+			cdfName){

 	is.snp = isSnp(cnSet)

 	snp.index = which(is.snp)

 ##	narrays = ncol(cnSet)

@@ -1379,6 +1380,7 @@ genotype.Illumina <- function(sampleSheet=NULL,

 	if(missing(cdfName)) stop("must specify cdfName")

 	if(!isValidCdfName(cdfName)) stop("cdfName not valid.  see validCdfNames")

 	stopifnot(!missing(batch))

+	if(is(batch, "factor")) batch <- as.character(batch)

         pkgname <- getCrlmmAnnotationName(cdfName)

 	message("Instantiate CNSet container.")

 	cnSet <- constructInf(sampleSheet=sampleSheet,

@@ -1408,7 +1410,8 @@ genotype.Illumina <- function(sampleSheet=NULL,

 				    fitMixture=fitMixture,

 				    eps=eps,

 				    verbose=verbose,

-				    seed=seed)

+				    seed=seed,

+				       cdfName=cdfName)

 	message("Preprocessing complete.  Begin genotyping...")

 	genotypeInf(cnSet=cnSet, mixtureParams=mixtureParams,

 		   probs=probs,

@@ -1419,7 +1422,8 @@ genotype.Illumina <- function(sampleSheet=NULL,

 		   returnParams=returnParams,

 		   badSNP=badSNP,

 		   gender=gender,

-		   DF=DF)

+		   DF=DF,

+		    cdfName=cdfName)

 	return(cnSet)

 }

@@ -216,3 +216,14 @@ setMethod("annotatedDataFrameFrom", "ffdf", Biobase:::annotatedDataFrameFromMatr

 getBAF <- function(theta, canonicalTheta)

     .Call('normalizeBAF', theta, ct)

+

+validCEL <- function(celfiles){

+	for(i in seq_along(celfiles)){

+		res <- tryCatch(read.celfile(celfiles[i], intensity.means.only=TRUE), error=function(e) NULL)

+		if(is.null(res)) {

+			msg <- message("Problem reading ", celfiles[i])

+			stop(msg)

+		}

+	}

+	return("Successfully read all cel files")

+}

Binary files a/data/cnSetExample.rda and b/data/cnSetExample.rda differ

Binary files a/data/cnSetExample2.rda and b/data/cnSetExample2.rda differ

similarity index 100%

rename from inst/doc/copynumber.Rnw

rename to inst/doc/AffyGW.Rnw

deleted file mode 100644

@@ -1,11 +0,0 @@

-%\VignetteIndexEntry{Preprocessing and genotyping Affymetrix array for copy number analysis}

-%\VignetteDepends{}

-%\VignetteKeywords{crlmm, SNP 6}

-%\VignettePackage{crlmm}

-

-%%% The real .Rnw is under inst/scripts.

-%%% It cannot be added here because it depends on external

-%%% data not yet available as a BioC package

-\documentclass{article}

-\begin{document}

-\end{document}

\ No newline at end of file

deleted file mode 100644

Binary files a/inst/doc/AffymetrixPreprocessCN.pdf and /dev/null differ

@@ -1,9 +1,8 @@

-All: copynumber clean

+All: vignettes clean

 ## all pdfs must be generated from the first target of the Makefile

-copynumber: copynumber.tex

-	cp ../scripts/copynumber.pdf .

-	cp ../scripts/AffymetrixPreprocessCN.pdf .

+vignettes: AffyGW.tex

+	cp ../scripts/AffyGW.pdf .

 	cp ../scripts/IlluminaPreprocessCN.pdf .

 	cp ../scripts/Infrastructure.pdf .

 	cp ../scripts/crlmmDownstream.pdf .

deleted file mode 100644

Binary files a/inst/doc/copynumber.pdf and /dev/null differ

deleted file mode 100644

Binary files a/inst/doc/illumina_copynumber.pdf and /dev/null differ

similarity index 88%

rename from inst/scripts/AffymetrixPreprocessCN.Rnw

rename to inst/scripts/AffyGW.Rnw

@@ -6,6 +6,7 @@

 \documentclass{article}

 \usepackage{graphicx}

 \usepackage{natbib}

+\usepackage{amsmath}

 \usepackage{url}

 \newcommand{\Rfunction}[1]{{\texttt{#1}}}

 \newcommand{\Rmethod}[1]{{\texttt{#1}}}

@@ -74,8 +75,7 @@ pathToCels <- "/thumper/ctsa/snpmicroarray/hapmap/raw/affy/1m"

 if(getRversion() < "2.13.0"){

 	rpath <- getRversion()

 } else rpath <- "trunk"

-##outdir <- paste("/amber1/archive/oncbb/rscharpf/crlmm_vignette/", rpath, sep="")

-outdir <- paste("/amber1/archive/oncbb/rscharpf/crlmm_vignette/", rpath, "/gw6v1.5/", sep="")

+outdir <- paste("/amber1/archive/oncbb/rscharpf/crlmm_vignette/", rpath, "/gw6/", sep="")

 dir.create(outdir, recursive=TRUE, showWarnings=FALSE)

 @

@@ -87,7 +87,6 @@ the genotyping step will be stored in \verb+outdir+.

 ldPath(outdir)

 @

-

 % only needed if cacheing

 <<cachedir, echo=FALSE>>=

 setCacheDir(outdir)

@@ -152,11 +151,7 @@ corrupt cel file. To check if any of the files are corrupt, try

 reading the files in one at a time:

 <<checkcorrupt,eval=FALSE>>=

-require(affyio)

-for(i in seq_along(celFiles)) {

-	cat("Reading ", i, "\n")

-	invisible(read.celfile(celFiles[i], intensity.means.only=TRUE))

-}

+validCEL(celFiles)

 @

@@ -188,10 +183,12 @@ saveObject <- function(outdir, cnSet) {

 (cnset.saved <- saveObject(outdir, cnSet))

 @

-Users can proceed to the \verb+copynumber+ vignette for copy number

-analyses.  See the \verb+Infrastructure+ vignette for additional

-details on the \Robject{CNSet} class, including an overview of the

-available accessors.

+%Users can proceed to the \verb+copynumber+ vignette for copy number

+%analyses.  See the \verb+Infrastructure+ vignette for additional

+%details on the \Robject{CNSet} class, including an overview of the

+%available accessors.

+

+\SweaveInput{copynumber}

 \section{Session information}

@@ -200,4 +197,20 @@ toLatex(sessionInfo())

 @

+\begin{figure}[f]

+  \begin{center}

+  \includegraphics[width=0.6\textwidth]{AffyGW-snr.pdf}

+  \caption{The signal to noise ratio (SNR) for 180 HapMap samples. For

+    Affymetrix platforms, SNR values below 5 can indicate possible

+    problems with sample quality.  In some circumstances, it may be

+    more helpful to exclude samples with poor DNA quality.}

+\end{center}

+\end{figure}

+

+

+\begin{bibliography}

+  \bibliographystyle{plain}

+  \bibliography{refs}

+\end{bibliography}

+

 \end{document}

new file mode 100644

Binary files /dev/null and b/inst/scripts/AffyGW.pdf differ

deleted file mode 100644

Binary files a/inst/scripts/AffymetrixPreprocessCN.pdf and /dev/null differ

@@ -5,6 +5,7 @@

 \documentclass{article}

 \usepackage{graphicx}

 \usepackage{natbib}

+\usepackage{amsmath}

 \usepackage[margin=1in]{geometry}

 \newcommand{\crlmm}{\Rpackage{crlmm}}

 \newcommand{\Rfunction}[1]{{\texttt{#1}}}

@@ -84,15 +85,15 @@ ocProbesets(150e3)

 ocSamples(500)

 @

-\section{Initializing a container for storing processed data}

-

-This section will initialize a container for storing processed forms

-of the data, including the normalized intensities for the A and B

-alleles and the CRLMM genotype calls and confidence scores.  In

-addition, the container will store information on the markers

-(physical position, chromosome, and a SNP indicator), the batch, and

-the samples (e.g., gender).  To construct this container for Infinium

-platforms, several steps are required.

+%\section{Initializing a container for storing processed data}

+%

+%This section will initialize a container for storing processed forms

+%of the data, including the normalized intensities for the A and B

+%alleles and the CRLMM genotype calls and confidence scores.  In

+%addition, the container will store information on the markers

+%(physical position, chromosome, and a SNP indicator), the batch, and

+%the samples (e.g., gender).  To construct this container for Infinium

+%platforms, several steps are required.

 We begin by specifying the path containing the raw IDAT files for a

 set of samples from the Infinium 370k platform.

@@ -143,129 +144,140 @@ processed at similar times (e.g., within a few weeks).

 batch <- rep("1", nrow(samplesheet))

 @

-Finally, we initialize an object of class \Robject{CNSet} using the

-function \Rfunction{constructInf}.

-

-<<container,cache=TRUE>>=

-cnSet <- constructInf(sampleSheet=samplesheet,

-		      arrayNames=arrayNames,

-		      batch=batch,

-		      arrayInfoColNames=arrayInfo,

-		      cdfName=cdfName,

-		      verbose=TRUE,

-		      saveDate=TRUE)

-@

-

-A concise summary of the object's contents can be viewed with the

-\Rfunction{print} function.

-

-<<cnset>>=

-print(cnSet)

-@

-

-Note that the above object does not yet contain any processed data

-(only \verb+NA+'s).  As the elements of the \verb+assayData+ slot are

-\Robject{ff} objects (not matrices), several \verb+.ff+ files now

-appear in the \verb+outdir+. The \verb+.ff+ files should not be

-removed and can be listed using the \R{} function

-\Rfunction{list.files}.  %For the most part, the \emph{appearance} that

+%Finally, we initialize an object of class \Robject{CNSet} using the

+%function \Rfunction{constructInf}.

+%

+%<<container,cache=TRUE>>=

+%cnSet <- constructInf(sampleSheet=samplesheet,

+%		      arrayNames=arrayNames,

+%		      batch=batch,

+%		      arrayInfoColNames=arrayInfo,

+%		      cdfName=cdfName,

+%		      verbose=TRUE,

+%		      saveDate=TRUE)

+%@

+%

+%A concise summary of the object's contents can be viewed with the

+%\Rfunction{print} function.

+%

+%<<cnset>>=

+%print(cnSet)

+%@

+%

+%Note that the above object does not yet contain any processed data

+%(only \verb+NA+'s).  As the elements of the \verb+assayData+ slot are

+%\Robject{ff} objects (not matrices), several \verb+.ff+ files now

+%appear in the \verb+outdir+. The \verb+.ff+ files should not be

+%removed and can be listed using the \R{} function

+%\Rfunction{list.files}.  %For the most part, the \emph{appearance} that

 %the data is stored in memory is preserved.

-

-<<listff>>=

-sapply(assayData(cnSet), function(x) class(x)[1])

-list.files(outdir, pattern=".ff")[1:5]

-@

-

-\section{Preprocessing}

+%

+%<<listff>>=

+%sapply(assayData(cnSet), function(x) class(x)[1])

+%list.files(outdir, pattern=".ff")[1:5]

+%@

+%

+\section{Preprocessing and genotyping}

 The raw intensities from the Infinium IDAT files are read and

 normalized using the function \Rfunction{preprocessInf}.  The function

 \Rfunction{preprocessInf} returns a \Robject{ff} object containing the

 parameters for the mixture model used by the CRLMM genotyping

-algorithm.

-

-<<preprocess,cache=TRUE, results=hide>>=

-mixtureParams <- preprocessInf(cnSet=cnSet, sampleSheet=samplesheet, arrayNames=arrayNames, arrayInfoColNames=arrayInfo)

-@

-

-<<showMixtureParams>>=

-invisible(open(mixtureParams))

-str(mixtureParams[])

-invisible(close(mixtureParams))

-@

-

-Note that the normalized intensities for the A and B alleles are no

-longer \verb+NA+s and can be inspected using the methods \Rfunction{A}

-and \Rfunction{B}, respectively.

-

-<<intensities>>=

-invisible(open(A(cnSet)))

-invisible(open(B(cnSet)))

-as.matrix(A(cnSet)[1:5, 1:5])

-as.matrix(B(cnSet)[1:5, 1:5])

-invisible(close(A(cnSet)))

-invisible(close(B(cnSet)))

-@

-

-\section{Genotyping}

-

-CRLMM genotype calls and confidence scores are estimated using the

-function \Rfunction{genotypeInf}.

-

-<<genotype,cache=TRUE>>=

-updated <- genotypeInf(cnSet, mixtureParams=mixtureParams)

-@

-\vspace{-0.5em}

-<<>>=

-updated

-@

-

-\textbf{Wrapper:} As an alternative to calling the functions

-\Rfunction{constructInf}, \Rfunction{preprocessInf} and

-\Rfunction{genotypeInf} in sequence, a convenience function called

-\Rfunction{genotype.Illumina} is a wrapper for the above functions and

-produces identical results.

+algorithm.  Following preprocessing, the \Rfunction{genotypeInf}

+genotypes the samples. The function \Rfunction{genotype.Illumina} is a

+wrapper to the above functions and returns an object of class

+\Rclass{CNSet}.

+

+%<<preprocess,cache=TRUE, results=hide>>=

+%mixtureParams <- preprocessInf(cnSet=cnSet, sampleSheet=samplesheet,

+%			       arrayNames=arrayNames,

+%			       arrayInfoColNames=arrayInfo,

+%			       cdfName=cdfName)

+%@

+%

+%<<showMixtureParams>>=

+%invisible(open(mixtureParams))

+%str(mixtureParams[])

+%invisible(close(mixtureParams))

+%@

+%

+%Note that the normalized intensities for the A and B alleles are no

+%longer \verb+NA+s and can be inspected using the methods \Rfunction{A}

+%and \Rfunction{B}, respectively.

+%

+%<<intensities>>=

+%invisible(open(A(cnSet)))

+%invisible(open(B(cnSet)))

+%as.matrix(A(cnSet)[1:5, 1:5])

+%as.matrix(B(cnSet)[1:5, 1:5])

+%invisible(close(A(cnSet)))

+%invisible(close(B(cnSet)))

+%@

+%

+%\section{Genotyping}

+%

+%CRLMM genotype calls and confidence scores are estimated using the

+%function \Rfunction{genotypeInf}.

+%

+%<<genotype,cache=TRUE>>=

+%updated <- genotypeInf(cnSet, mixtureParams=mixtureParams, cdfName=cdfName)

+%@

+%\vspace{-0.5em}

+%<<>>=

+%updated

+%@

+%

+%\textbf{Wrapper:} As an alternative to calling the functions

+%\Rfunction{constructInf}, \Rfunction{preprocessInf} and

+%\Rfunction{genotypeInf} in sequence, a convenience function called

+%\Rfunction{genotype.Illumina} is a wrapper for the above functions and

+%produces identical results.

 <<genotype.Illumina,cache=TRUE,results=hide>>=

-cnSet2 <- genotype.Illumina(sampleSheet=samplesheet,

+cnSet <- genotype.Illumina(sampleSheet=samplesheet,

 			    arrayNames=arrayNames,

 			    arrayInfoColNames=arrayInfo,

 			    cdfName="human370v1c",

 			    batch=batch)

 @

-<<checkIdenticalCalls>>=

-invisible(open(calls(cnSet)))

-invisible(open(calls(cnSet2)))

-snp.index <- which(isSnp(cnSet))

-identical(calls(cnSet)[snp.index, 1:20], calls(cnSet2)[snp.index, 1:20])

-invisible(close(calls(cnSet)))

-invisible(close(calls(cnSet2)))

-@

-To fully remove the data associated with the \Robject{cnSet2} object,

-one should use the \Rfunction{delete} function in the \Rpackage{ff}

-package followed by the \Rfunction{rm} function.  The following code

-is not evaluated is it would change the results of the cached

-computations in the previous code chunk.

+Note, to fully remove the data associated with the \Robject{cnSet2}

+object, one should use the \Rfunction{delete} function in the

+\Rpackage{ff} package followed by the \Rfunction{rm} function.  The

+following code is not evaluated is it would change the results of the

+cached computations in the previous code chunk.

 <<delete, eval=FALSE>>=

-lapply(assayData(cnSet2), delete)

-lapply(batchStatistics(cnSet2), delete)

-delete(cnSet2$gender)

-delete(cnSet2$SNR)

-delete(cnSet2$SKW)

-rm(cnSet2)

+lapply(assayData(cnSet), delete)

+lapply(batchStatistics(cnSet), delete)

+delete(cnSet$gender)

+delete(cnSet$SNR)

+delete(cnSet$SKW)

+rm(cnSet)

 @

-Users can proceed to the \verb+copynumber+ vignette for copy number

-analyses.  See the \verb+Infrastructure+ vignette for additional

-details on the \Robject{CNSet} class, including an overview of the

-available accessors.

+%\section{Copy number estimation}

+

+\SweaveInput{copynumber}

 \section{Session information}

 <<sessionInfo, results=tex>>=

 toLatex(sessionInfo())

 @

+\begin{figure}[f]

+  \begin{center}

+  \includegraphics[width=0.6\textwidth]{IlluminaPreprocessCN-snr.pdf}

+  \caption{The signal to noise ratio (SNR) for 180 HapMap samples. For

+    Affymetrix platforms, SNR values below 5 can indicate possible

+    problems with sample quality.  In some circumstances, it may be