@@ -49,9 +49,12 @@ BioC data packages

 2009-03-25 Rob Scharpf - committed version 1.0.61

 * update to copynumber vignette

+2009-03-27 B Carvalho - committed version 1.0.62

+* crlmm() returns a SnpSet object (from Biobase)

+* fixed standard vignette to accommodate this change

+* redefined calls/confs to accommodate this change

-

-

+* fixed documentation to acommodate this change

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

 Package: crlmm

 Type: Package

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

-Version: 1.0.61

+Version: 1.0.62

 Date: 2008-12-28

 Author: Rafael A Irizarry, Benilton S Carvalho <bcarvalh@jhsph.edu>, Robert Scharpf <rscharpf@jhsph.edu>

 Maintainer: Benilton S Carvalho <bcarvalh@jhsph.edu>, Robert Scharpf <rscharpf@jhsph.edu>

@@ -10,7 +10,7 @@ License: Artistic-2.0

 Imports: affyio, preprocessCore, utils, stats, genefilter, splines, Biobase, mvtnorm

 Suggests: hapmapsnp5, genomewidesnp5Crlmm, genomewidesnp6Crlmm (>= 1.0.1), methods, GGdata, snpMatrix

 Collate: AllClasses.R

-         crlmm-methods.R

+         methods-SnpSet.R

 	 cnrma-functions.R

          crlmm-functions.R

          snprma-functions.R

@@ -1,7 +1,4 @@

 useDynLib("crlmm", .registration=TRUE)

-export("crlmm", "list.celfiles", "computeCopynumber", "cnrma", "celDates")

-exportMethods("list2crlmmSet", "calls", "confs")

-exportClasses("crlmmSet")

 import(Biobase)

 importFrom(affyio, read.celfile.header, read.celfile)

 importFrom(preprocessCore, normalize.quantiles.use.target)

@@ -10,3 +7,5 @@ importFrom(stats, coef, cov, dnorm, kmeans, lm, mad, median, quantile, sd)

 importFrom(genefilter, rowSds)

 importFrom(mvtnorm, dmvnorm)

+export("crlmm", "list.celfiles", "computeCopynumber", "cnrma", "celDates")

+exportMethods("calls", "confs")

@@ -189,7 +189,8 @@ instantiateObjects <- function(calls, NP, plate, envir, chrom, A, B,

 			       gender, SNRmin=5, SNR){

 	envir[["chrom"]] <- chrom

 	CHR_INDEX <- paste(chrom, "index", sep="")

-	data(list=CHR_INDEX, package="genomewidesnp6Crlmm")	

+	data(list=CHR_INDEX, package="genomewidesnp6Crlmm")

+        ## Rob, where does 'index[[1]]' come from?

 	A <- A[index[[1]], SNR > SNRmin]

 	B <- B[index[[1]], SNR > SNRmin]

 	calls <- calls[index[[1]], SNR > SNRmin]

@@ -43,7 +43,7 @@ crlmm <- function(filenames, row.names=TRUE, col.names=TRUE,

   res2[["SNR"]] <- res[["SNR"]]

   res2[["SKW"]] <- res[["SKW"]]

-  return(res2)

+  return(list2SnpSet(res2, returnParams=returnParams))

 }

 crlmmGT <- function(A, B, SNR, mixtureParams, cdfName, row.names=NULL,

@@ -122,9 +122,8 @@ crlmmGT <- function(A, B, SNR, mixtureParams, cdfName, row.names=NULL,

                                 newparams[["N"]][, 3]) > recallMin &

                                 !apply(regionInfo, 1, any)),

                                 autosomeIndex)

-  

   if(length(Index) < recallRegMin){

-    warning("Recallibration not possible.")

+    warning("Recalibration not possible.")

     newparams <- params

     dev <- vector("numeric", nrow(newparams[["centers"]]))

     SS <- matrix(Inf, 3, 3)

@@ -221,9 +220,9 @@ crlmmGT <- function(A, B, SNR, mixtureParams, cdfName, row.names=NULL,

   if(verbose) message("Done.")

   if (returnParams){

-    return(list(calls=A, confs=B, SNPQC=dev, batchQC=batchQC, params=params, DD=DD, covDD=SS))

+    return(list(calls=A, confs=B, SNPQC=dev, batchQC=batchQC, params=params, DD=DD, covDD=SS, gender=gender, pkgname=pkgname))

   }else{

-    return(list(calls=A, confs=B, SNPQC=dev, batchQC=batchQC, DD=DD, covDD=SS))

+    return(list(calls=A, confs=B, SNPQC=dev, batchQC=batchQC, DD=DD, covDD=SS, gender=gender, pkgname=pkgname))

   }

 }

deleted file mode 100644

@@ -1,27 +0,0 @@

-## Methods for crlmm

-

-setGeneric("calls", function(x) standardGeneric("calls"))

-setMethod("calls", "crlmmSet", function(x) assayData(x)$calls)

-

-setGeneric("confs", function(x) standardGeneric("confs"))

-setMethod("confs", "crlmmSet", function(x) assayData(x)$confs)

-

-setGeneric("list2crlmmSet", function(x) standardGeneric("list2crlmmSet"))

-setMethod("list2crlmmSet", "list",

-          function(x){

-            pd <- data.frame(SNR=x[["SNR"]],

-                             row.names=colnames(x[["calls"]]))

-            pdv <- data.frame(labelDescription=c("Signal-to-noise Ratio"),

-                              row.names=c("SNR"))

-            fd <- data.frame(SNPQC=x[["SNPQC"]],

-                             row.names=rownames(x[["calls"]]))

-            fdv <- data.frame(labelDescription=c("SNP Quality Score"),

-                              row.names=c("SNPQC"))

-            new("crlmmSet",

-                assayData=assayDataNew("lockedEnvironment",

-                  calls=x[["calls"]], confs=x[["confs"]]),

-                phenoData=new("AnnotatedDataFrame",

-                  data=pd, varMetadata=pdv),

-                featureData=new("AnnotatedDataFrame",

-                  data=fd, varMetadata=fdv))

-          })

new file mode 100644

@@ -0,0 +1,7 @@

+## Methods for crlmm

+

+setGeneric("calls", function(x) standardGeneric("calls"))

+setMethod("calls", "SnpSet", function(x) assayData(x)$call)

+

+setGeneric("confs", function(x) standardGeneric("confs"))

+setMethod("confs", "SnpSet", function(x) assayData(x)$callProbability)

@@ -42,3 +42,95 @@ getVarInEnv <- function(dataset, environ=.crlmmPkgEnv){

 		stop("Variable ", dataset, " not found in .crlmmPkgEnv")

 	environ[[dataset]]

 }

+

+list2SnpSet <- function(x, returnParams=FALSE){

+  pd <- data.frame(SNR=x[["SNR"]], gender=x[["gender"]],

+                   row.names=colnames(x[["calls"]]))

+  pdv <- data.frame(labelDescription=c("Signal-to-noise Ratio",

+                      "Gender: Male (1) and Female (2)"),

+                    row.names=c("SNR", "gender"))

+  recall <- length(x[["DD"]]) > 1

+  if (returnParams){

+    if (recall){

+      fd <- data.frame(SNPQC=x[["SNPQC"]],

+                       cAA=x[["params"]][["centers"]][,1],

+                       cAB=x[["params"]][["centers"]][,2],

+                       cBB=x[["params"]][["centers"]][,3],

+                       sAA=x[["params"]][["scales"]][,1],

+                       sAB=x[["params"]][["scales"]][,2],

+                       sBB=x[["params"]][["scales"]][,3],

+                       nAA=x[["params"]][["N"]][,1],

+                       nAB=x[["params"]][["N"]][,2],

+                       nBB=x[["params"]][["N"]][,3],

+                       spAA=x[["DD"]][,1],

+                       spAB=x[["DD"]][,2],

+                       spBB=x[["DD"]][,3],

+                       row.names=rownames(x[["calls"]]))

+      fdv <- data.frame(labelDescription=c(

+                          "SNP Quality Score",

+                          "Center AA", "Center AB", "Center BB",

+                          "Scale AA", "Scale AB", "Scale BB",

+                          "N AA", "N AB", "N BB",

+                          "Shift in parameters AA",

+                          "Shift in parameters AB",

+                          "Shift in parameters BB"),

+                        row.names=c(

+                          "SNPQC",

+                          "cAA", "cAB", "cBB",

+                          "sAA", "sAB", "sBB",

+                          "nAA", "nAB", "nBB",

+                          "spAA", "spAB", "spBB"))

+    }else{

+      fd <- data.frame(SNPQC=x[["SNPQC"]],

+                       cAA=x[["params"]][["centers"]][,1],

+                       cAB=x[["params"]][["centers"]][,2],

+                       cBB=x[["params"]][["centers"]][,3],

+                       sAA=x[["params"]][["scales"]][,1],

+                       sAB=x[["params"]][["scales"]][,2],

+                       sBB=x[["params"]][["scales"]][,3],

+                       nAA=x[["params"]][["N"]][,1],

+                       nAB=x[["params"]][["N"]][,2],

+                       nBB=x[["params"]][["N"]][,3],

+                       row.names=rownames(x[["calls"]]))

+      fdv <- data.frame(labelDescription=c(

+                          "SNP Quality Score",

+                          "Center AA", "Center AB", "Center BB",

+                          "Scale AA", "Scale AB", "Scale BB",

+                          "N AA", "N AB", "N BB",

+                          "Shift in parameters AA",

+                          "Shift in parameters AB",

+                          "Shift in parameters BB"),

+                        row.names=c(

+                          "SNPQC",

+                          "cAA", "cAB", "cBB",

+                          "sAA", "sAB", "sBB",

+                          "nAA", "nAB", "nBB"))

+    }

+  }else{

+    if (recall){

+      fd <- data.frame(SNPQC=x[["SNPQC"]],

+                       spAA=x[["DD"]][,1],

+                       spAB=x[["DD"]][,2],

+                       spBB=x[["DD"]][,3],

+                       row.names=rownames(x[["calls"]]))

+      fdv <- data.frame(labelDescription=c("SNP Quality Score",

+                          "Shift in parameters AA",

+                          "Shift in parameters AB",

+                          "Shift in parameters BB"),

+                        row.names=c("SNPQC", "spAA", "spAB", "spBB"))

+    }else{

+      fd <- data.frame(SNPQC=x[["SNPQC"]],

+                       row.names=rownames(x[["calls"]]))

+      fdv <- data.frame(labelDescription=c("SNP Quality Score"),

+                        row.names=c("SNPQC"))

+    }

+  }

+  new("SnpSet",

+      assayData=assayDataNew("lockedEnvironment",

+        call=x[["calls"]], callProbability=x[["confs"]]),

+      phenoData=new("AnnotatedDataFrame",

+        data=pd, varMetadata=pdv),

+      featureData=new("AnnotatedDataFrame",

+        data=fd, varMetadata=fdv),

+      annotation=x[["pkgname"]])

+}

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

 #####################################

 ### FOR CRLMM

 #####################################

-- Decide on output format (BC vote: eSet-like)

+- Fix downstream vignette

 - Add RS ids to annotation packages

 - Allele plots

 - M v S plots

@@ -14,7 +14,7 @@

 \begin{document}

 \title{Genotyping with the \Rpackage{crlmm} Package}

-\date{January, 2009}

+\date{March, 2009}

 \author{Benilton Carvalho}

 \maketitle

@@ -56,20 +56,21 @@ celFiles <- list.celfiles(path, full.names=TRUE)

 system.time(crlmmResult <- crlmm(celFiles, verbose=FALSE))

 @ 

-The \Robject{crlmmResult} is a list with the following components:

+The \Robject{crlmmResult} is a \Rclass{SnpSet} (see Biobase) object.

 \begin{itemize}

 \item \Robject{calls}: genotype calls (1 - AA; 2 - AB; 3 - BB);

 \item \Robject{confs}: confidence scores, which can be translated to probabilities by using:

   \[ 1-2^-(\mbox{confs}/1000), \] although we prefer this

   representation as it saves a significant amount of memory;

 \item \Robject{SNPQC}: SNP quality score;

-\item \Robject{batchQC}: Batch quality score;

+%%\item \Robject{batchQC}: Batch quality score;

 \item \Robject{SNR}: Signal-to-noise ratio.

 \end{itemize}

 <<out>>=

-crlmmResult[["calls"]][1:10,]

-crlmmResult[["confs"]][1:10,]

+calls(crlmmResult)[1:10,]

+confs(crlmmResult)[1:10,]

+crlmmResult[["SNR"]]

 @ 

 \section{Details}

@@ -10,8 +10,10 @@ Faster implementation of CRLMM specific to SNP 5.0 and 6.0 arrays.

 \details{

 Index:

 \preformatted{

-crlmm                   Genotype SNP 5.0 or 6.0 samples.

 crlmm-package           New implementation of the CRLMM Algorithm.

+crlmm                   Genotype SNP 5.0 or 6.0 samples.

+calls                   Accessor for genotype calls.

+confs                   Accessor for confidences.

 }

 The 'crlmm' package reimplements the CRLMM algorithm present in the

 'oligo' package. This implementation primes for efficient genotyping of

@@ -1,6 +1,10 @@

 \name{crlmm}

 \alias{crlmm}

-%- Also NEED an '\alias' for EACH other topic documented here.

+\alias{calls}

+\alias{calls,SnpSet-method}

+\alias{confs}

+\alias{confs,SnpSet-method}

+

 \title{Genotype oligonucleotide arrays with CRLMM}

 \description{

   This is a faster and more efficient implementation of the CRLMM

@@ -8,16 +12,18 @@

   be soon extended to other platforms).

 }

 \usage{

-

 crlmm(filenames, row.names=TRUE, col.names=TRUE,

       probs=c(1/3, 1/3, 1/3), DF=6, SNRMin=5,

       gender=NULL, save.it=FALSE, load.it=FALSE,

       intensityFile, mixtureSampleSize=10^5,

       eps=0.1, verbose=TRUE, cdfName, sns, recallMin=10,

       recallRegMin=1000, returnParams=FALSE, badSNP=0.7)

+calls(x)

+confs(x)

 }

-%- maybe also 'usage' for other objects documented here.

+

 \arguments{

+  \item{x}{'SnpSet' object. Usually the result of crlmm.}

   \item{filenames}{'character' vector with CEL files to be genotyped.}

   \item{row.names}{'logical'. Use rownames - SNP names?}

   \item{col.names}{'logical'. Use colnames - Sample names?}

@@ -43,6 +49,7 @@ crlmm(filenames, row.names=TRUE, col.names=TRUE,

   \item{badSNP}{'numeric'. Threshold to flag as bad SNP (affects batchQC)}

 }

 \value{

+  A \code{SnpSet} object.

   \item{calls}{Genotype calls (1 - AA, 2 - AB, 3 - BB)}

   \item{confs}{Confidence scores 'round(-1000*log2(1-p))'}

   \item{SNPQC}{SNP Quality Scores}

@@ -67,8 +74,8 @@ if (require(genomewidesnp5Crlmm) & require(hapmapsnp5)){

   ## very useful when genotyping samples not in the working directory

   cels <- list.celfiles(path, full.names=TRUE)

   crlmmOutput <- crlmm(cels)

-  crlmmOutput[["calls"]][1:10, 1:2]

-  crlmmOutput[["confs"]][1:10, 1:2]

+  (calls(crlmmOutput)[1:10, 1:2])

+  (confs(crlmmOutput)[1:10, 1:2])

 }

 }

 \keyword{classif}

deleted file mode 100644

@@ -1,58 +0,0 @@

-\name{crlmmSet-class}

-\Rdversion{1.1}

-\docType{class}

-\alias{crlmmSet-class}

-\alias{calls}

-\alias{calls,crlmmSet-method}

-\alias{confs}

-\alias{confs,crlmmSet-method}

-\alias{list2crlmmSet}

-\alias{list2crlmmSet,list-method}

-

-\title{Class "crlmmSet"}

-\description{An eSet extension to store genotype results.}

-\section{Objects from the Class}{

-  Objects can be created by calls of the form \code{new("crlmmSet",

-    assayData, phenoData, featureData, experimentData, annotation,

-    ...)}.

-}

-\section{Slots}{

-	 \describe{

-    \item{\code{assayData}:}{Object of class \code{"AssayData"}. For an

-      object to be valid, the assayData slot must have two matrices

-      named 'calls' and 'confs'.}

-    \item{\code{phenoData}:}{Object of class \code{"AnnotatedDataFrame"}}

-    \item{\code{featureData}:}{Object of class \code{"AnnotatedDataFrame"}}

-    \item{\code{experimentData}:}{Object of class \code{"MIAME"}}

-    \item{\code{annotation}:}{Object of class \code{"character"}}

-    \item{\code{.__classVersion__}:}{Object of class \code{"Versions"}}

-  }

-}

-\section{Extends}{

-Class \code{"\link[Biobase]{eSet-class}"}, directly.

-Class \code{"\link[Biobase]{VersionedBiobase-class}"}, by class "eSet", distance 2.

-Class \code{"\link[Biobase]{Versioned-class}"}, by class "eSet", distance 3.

-}

-\section{Methods}{

-  \describe{

-    \item{calls}{\code{signature(x = "crlmmSet")}: extractor for

-      genotype calls}

-    \item{confs}{\code{signature(x = "crlmmSet")}: extractor for confidences}

-	 }

-}

-\examples{

-if (require(genomewidesnp5Crlmm) & require(hapmapsnp5)){

-  path <- system.file("celFiles", package="hapmapsnp5")

-

-  ## the filenames with full path...

-  ## very useful when genotyping samples not in the working directory

-  cels <- list.celfiles(path, full.names=TRUE)

-  crlmmOutput <- crlmm(cels)

-  crlmmSetObject <- list2crlmmSet(crlmmOutput)

-  crlmmSetObject

-  calls(crlmmSetObject)[1:10, 1:2]

-  confs(crlmmSetObject)[1:10, 1:2]

-}

-}

-\keyword{classes}

-\keyword{methods}