From 4d4d33404e77532ec625b25428d477cf7b7e5e05 Mon Sep 17 00:00:00 2001
From: Phillip Dexheimer <phildex@broadinstitute.org>
Date: Wed, 4 Feb 2015 23:02:54 -0500
Subject: [PATCH] Added gsa.reshape.concordance.table function to gsalib

---
 .../variantutils/GenotypeConcordance.java     |  2 +-
 public/gsalib/src/R/DESCRIPTION               |  4 +-
 public/gsalib/src/R/NAMESPACE                 |  3 +-
 .../src/R/R/gsa.reshape.concordance.table.R   | 20 ++++++++
 .../R/man/gsa.reshape.concordance.table.Rd    | 48 +++++++++++++++++++
 public/gsalib/src/R/man/gsalib-package.Rd     |  4 +-
 6 files changed, 75 insertions(+), 6 deletions(-)
 create mode 100644 public/gsalib/src/R/R/gsa.reshape.concordance.table.R
 create mode 100644 public/gsalib/src/R/man/gsa.reshape.concordance.table.Rd

diff --git a/public/gatk-tools-public/src/main/java/org/broadinstitute/gatk/tools/walkers/variantutils/GenotypeConcordance.java b/public/gatk-tools-public/src/main/java/org/broadinstitute/gatk/tools/walkers/variantutils/GenotypeConcordance.java
index d87c7ec71..a9e578058 100644
--- a/public/gatk-tools-public/src/main/java/org/broadinstitute/gatk/tools/walkers/variantutils/GenotypeConcordance.java
+++ b/public/gatk-tools-public/src/main/java/org/broadinstitute/gatk/tools/walkers/variantutils/GenotypeConcordance.java
@@ -82,7 +82,7 @@ import java.util.*;
  *  <p>
  *  It may be informative to reshape rows of the GenotypeConcordance counts and proportions tables into separate row-major tables
  *  where the columns indicate the COMP genotype and the rows indicate the EVAL genotype for easy comparison between the
- *  two callsets. This can be done with a command similar to d <- matrix(sampleRow,nrow=6,byrow=T) in R where sampleRow is the 36-value row corresponding to the sample of interest, excluding "Mismatching_Alleles".
+ *  two callsets. This can be done with the gsa.reshape.concordance.table function in the gsalib R library.
  *  In Excel this can be accomplished using the OFFSET function.
  *  </p>
  *  <ul>
diff --git a/public/gsalib/src/R/DESCRIPTION b/public/gsalib/src/R/DESCRIPTION
index ff535e221..a3736d0d2 100644
--- a/public/gsalib/src/R/DESCRIPTION
+++ b/public/gsalib/src/R/DESCRIPTION
@@ -1,8 +1,8 @@
 Package: gsalib
 Type: Package
 Title: Utility Functions For GATK
-Version: 2.1
-Date: 2014-12-09
+Version: 2.2
+Date: 2015-03-17
 Author: Kiran Garimella
 Maintainer: Geraldine Van der Auwera <vdauwera@broadinstitute.org>
 Description: This package contains utility functions used by the Genome Analysis Toolkit (GATK) to load tables and plot data. The GATK is a toolkit for variant discovery in high-throughput sequencing data.
diff --git a/public/gsalib/src/R/NAMESPACE b/public/gsalib/src/R/NAMESPACE
index 73b91cfa8..a976c2089 100644
--- a/public/gsalib/src/R/NAMESPACE
+++ b/public/gsalib/src/R/NAMESPACE
@@ -1 +1,2 @@
-export(gsa.read.gatkreport)
\ No newline at end of file
+export(gsa.read.gatkreport)
+export(gsa.reshape.concordance.table)
\ No newline at end of file
diff --git a/public/gsalib/src/R/R/gsa.reshape.concordance.table.R b/public/gsalib/src/R/R/gsa.reshape.concordance.table.R
new file mode 100644
index 000000000..46c3a9e93
--- /dev/null
+++ b/public/gsalib/src/R/R/gsa.reshape.concordance.table.R
@@ -0,0 +1,20 @@
+gsa.reshape.concordance.table <- function(data, table.name="GenotypeConcordance_Counts", sample.name="ALL") {
+  if (!is.null(table.name)) {
+    data <- data[[table.name]]
+  }
+  if (is.null(data)) {
+    return NULL
+  }
+  d <- data[data$Sample==sample.name,2:(length(data[1,])-1)]
+  
+  possible.genotypes <- c('NO_CALL', 'HOM_REF', 'HET', 'HOM_VAR', 'UNAVAILABLE', 'MIXED')
+  combinations <- outer(possible.genotypes, possible.genotypes, function(a,b) {paste(a,b,sep='_')})
+  existing.combi <- matrix(combinations %in% colnames(d), nrow=length(possible.genotypes))
+  eval.genotypes <- apply(existing.combi, 1, any)
+  comp.genotypes <- apply(existing.combi, 2, any)
+  
+  m <- matrix(d, nrow=sum(eval.genotypes), byrow=T)
+  dimnames(m) <- list(EvalGenotypes=possible.genotypes[eval.genotypes],
+                      CompGenotypes=possible.genotypes[comp.genotypes])
+  m
+}
diff --git a/public/gsalib/src/R/man/gsa.reshape.concordance.table.Rd b/public/gsalib/src/R/man/gsa.reshape.concordance.table.Rd
new file mode 100644
index 000000000..f3e85047a
--- /dev/null
+++ b/public/gsalib/src/R/man/gsa.reshape.concordance.table.Rd
@@ -0,0 +1,48 @@
+\name{gsa.reshape.concordance.table}
+\alias{gsa.reshape.concordance.table}
+\title{
+Reshape a Concordance Table
+}
+\description{
+Given a GATKReport generated by GenotypeConcordance (as output by \code{gsa.read.gatkreport}), this function reshapes the concordance for a specified sample into a matrix with the EvalGenotypes in rows and the CompGenotypes in columns (see the documentation for GenotypeConcordance for the definition of Eval and Comp)
+}
+\usage{
+gsa.reshape.concordance.table(x, table="GenotypeConcordance_Counts", sample.name="ALL")
+}
+\arguments{
+  \item{x}{
+A GATKReport as output by \code{gsa.read.gatkreport}.  If \code{table} is \code{NULL}, \code{x} is assumed to be the vector of concordance values to reshape.
+}
+  \item{table}{
+The table name in the GATKReport to reshape.  Defaults to "GenotypeConcordance_Counts", but could also be one of the proportion tables ("GenotypeConcordance_EvalProportions", "GenotypeConcordance_CompProportions").  This value can also be \code{NULL}, in which case \code{x} is reshaped directly.  
+}
+  \item{sample.name}{
+The sample name within \code{table} to use.
+}
+}
+\value{
+Returns a two-dimensional matrix with Eval genotypes in the rows and Comp genotypes in the columns.  The genotypes themselves (\code{HOM_REF}, \code{NO_CALL}, etc) are specified in the row/col names of the matrix.
+}
+\author{
+Phillip Dexheimer
+}
+
+\seealso{
+\code{\link{gsa.read.gatkreport}}
+}
+\examples{
+test_file = system.file("inst", "extdata", "test_gatkreport.table", package = "gsalib")
+report = gsa.read.gatkreport(test_file)
+gsa.reshape.concordance.table(report)
+
+## Output looks like:
+##             CompGenotypes
+##EvalGenotypes NO_CALL HOM_REF HET HOM_VAR UNAVAILABLE MIXED
+##  NO_CALL     0       0       0   0       0           0    
+##  HOM_REF     0       2       0   0       0           0    
+##  HET         0       3       0   0       0           0    
+##  HOM_VAR     0       2       0   0       0           0    
+##  UNAVAILABLE 0       0       0   0       0           0    
+##  MIXED       0       0       0   0       0           0    
+}
+\keyword{ manip }
diff --git a/public/gsalib/src/R/man/gsalib-package.Rd b/public/gsalib/src/R/man/gsalib-package.Rd
index c758be138..f1e0e2ac4 100644
--- a/public/gsalib/src/R/man/gsalib-package.Rd
+++ b/public/gsalib/src/R/man/gsalib-package.Rd
@@ -12,8 +12,8 @@ Utility functions for analysis of genome sequence data with the GATK
 \tabular{ll}{
 Package: \tab gsalib\cr
 Type: \tab Package\cr
-Version: \tab 2.1\cr
-Date: \tab 2014-12-09\cr
+Version: \tab 2.2\cr
+Date: \tab 2015-03-17\cr
 License: \tab MIT\cr
 LazyLoad: \tab yes\cr
 }