Updated EBSeq from v1.1.5 to v1.1.6 and fixed a bug in 'rsem-generate-data-matrix...

diff --git a/EBSeq/EBSeq_1.1.5.tar.gz b/EBSeq/EBSeq_1.1.5.tar.gz
deleted file mode 100644 (file)
index e0bf8de..0000000
Binary files a/EBSeq/EBSeq_1.1.5.tar.gz and /dev/null differ

diff --git a/EBSeq/EBSeq_1.1.6.tar.gz b/EBSeq/EBSeq_1.1.6.tar.gz
new file mode 100644 (file)
index 0000000..f2b0d5e
Binary files /dev/null and b/EBSeq/EBSeq_1.1.6.tar.gz differ

diff --git a/EBSeq/makefile b/EBSeq/makefile
index 020a32a292fb3ed2077cf388cba5bdcbf7ca92cf..ac97910487878543d80f7b1ec2a106d30acb2605 100644 (file)
--- a/EBSeq/makefile
+++ b/EBSeq/makefile
@@ -6,8 +6,8 @@ all : $(PROGRAMS)
 blockmodeling : blockmodeling_0.1.8.tar.gz
        R CMD INSTALL -l "." blockmodeling_0.1.8.tar.gz
 
-EBSeq : blockmodeling EBSeq_1.1.5.tar.gz
-       R CMD INSTALL -l "." EBSeq_1.1.5.tar.gz
+EBSeq : blockmodeling EBSeq_1.1.6.tar.gz
+       R CMD INSTALL -l "." EBSeq_1.1.6.tar.gz
 
 rsem-for-ebseq-calculate-clustering-info : calcClusteringInfo.cpp
        $(CC) -O3 -Wall calcClusteringInfo.cpp -o $@

diff --git a/README.md b/README.md
index 087493d515369c29d28063f72a4134e2d4a0762d..7b9a4b83ca83a7ec87926af31cab8bf292a29e8c 100644 (file)
--- a/README.md
+++ b/README.md
@@ -375,13 +375,13 @@ you.
 
 Usage: 
 
-    rsem-find-DE data_matrix_file [--ngvector ngvector_file] number_sample_condition1 FDR_rate output_file
+    rsem-find-DE data_matrix_file [--ngvector ngvector_file] number_of_samples_in_condition_1 FDR_rate output_file
 
 This script calls EBSeq to find differentially expressed genes/transcripts in two conditions.
 
 data_matrix_file: m by n matrix containing expected counts, m is the number of transcripts/genes, n is the number of total samples.   
 [--ngvector ngvector_file]: optional field. 'ngvector_file' is calculated by 'rsem-generate-ngvector'. Having this field is recommended for transcript data.   
-number_sample_condition1: the number of samples in condition 1. A condition's samples must be adjacent. The left group of samples are defined as condition 1.   
+number_of_samples_in_condition_1: the number of samples in condition 1. A condition's samples must be adjacent. The left group of samples are defined as condition 1.   
 FDR_rate: false discovery rate.   
 output_file: the output file. Three files will be generated: 'output_file', 'output_file.hard_threshold' and 'output_file.all'. The first file reports all DE genes/transcripts using a soft threshold (calculated by crit_func in EBSeq). The second file reports all DE genes/transcripts using a hard threshold (only report if PPEE <= fdr). The third file reports all genes/transcripts. The first file is recommended to be used as DE results because it generally contains more called genes/transcripts.   
 
@@ -414,6 +414,7 @@ RSEM uses the [Boost C++](http://www.boost.org) and
 differential expression analysis.
 
 We thank earonesty for contributing patches.
+
 We thank Han Lin for suggesting possible fixes. 
 
 ## <a name="license"></a> License

diff --git a/WHAT_IS_NEW b/WHAT_IS_NEW
index 28108dcfac60eb05c39224ee73e09b58b80d3763..86e21949e7f52f0e0b8e0356393b3e1cfa001d80 100644 (file)
--- a/WHAT_IS_NEW
+++ b/WHAT_IS_NEW
@@ -1,3 +1,10 @@
+RSEM v1.2.5
+
+- Updated EBSeq from v1.1.5 to v1.1.6
+- Fixed a bug in 'rsem-generate-data-matrix', which can cause 'rsem-find-DE' to crash 
+
+--------------------------------------------------------------------------------------------
+
 RSEM v1.2.4
 
 - Fixed a bug that leads to poor parallelization performance in Mac OS systems

diff --git a/rsem-find-DE b/rsem-find-DE
index e9d65cfbbd5fd0cb1a1446e0de1240b81e6339e1..a69ebd427e67d1381691e6fde9b1ff552326bb75 100755 (executable)
--- a/rsem-find-DE
+++ b/rsem-find-DE
@@ -1,11 +1,11 @@
 #!/usr/bin/env Rscript
 
 printUsage <- function() {
-  cat("Usage: rsem-find-DE data_matrix_file [--ngvector ngvector_file] number_sample_condition1 FDR_rate output_file\n\n")
+  cat("Usage: rsem-find-DE data_matrix_file [--ngvector ngvector_file] number_of_samples_in_condition_1 FDR_rate output_file\n\n")
   cat("This script calls EBSeq to find differentially expressed genes/transcripts in two conditions.\n\n") 
   cat("data_matrix_file: m by n matrix containing expected counts, m is the number of transcripts/genes, n is the number of total samples.\n")
   cat("[--ngvector ngvector_file]: optional field. 'ngvector_file' is calculated by 'rsem-generate-ngvector'. Having this field is recommended for transcript data.\n")
-  cat("number_sample_condition1: the number of samples in condition 1. A condition's samples must be adjacent. The left group of samples are defined as condition 1.\n")
+  cat("number_of_samples_in_condition_1: the number of samples in condition 1. A condition's samples must be adjacent. The left group of samples are defined as condition 1.\n")
   cat("FDR_rate: false discovery rate.\n")
   cat("output_file: the output file. Three files will be generated: 'output_file', 'output_file.hard_threshold' and 'output_file.all'. The first file reports all DE genes/transcripts using a soft threshold (calculated by crit_func in EBSeq). The second file reports all DE genes/transcripts using a hard threshold (only report if PPEE <= fdr). The third file reports all genes/transcripts. The first file is recommended to be used as DE results because it generally contains more called genes/transcripts.\n\n")
   cat("The results are written as a matrix with row and column names. The row names are the genes'/transcripts' ids. The column names are 'PPEE', 'PPDE', 'PostFC' and 'RealFC'.\n\n")

diff --git a/rsem-generate-data-matrix b/rsem-generate-data-matrix
index 951f415df24a8de4e5efd1e594daa11608abe87c..b1b4f63e9713babf3730565261f90c6738fdcdfe 100755 (executable)
--- a/rsem-generate-data-matrix
+++ b/rsem-generate-data-matrix
@@ -22,7 +22,7 @@ sub loadData {
     while ($line = <INPUT>) {
        chomp($line); 
        my @fields = split(/\t/, $line);
-       push(@{$_[2]}, $fields[0]);
+       push(@{$_[2]}, "\"$fields[0]\"");
        push(@{$_[1]}, $fields[$offsite]);
     }
     close(INPUT);
@@ -59,7 +59,11 @@ for (my $i = 0; $i < $n; $i++) {
        exit(-1); 
     }
 
-    @ecs = ($ARGV[$i], @ecs);
+    my $colname;
+    if (substr($ARGV[$i], 0, 2) eq "./") { $colname = substr($ARGV[$i], 2); }
+    else { $colname = $ARGV[$i]; }
+    $colname = "\"$colname\"";
+    @ecs = ($colname, @ecs);
     push(@matrix, \@ecs);
 }