From 65e377d29c35579a13f39f478aa44d9f3c3c8b9e Mon Sep 17 00:00:00 2001
From: Aaron <aaronquinlan@gmail.com>
Date: Tue, 29 Jan 2013 08:16:03 -0500
Subject: [PATCH] [EHN] add -delim option to bedtools merge
---
docs/content/tools/genomecov.rst | 2 +-
docs/content/tools/merge.rst | 39 ++++++++++++++++++++++++++-----
src/mergeBed/mergeBed.cpp | 14 +++++++----
src/mergeBed/mergeBed.h | 16 ++++++++++---
src/mergeBed/mergeMain.cpp | 36 ++++++++++++++++++++++++----
src/utils/VectorOps/VectorOps.cpp | 4 ++--
src/utils/VectorOps/VectorOps.h | 4 ++--
test/merge/test-merge.sh | 28 ++++++++++++++++++----
8 files changed, 114 insertions(+), 29 deletions(-)
diff --git a/docs/content/tools/genomecov.rst b/docs/content/tools/genomecov.rst
index c8e86566..e0c73a6e 100755
--- a/docs/content/tools/genomecov.rst
+++ b/docs/content/tools/genomecov.rst
@@ -35,7 +35,7 @@ Usage and option summary
**(or)**:
::
- genomeCovergaBed [OPTIONS] [-i|-ibam] -g (iff. -i)
+ genomeCoverageBed [OPTIONS] [-i|-ibam] -g (iff. -i)
diff --git a/docs/content/tools/merge.rst b/docs/content/tools/merge.rst
index 1b9eb767..c8c17a4a 100755
--- a/docs/content/tools/merge.rst
+++ b/docs/content/tools/merge.rst
@@ -45,12 +45,15 @@ Option Description
**-s** Force strandedness. That is, only merge features that are the same strand. *By default, this is disabled*.
**-n** Report the number of BED entries that were merged. *1 is reported if no merging occurred*.
**-d** Maximum distance between features allowed for features to be merged. *Default is 0. That is, overlapping and/or book-ended features are merged*.
-**-nms** Report the names of the merged features separated by semicolons.
+**-nms** Report the names of the merged features separated by commas. Change delimiter with ``-delim``
**-scores** | Report the scores of the merged features. Specify one of
- | the following options for reporting scores:
- | sum, min, max,
- | mean, median, mode, antimode,
- | collapse (i.e., print a semicolon-separated list)
+ | the following options for reporting scores:
+ | sum, min, max,
+ | mean, median, mode, antimode,
+ | collapse (i.e., print a semicolon-separated list)
+**-delim** | Specify a custom delimiter for the -nms and -scores concat options
+ | Example: ``-delim "|"``
+ | ``Default: ","``
=========================== ===============================================================================================================================================================================================================
@@ -156,7 +159,7 @@ merged features.
chr1 250 500 A3
$ bedtools merge -i A.bed -nms
- chr1 100 500 A1;A2;A3
+ chr1 100 500 A1,A2,A3
==========================================================================
@@ -183,5 +186,29 @@ how the scores from each overlapping interval should be reported.
chr1 100 500 1,2,3
+==========================================================================
+``-delim`` Change the delimiter for ``-nms`` and ``-scores collapse``
+==========================================================================
+One can override the use of a comma as the delimiter for the ``-nms`` and
+``-scores collapse`` options via the use of the ``-delim`` option.
+
+.. code-block:: bash
+
+ $ cat A.bed
+ chr1 100 200 A1
+ chr1 150 300 A2
+ chr1 250 500 A3
+
+Compare:
+
+.. code-block:: bash
+
+ $ bedtools merge -i A.bed -nms
+ chr1 100 500 A1,A2,A3
+to:
+
+.. code-block:: bash
+ $ bedtools merge -i A.bed -nms -delim "|"
+ chr1 100 500 A1|A2|A3
diff --git a/src/mergeBed/mergeBed.cpp b/src/mergeBed/mergeBed.cpp
index 0c1bb27d..e556e44b 100644
--- a/src/mergeBed/mergeBed.cpp
+++ b/src/mergeBed/mergeBed.cpp
@@ -21,7 +21,7 @@ void BedMerge::ReportMergedNames(const vector<string> &names) {
vector<string>::const_iterator nameEnd = names.end();
for (; nameItr != nameEnd; ++nameItr) {
if (nameItr < (nameEnd - 1))
- cout << *nameItr << ";";
+ cout << *nameItr << _delimiter;
else
cout << *nameItr;
}
@@ -29,7 +29,9 @@ void BedMerge::ReportMergedNames(const vector<string> &names) {
else {
cerr << endl
<< "*****" << endl
- << "*****ERROR: No names found to report for the -names option. Exiting." << endl
+ << "*****ERROR: "
+ << "No names found to report for the -names option. Exiting."
+ << endl
<< "*****" << endl;
exit(1);
}
@@ -58,7 +60,7 @@ void BedMerge::ReportMergedScores(const vector<string> &scores) {
else if (_scoreOp == "antimode")
buffer << setprecision (PRECISION) << vo.GetAntiMode();
else if (_scoreOp == "collapse")
- buffer << setprecision (PRECISION) << vo.GetCollapse();
+ buffer << setprecision (PRECISION) << vo.GetCollapse(_delimiter);
cout << "\t" << buffer.str();
}
else {
@@ -79,14 +81,16 @@ BedMerge::BedMerge(string &bedFile,
bool forceStrand,
bool reportNames,
bool reportScores,
- const string &scoreOp) :
+ const string &scoreOp,
+ const string &delimiter) :
_bedFile(bedFile),
_numEntries(numEntries),
_forceStrand(forceStrand),
_reportNames(reportNames),
_reportScores(reportScores),
_scoreOp(scoreOp),
- _maxDistance(maxDistance)
+ _maxDistance(maxDistance),
+ _delimiter(delimiter)
{
_bed = new BedFile(bedFile);
diff --git a/src/mergeBed/mergeBed.h b/src/mergeBed/mergeBed.h
index 52eb7cb1..d9b0c143 100644
--- a/src/mergeBed/mergeBed.h
+++ b/src/mergeBed/mergeBed.h
@@ -34,7 +34,8 @@ public:
// constructor
BedMerge(string &bedFile, bool numEntries,
int maxDistance, bool forceStrand,
- bool reportNames, bool reportScores, const string &scoreOp);
+ bool reportNames, bool reportScores,
+ const string &scoreOp, const string &delimiter);
// destructor
~BedMerge(void);
@@ -51,11 +52,20 @@ private:
bool _reportScores;
string _scoreOp;
int _maxDistance;
+ string _delimiter;
// instance of a bed file class.
BedFile *_bed;
- void Report(string chrom, int start, int end, const vector<string> &names, const vector<string> &scores, int mergeCount);
- void ReportStranded(string chrom, int start, int end, const vector<string> &names, const vector<string> &scores, int mergeCount, string strand);
+ void Report(string chrom, int start, int end,
+ const vector<string> &names,
+ const vector<string> &scores,
+ int mergeCount);
+
+ void ReportStranded(string chrom, int start, int end,
+ const vector<string> &names,
+ const vector<string> &scores,
+ int mergeCount,
+ string strand);
void ReportMergedNames(const vector<string> &names);
void ReportMergedScores(const vector<string> &scores);
diff --git a/src/mergeBed/mergeMain.cpp b/src/mergeBed/mergeMain.cpp
index 2f1a164c..28b869af 100644
--- a/src/mergeBed/mergeMain.cpp
+++ b/src/mergeBed/mergeMain.cpp
@@ -41,6 +41,7 @@ int merge_main(int argc, char* argv[]) {
bool forceStrand = false;
bool reportNames = false;
bool reportScores = false;
+ string delimiter = ",";
for(int i = 1; i < argc; i++) {
int parameterLength = (int)strlen(argv[i]);
@@ -87,6 +88,12 @@ int merge_main(int argc, char* argv[]) {
i++;
}
}
+ else if (PARAMETER_CHECK("-delim", 6, parameterLength)) {
+ if ((i+1) < argc) {
+ delimiter = argv[i + 1];
+ i++;
+ }
+ }
else {
cerr << endl << "*****ERROR: Unrecognized parameter: " << argv[i] << " *****" << endl << endl;
showHelp = true;
@@ -98,15 +105,28 @@ int merge_main(int argc, char* argv[]) {
cerr << endl << "*****" << endl << "*****ERROR: Need -i BED file. " << endl << "*****" << endl;
showHelp = true;
}
- if ((reportScores == true) && (scoreOp != "sum") && (scoreOp != "max") && (scoreOp != "min") && (scoreOp != "mean") &&
- (scoreOp != "mode") && (scoreOp != "median") && (scoreOp != "antimode") && (scoreOp != "collapse"))
+ if ((reportScores == true) && (scoreOp != "sum")
+ && (scoreOp != "max") && (scoreOp != "min")
+ && (scoreOp != "mean") && (scoreOp != "mode")
+ && (scoreOp != "median") && (scoreOp != "antimode")
+ && (scoreOp != "collapse"))
{
- cerr << endl << "*****" << endl << "*****ERROR: Invalid scoreOp selection \"" << scoreOp << endl << "\" *****" << endl;
+ cerr << endl
+ << "*****"
+ << endl
+ << "*****ERROR: Invalid scoreOp selection \""
+ << scoreOp
+ << endl
+ << "\" *****"
+ << endl;
showHelp = true;
}
if (!showHelp) {
- BedMerge *bm = new BedMerge(bedFile, numEntries, maxDistance, forceStrand, reportNames, reportScores, scoreOp);
+ BedMerge *bm = new BedMerge(bedFile, numEntries,
+ maxDistance, forceStrand,
+ reportNames, reportScores,
+ scoreOp, delimiter);
delete bm;
}
else {
@@ -137,7 +157,8 @@ void merge_help(void) {
cerr << "\t\t- Def. 0. That is, overlapping & book-ended features are merged." << endl;
cerr << "\t\t- (INTEGER)" << endl << endl;
- cerr << "\t-nms\t" << "Report the names of the merged features separated by semicolons." << endl << endl;
+ cerr << "\t-nms\t" << "Report the names of the merged features separated by commas." << endl;
+ cerr << "\t\tChange delim. with -delim." << endl << endl;
cerr << "\t-scores\t" << "Report the scores of the merged features. Specify one of " << endl;
cerr << "\t\tthe following options for reporting scores:" << endl;
@@ -146,6 +167,11 @@ void merge_help(void) {
cerr << "\t\t collapse (i.e., print a semicolon-separated list)," << endl;
cerr << "\t\t- (INTEGER)" << endl << endl;
+ cerr << "\t-delim\t" << "Specify a custom delimiter for the -nms and -scores concat options" << endl;
+ cerr << "\t\t- Example: -delim \"|\"" << endl;
+ cerr << "\t\t- Default: \",\"." << endl << endl;
+
+
cerr << "Notes: " << endl;
cerr << "\t(1) All output, regardless of input type (e.g., GFF or VCF)" << endl;
cerr << "\t will in BED format with zero-based starts" << endl << endl;
diff --git a/src/utils/VectorOps/VectorOps.cpp b/src/utils/VectorOps/VectorOps.cpp
index 53c2c374..97c5d629 100644
--- a/src/utils/VectorOps/VectorOps.cpp
+++ b/src/utils/VectorOps/VectorOps.cpp
@@ -208,12 +208,12 @@ uint32_t VectorOps::GetCountDistinct(void)
return _vecs.size();
}
-string VectorOps::GetCollapse(void)
+string VectorOps::GetCollapse(string delimiter)
{
ostringstream collapse;
for( size_t i = 0; i < _vecs.size(); i++ ) {
if (i>0)
- collapse << ",";
+ collapse << delimiter;
collapse << _vecs[i];
}
return collapse.str();
diff --git a/src/utils/VectorOps/VectorOps.h b/src/utils/VectorOps/VectorOps.h
index 3f4b53c9..5e979fa9 100644
--- a/src/utils/VectorOps/VectorOps.h
+++ b/src/utils/VectorOps/VectorOps.h
@@ -63,8 +63,8 @@ public:
uint32_t GetCount(void);
// return a the count of _unique_ elements in the vector
uint32_t GetCountDistinct(void);
- // return a comma-separated list of elements
- string GetCollapse(void);
+ // return a delimiter-separated list of elements
+ string GetCollapse(string delimiter = ",");
// return a concatenation of all elements in the vector
string GetConcat(void);
// return a comma-separated list of the _unique_ elements
diff --git a/test/merge/test-merge.sh b/test/merge/test-merge.sh
index b616be0a..ee7f6290 100644
--- a/test/merge/test-merge.sh
+++ b/test/merge/test-merge.sh
@@ -87,7 +87,7 @@ rm obs exp
echo " merge.t5...\c"
echo \
"chr1 10 20 a1
-chr1 30 100 a2;a3;a4" > exp
+chr1 30 100 a2,a3,a4" > exp
$BT merge -i a.names.bed -nms > obs
check obs exp
rm obs exp
@@ -98,10 +98,10 @@ rm obs exp
echo " merge.t6...\c"
echo \
"chr1 10 20 a1 1
-chr1 30 100 a2;a3;a4 9
+chr1 30 100 a2,a3,a4 9
chr2 10 20 a1 5
chr2 30 40 a2 6
-chr2 42 100 a3;a4 15" > exp
+chr2 42 100 a3,a4 15" > exp
$BT merge -i a.full.bed -nms -scores sum> obs
check obs exp
rm obs exp
@@ -126,10 +126,10 @@ rm obs exp
echo " merge.t8...\c"
echo \
"chr1 10 20 a1 1 1
-chr1 30 100 a2;a3;a4 9 3
+chr1 30 100 a2,a3,a4 9 3
chr2 10 20 a1 5 1
chr2 30 40 a2 6 1
-chr2 42 100 a3;a4 15 2" > exp
+chr2 42 100 a3,a4 15 2" > exp
$BT merge -i a.full.bed -nms -n -scores sum> obs
check obs exp
rm obs exp
@@ -150,3 +150,21 @@ chr2 45 100 a4 8 - 1" > exp
$BT merge -i a.full.bed -s -nms -n -scores sum> obs
check obs exp
rm obs exp
+
+###########################################################
+# Test #10
+# Test the use of a custom delimiter for -nms
+#
+# cat a.names.bed
+# chr1 10 20 a1
+# chr1 30 40 a2
+# chr1 40 50 a3
+# chr1 45 100 a4
+###########################################################
+echo " merge.t10...\c"
+echo \
+"chr1 10 20 a1
+chr1 30 100 a2|a3|a4" > exp
+$BT merge -i a.names.bed -nms -delim "|" > obs
+check obs exp
+rm obs exp
--
GitLab