// Marth Lab, Department of Biology, Boston College
// All rights reserved.
// ---------------------------------------------------------------------------
-// Last modified: 23 March 2011
+// Last modified: 7 April 2011
// ---------------------------------------------------------------------------
// Prints alignment count for BAM file(s)
// ***************************************************************************
#include "bamtools_count.h"
+#include <api/BamAlgorithms.h>
#include <api/BamMultiReader.h>
#include <utils/bamtools_options.h>
#include <utils/bamtools_utilities.h>
};
// ---------------------------------------------
-// CountTool implementation
+// CountToolPrivate implementation
-CountTool::CountTool(void)
- : AbstractTool()
- , m_settings(new CountSettings)
-{
- // set program details
- Options::SetProgramInfo("bamtools count", "prints number of alignments in BAM file(s)", "[-in <filename> -in <filename> ...] [-region <REGION>]");
-
- // set up options
- OptionGroup* IO_Opts = Options::CreateOptionGroup("Input & Output");
- Options::AddValueOption("-in", "BAM filename", "the input BAM file(s)", "", m_settings->HasInput, m_settings->InputFiles, IO_Opts, Options::StandardIn());
- Options::AddValueOption("-region", "REGION", "genomic region. Index file is recommended for better performance, and is used automatically if it exists. See \'bamtools help index\' for more details on creating one", "", m_settings->HasRegion, m_settings->Region, IO_Opts);
-}
+struct CountTool::CountToolPrivate {
-CountTool::~CountTool(void) {
- delete m_settings;
- m_settings = 0;
-}
+ // ctor & dtro
+ public:
+ CountToolPrivate(CountTool::CountSettings* settings)
+ : m_settings(settings)
+ { }
-int CountTool::Help(void) {
- Options::DisplayHelp();
- return 0;
-}
+ ~CountToolPrivate(void) { }
-int CountTool::Run(int argc, char* argv[]) {
+ // interface
+ public:
+ bool Run(void);
- // parse command line arguments
- Options::Parse(argc, argv, 1);
+ // data members
+ private:
+ CountTool::CountSettings* m_settings;
+};
+
+static
+void printAlignments(const vector<BamAlignment>& alignments) {
+
+ vector<BamAlignment>::const_iterator alIter = alignments.begin();
+ vector<BamAlignment>::const_iterator alEnd = alignments.end();
+ for ( ; alIter != alEnd; ++alIter ) {
+ const BamAlignment& a = (*alIter);
+
+ cerr << a.Name
+ << "\t" << a.RefID << ":" << a.Position;
+
+ int aqValue;
+ bool hasTag = a.GetTag("Aq", aqValue);
+ cerr << "\tAq=";
+ if ( hasTag ) cerr << aqValue;
+ else cerr << "?";
+ cerr << endl;
+ }
+}
+
+bool CountTool::CountToolPrivate::Run(void) {
// if no '-in' args supplied, default to stdin
- if ( !m_settings->HasInput )
+ if ( !m_settings->HasInput )
m_settings->InputFiles.push_back(Options::StandardIn());
-
+
// open reader without index
BamMultiReader reader;
if ( !reader.Open(m_settings->InputFiles) ) {
cerr << "bamtools count ERROR: could not open input BAM file(s)... Aborting." << endl;
- return 1;
+ return false;
}
// alignment counter
BamAlignment al;
int alignmentCount(0);
-
- // if no region specified, count entire file
+
+ // if no region specified, count entire file
if ( !m_settings->HasRegion ) {
- while ( reader.GetNextAlignmentCore(al) )
+
+
+ vector<BamAlignment> alignments;
+ while ( reader.GetNextAlignment(al) ) {
++alignmentCount;
+
+ if ( alignments.size() < 100 )
+ alignments.push_back(al);
+ }
+
+ using namespace BamTools::Algorithms;
+
+ cerr << endl
+ << "------------------------------" << endl
+ << "Unsorted Alignments" << endl
+ << "------------------------------" << endl
+ << endl;
+ std::stable_sort(alignments.begin(), alignments.end(), Sort::Unsorted());
+ printAlignments(alignments);
+ cerr << "------------------------------" << endl
+ << endl;
+
+ cerr << endl
+ << "------------------------------" << endl
+ << "Sorted Alignments (by name)" << endl
+ << "------------------------------" << endl
+ << endl;
+// std::sort(alignments.begin(), alignments.end(), Sort::ByName());
+ Algorithms::SortAlignments(alignments, Sort::ByName());
+ printAlignments(alignments);
+ cerr << endl
+ << "------------------------------" << endl
+ << endl;
+
+ cerr << endl
+ << "------------------------------" << endl
+ << "Sorted Alignments (by tag Aq)" << endl
+ << "------------------------------" << endl
+ << endl;
+// std::sort(alignments.begin(), alignments.end(), Sort::ByTag<int>("Aq"));
+ Algorithms::SortAlignments(alignments, Sort::ByTag<int>("Aq"));
+ printAlignments(alignments);
+ cerr << endl
+ << "------------------------------" << endl
+ << endl;
+
+ cerr << endl
+ << "------------------------------" << endl
+ << "Sorted Alignments (by tag Aq) desc" << endl
+ << "------------------------------" << endl
+ << endl;
+ std::sort(alignments.begin(), alignments.end(), Sort::ByTag<int>("Aq", Sort::DescendingOrder));
+ printAlignments(alignments);
+ cerr << endl
+ << "------------------------------" << endl
+ << endl;
+
+
+
+
+// // ########################################
+// // original
+// // ########################################
+//
+// while ( reader.GetNextAlignmentCore(al) )
+// ++alignmentCount;
+//
+// //#########################################
}
-
+
// otherwise attempt to use region as constraint
else {
-
+
// if region string parses OK
BamRegion region;
if ( Utilities::ParseRegionString(m_settings->Region, reader, region) ) {
reader.LocateIndexes();
// if index data available for all BAM files, we can use SetRegion
- if ( reader.IsIndexLoaded() ) {
-
- // attempt to set region on reader
- if ( !reader.SetRegion(region.LeftRefID, region.LeftPosition, region.RightRefID, region.RightPosition) ) {
- cerr << "bamtools count ERROR: set region failed. Check that REGION describes a valid range" << endl;
- reader.Close();
- return 1;
- }
-
- // everything checks out, just iterate through specified region, counting alignments
- while ( reader.GetNextAlignmentCore(al) )
- ++alignmentCount;
- }
-
+ if ( reader.HasIndexes() ) {
+
+ vector<BamAlignment> alignments;
+ using namespace BamTools::Algorithms;
+
+ alignments = GetSortedRegion(reader, region, Sort::ByName() );
+ printAlignments(alignments);
+
+ cerr << "################################" << endl;
+
+ alignments = GetSortedRegion(reader, region, Sort::ByTag<int>("Aq"));
+ printAlignments(alignments);
+
+// // attempt to set region on reader
+// if ( !reader.SetRegion(region.LeftRefID, region.LeftPosition, region.RightRefID, region.RightPosition) ) {
+// cerr << "bamtools count ERROR: set region failed. Check that REGION describes a valid range" << endl;
+// reader.Close();
+// return false;
+// }
+
+// // everything checks out, just iterate through specified region, counting alignments
+// while ( reader.GetNextAlignmentCore(al) )
+// ++alignmentCount;
+ }
+
// no index data available, we have to iterate through until we
// find overlapping alignments
else {
while ( reader.GetNextAlignmentCore(al) ) {
if ( (al.RefID >= region.LeftRefID) && ( (al.Position + al.Length) >= region.LeftPosition ) &&
- (al.RefID <= region.RightRefID) && ( al.Position <= region.RightPosition) )
+ (al.RefID <= region.RightRefID) && ( al.Position <= region.RightPosition) )
{
++alignmentCount;
}
}
}
- }
-
+ }
+
// error parsing REGION string
else {
cerr << "bamtools count ERROR: could not parse REGION - " << m_settings->Region << endl;
cerr << "Check that REGION is in valid format (see documentation) and that the coordinates are valid"
<< endl;
reader.Close();
- return 1;
+ return false;
}
}
-
- // print results
+
+ // print results
cout << alignmentCount << endl;
-
+
// clean up & exit
reader.Close();
+ return true;
+}
+
+// ---------------------------------------------
+// CountTool implementation
+
+CountTool::CountTool(void)
+ : AbstractTool()
+ , m_settings(new CountSettings)
+ , m_impl(0)
+{
+ // set program details
+ Options::SetProgramInfo("bamtools count", "prints number of alignments in BAM file(s)", "[-in <filename> -in <filename> ...] [-region <REGION>]");
+
+ // set up options
+ OptionGroup* IO_Opts = Options::CreateOptionGroup("Input & Output");
+ Options::AddValueOption("-in", "BAM filename", "the input BAM file(s)", "", m_settings->HasInput, m_settings->InputFiles, IO_Opts, Options::StandardIn());
+ Options::AddValueOption("-region", "REGION", "genomic region. Index file is recommended for better performance, and is used automatically if it exists. See \'bamtools help index\' for more details on creating one", "", m_settings->HasRegion, m_settings->Region, IO_Opts);
+}
+
+CountTool::~CountTool(void) {
+
+ delete m_settings;
+ m_settings = 0;
+
+ delete m_impl;
+ m_impl = 0;
+}
+
+int CountTool::Help(void) {
+ Options::DisplayHelp();
return 0;
+}
+
+int CountTool::Run(int argc, char* argv[]) {
+
+ // parse command line arguments
+ Options::Parse(argc, argv, 1);
+
+ // initialize CountTool with settings
+ m_impl = new CountToolPrivate(m_settings);
+
+ // run CountTool, return success/fail
+ if ( m_impl->Run() )
+ return 0;
+ else
+ return 1;
}