#include "bamtools_options.h"
#include "bamtools_utilities.h"
#include "BamReader.h"
+#include "BamMultiReader.h"
using namespace std;
using namespace BamTools;
struct CountTool::CountSettings {
// flags
- bool HasBamIndexFilename;
- bool HasInputBamFilename;
+ bool HasInput;
bool HasRegion;
// filenames
- string BamIndexFilename;
- string InputBamFilename;
+ vector<string> InputFiles;
string Region;
// constructor
CountSettings(void)
- : HasBamIndexFilename(false)
- , HasInputBamFilename(false)
+ : HasInput(false)
, HasRegion(false)
- , InputBamFilename(Options::StandardIn())
{ }
};
, m_settings(new CountSettings)
{
// set program details
- Options::SetProgramInfo("bamtools count", "prints alignment counts for a BAM file", "-in <filename> [-region <REGION> [-index <filename>]]");
+ Options::SetProgramInfo("bamtools count", "prints alignment counts for a BAM file", "-in <filename> [-region <REGION>]");
// set up options
OptionGroup* IO_Opts = Options::CreateOptionGroup("Input & Output");
- Options::AddValueOption("-in", "BAM filename", "the input BAM file", "", m_settings->HasInputBamFilename, m_settings->InputBamFilename, IO_Opts, Options::StandardIn());
- Options::AddValueOption("-index", "BAM index filename", "the BAM index file", "", m_settings->HasBamIndexFilename, m_settings->BamIndexFilename, IO_Opts);
+ //Options::AddValueOption("-in", "BAM filename", "the input BAM file", "", m_settings->HasInputBamFilename, m_settings->InputBamFilename, IO_Opts, Options::StandardIn());
+ Options::AddValueOption("-in", "BAM filename", "the input BAM file(s)", "", m_settings->HasInput, m_settings->InputFiles, IO_Opts);
+ //Options::AddValueOption("-index", "BAM index filename", "the BAM index file", "", m_settings->HasBamIndexFilename, m_settings->BamIndexFilename, IO_Opts);
OptionGroup* FilterOpts = Options::CreateOptionGroup("Filters");
- Options::AddValueOption("-region", "REGION", "genomic region. Index file is recommended for better performance. See \'bamtools help index\' for more details on creating one", "", m_settings->HasRegion, m_settings->Region, FilterOpts);
+ Options::AddValueOption("-region", "REGION", "genomic region. Index file is recommended for better performance, and is read automatically if it exists as <filename>.bai. See \'bamtools help index\' for more details on creating one", "", m_settings->HasRegion, m_settings->Region, FilterOpts);
}
CountTool::~CountTool(void) {
// parse command line arguments
Options::Parse(argc, argv, 1);
- //open our BAM reader
- BamReader reader;
- reader.Open(m_settings->InputBamFilename);
+ BamMultiReader reader;
+ reader.Open(m_settings->InputFiles, false, true);
// alignment counter
int alignmentCount(0);
// if no region specified, count entire file
if ( !m_settings->HasRegion ) {
BamAlignment al;
- while ( reader.GetNextAlignment(al) )
+ while ( reader.GetNextAlignmentCore(al) )
++alignmentCount;
- }
+ }
// more complicated - region specified
else {
- // parse region string for desired region
- string startChrom;
- string stopChrom;
- int startPos;
- int stopPos;
- if ( Utilities::ParseRegionString(m_settings->Region, startChrom, startPos, stopChrom, stopPos) ) {
-
- const RefVector references = reader.GetReferenceData();
-
- // -------------------------------
- // validate start ref & position
-
- int startRefID = reader.GetReferenceID(startChrom);
-
- // startRefID not found
- if ( startRefID == (int)references.size() ) {
- foundError = true;
- errorStream << "Start chrom: " << startChrom << " not found in BAM file." << endl;
- }
-
- // valid startRefID, check startPos
- else {
- const RefData& reference = references.at(startRefID);
-
- // startPos too large
- if ( startPos > reference.RefLength ) {
- foundError = true;
- errorStream << "Start pos: " << startPos << " is larger than expected." << endl;
+ BamRegion region;
+ if ( Utilities::ParseRegionString(m_settings->Region, reader, region) ) {
+
+ // check if there are index files *.bai corresponding to the input files
+ bool hasIndexes = true;
+ for (vector<string>::const_iterator f = m_settings->InputFiles.begin(); f != m_settings->InputFiles.end(); ++f) {
+ if (!Utilities::FileExists(*f + ".bai")) {
+ errorStream << "could not find index file " << *f + ".bai"
+ << ", parsing whole file(s) to get alignment counts for target region" << endl;
+ hasIndexes = false;
+ break;
}
}
-
- // -------------------------------
- // validate stop ref & position
-
- int stopRefID = reader.GetReferenceID(stopChrom);
-
- // skip if error already found
- if ( !foundError ) {
+ // has index, we can jump directly to
+ if ( hasIndexes ) {
- // stopRefID not found
- if ( stopRefID == (int)references.size() ) {
- foundError = true;
- errorStream << "Stop chrom: " << stopChrom << " not found in BAM file." << endl;
- }
-
- // valid stopRefID, check stopPos
- else {
- const RefData& reference = references.at(stopRefID);
-
- // stopPos too large
- if ( stopPos > reference.RefLength ) {
- foundError = true;
- errorStream << "Stop pos: " << stopPos << " is larger than expected." << endl;
- }
-
- // no stopPos specified, set to reference end
- else if ( stopPos == -1 ) {
- stopPos = reference.RefLength;
- }
+ // this is kind of a hack...?
+ BamMultiReader reader;
+ reader.Open(m_settings->InputFiles, true, true);
+
+ if ( !reader.SetRegion(region.LeftRefID, region.LeftPosition, region.RightRefID, region.RightPosition) ) {
+ foundError = true;
+ errorStream << "Could not set BamReader region to REGION: " << m_settings->Region << endl;
+ } else {
+ BamAlignment al;
+ while ( reader.GetNextAlignmentCore(al) )
+ ++alignmentCount;
}
- }
- // -------------------------------
- // if refs & positions valid, retrieve data
-
- if ( !foundError ) {
-
- // has index, we can jump directly to
- if ( m_settings->HasBamIndexFilename ) {
-
- // this is kind of a hack...?
- // re-open BamReader, this time with the index file
- reader.Close();
- reader.Open(m_settings->InputBamFilename, m_settings->BamIndexFilename);
-
- // attempt Jump(), catch error
- if ( !reader.Jump(startRefID, startPos) ) {
- foundError = true;
- errorStream << "Could not jump to desired REGION: " << m_settings->Region << endl;
- }
- }
-
- else {
-
- // read through sequentially, until first overlapping read is found
- BamAlignment al;
- bool alignmentFound(false);
- while( reader.GetNextAlignment(al) ) {
- if ( (al.RefID == startRefID ) && ( (al.Position + al.Length) >= startPos) ) {
- alignmentFound = true;
- break;
- }
+ } else {
+
+ // read through sequentially, until first overlapping read is found
+ BamAlignment al;
+ bool alignmentFound(false);
+ //reader.Open(m_settings->InputFiles, false, true);
+ while( reader.GetNextAlignmentCore(al) ) {
+ if ( (al.RefID == region.LeftRefID) && ( (al.Position + al.Length) >= region.LeftPosition ) ) {
+ alignmentFound = true;
+ break;
}
-
- // if overlapping alignment found (not EOF), increment counter
- if ( alignmentFound) ++ alignmentCount;
}
- // -----------------------------
- // count alignments until stop hit
-
- if ( !foundError ) {
- // while valid alignment AND
- // ( either current ref is before stopRefID OR
- // current ref stopRefID but we're before stopPos )
- BamAlignment al;
- while ( reader.GetNextAlignment(al) && ((al.RefID < stopRefID ) || (al.RefID == stopRefID && al.Position <= stopPos)) )
- ++alignmentCount;
- }
+ // if overlapping alignment found (not EOF), increment counter
+ if ( alignmentFound ) ++ alignmentCount;
}
- }
-
- // could not parse REGION string, set error
- else {
+
+ // -----------------------------
+ // count alignments until stop hit
+
+ } else {
foundError = true;
- errorStream << "Could not parse desired REGION: " << m_settings->Region << endl;
- errorStream << "Please see README for details on accepted REGION formats" << endl;
+ errorStream << "Could not parse REGION: " << m_settings->Region << endl;
+ errorStream << "Be sure REGION is in valid format (see README) and that coordinates are valid for selected references" << endl;
}
}
// clean & exit
reader.Close();
return (int)foundError;
-}
\ No newline at end of file
+}