1 // ***************************************************************************
2 // BamMultiReader.cpp (c) 2010 Erik Garrison, Derek Barnett
3 // Marth Lab, Department of Biology, Boston College
4 // ---------------------------------------------------------------------------
5 // Last modified: 7 October 2011 (DB)
6 // ---------------------------------------------------------------------------
7 // Convenience class for reading multiple BAM files.
9 // This functionality allows applications to work on very large sets of files
10 // without requiring intermediate merge, sort, and index steps for each file
11 // subset. It also improves the performance of our merge system as it
12 // precludes the need to sort merged files.
13 // ***************************************************************************
15 #include <api/BamMultiReader.h>
16 #include <api/internal/BamMultiReader_p.h>
17 using namespace BamTools;
23 /*! \class BamTools::BamReader
24 \brief Convenience class for reading multiple BAM files.
27 /*! \fn BamMultiReader::BamMultiReader(void)
30 BamMultiReader::BamMultiReader(void)
31 : d(new Internal::BamMultiReaderPrivate)
34 /*! \fn BamMultiReader::~BamMultiReader(void)
37 BamMultiReader::~BamMultiReader(void) {
42 /*! \fn void BamMultiReader::Close(void)
43 \brief Closes all open BAM files.
45 Also clears out all header and reference data.
47 \sa CloseFile(), IsOpen(), Open(), BamReader::Close()
49 bool BamMultiReader::Close(void) {
53 /*! \fn void BamMultiReader::CloseFile(const std::string& filename)
54 \brief Closes requested BAM file.
56 Leaves any other file(s) open, along with header and reference data.
58 \sa Close(), IsOpen(), Open(), BamReader::Close()
60 bool BamMultiReader::CloseFile(const std::string& filename) {
61 return d->CloseFile(filename);
64 /*! \fn bool BamMultiReader::CreateIndexes(const BamIndex::IndexType& type)
65 \brief Creates index files for the current BAM files.
67 \param type file format to create, see BamIndex::IndexType for available formats
68 \return \c true if index files created OK
69 \sa LocateIndexes(), OpenIndexes(), BamReader::CreateIndex()
71 bool BamMultiReader::CreateIndexes(const BamIndex::IndexType& type) {
72 return d->CreateIndexes(type);
75 /*! \fn const std::vector<std::string> BamMultiReader::Filenames(void) const
76 \brief Returns list of filenames for all open BAM files.
78 Retrieved filenames will contain whatever was passed via Open().
79 If you need full directory paths here, be sure to include them
80 when you open the BAM files.
82 \returns names of open BAM files. If no files are open, returns an empty vector.
83 \sa IsOpen(), BamReader::GetFilename()
85 const std::vector<std::string> BamMultiReader::Filenames(void) const {
86 return d->Filenames();
89 // returns a description of the last error that occurred
90 std::string BamMultiReader::GetErrorString(void) const {
91 return d->GetErrorString();
94 /*! \fn SamHeader BamMultiReader::GetHeader(void) const
95 \brief Returns unified SAM-format header for all files
97 N.B. - Modifying the retrieved text does NOT affect the current
98 BAM files. Thesse file have been opened in a read-only mode. However,
99 your modified header text can be used in conjunction with BamWriter
100 to generate a new BAM file with the appropriate header information.
102 \returns header data wrapped in SamHeader object
103 \sa GetHeaderText(), BamReader::GetHeader()
105 SamHeader BamMultiReader::GetHeader(void) const {
106 return d->GetHeader();
109 /*! \fn std::string BamMultiReader::GetHeaderText(void) const
110 \brief Returns unified SAM-format header text for all files
112 N.B. - Modifying the retrieved text does NOT affect the current
113 BAM files. Thesse file have been opened in a read-only mode. However,
114 your modified header text can be used in conjunction with BamWriter
115 to generate a new BAM file with the appropriate header information.
117 \returns SAM-formatted header text
118 \sa GetHeader(), BamReader::GetHeaderText()
120 std::string BamMultiReader::GetHeaderText(void) const {
121 return d->GetHeaderText();
124 /*! \fn bool BamMultiReader::GetNextAlignment(BamAlignment& alignment)
125 \brief Retrieves next available alignment.
127 Equivalent to BamReader::GetNextAlignment() with respect to what is a valid
128 overlapping alignment and what data gets populated.
130 This method takes care of determining which alignment actually is 'next'
131 across multiple files, depending on current SortOrder.
133 \param alignment destination for alignment record data
134 \returns \c true if a valid alignment was found
135 \sa GetNextAlignmentCore(), SetRegion(), SetSortOrder(), BamReader::GetNextAlignment()
137 bool BamMultiReader::GetNextAlignment(BamAlignment& nextAlignment) {
138 return d->GetNextAlignment(nextAlignment);
141 /*! \fn bool BamMultiReader::GetNextAlignmentCore(BamAlignment& alignment)
142 \brief Retrieves next available alignment.
144 Equivalent to BamReader::GetNextAlignmentCore() with respect to what is a valid
145 overlapping alignment and what data gets populated.
147 This method takes care of determining which alignment actually is 'next'
148 across multiple files, depending on current SortOrder.
150 \param alignment destination for alignment record data
151 \returns \c true if a valid alignment was found
152 \sa GetNextAlignment(), SetRegion(), SetSortOrder(), BamReader::GetNextAlignmentCore()
154 bool BamMultiReader::GetNextAlignmentCore(BamAlignment& nextAlignment) {
155 return d->GetNextAlignmentCore(nextAlignment);
158 /*! \fn int BamMultiReader::GetReferenceCount(void) const
159 \brief Returns number of reference sequences.
160 \sa BamReader::GetReferenceCount()
162 int BamMultiReader::GetReferenceCount(void) const {
163 return d->GetReferenceCount();
166 /*! \fn const RefVector& BamMultiReader::GetReferenceData(void) const
167 \brief Returns all reference sequence entries.
168 \sa RefData, BamReader::GetReferenceData()
170 const BamTools::RefVector BamMultiReader::GetReferenceData(void) const {
171 return d->GetReferenceData();
174 /*! \fn int BamMultiReader::GetReferenceID(const std::string& refName) const
175 \brief Returns the ID of the reference with this name.
177 If \a refName is not found, returns -1.
179 \sa BamReader::GetReferenceID()
181 int BamMultiReader::GetReferenceID(const std::string& refName) const {
182 return d->GetReferenceID(refName);
185 /*! \fn bool BamMultiReader::HasIndexes(void) const
186 \brief Returns \c true if all BAM files have index data available.
187 \sa BamReader::HasIndex()
189 bool BamMultiReader::HasIndexes(void) const {
190 return d->HasIndexes();
193 /*! \fn bool BamMultiReader::HasOpenReaders(void) const
194 \brief Returns \c true if there are any open BAM files.
196 bool BamMultiReader::HasOpenReaders(void) const {
197 return d->HasOpenReaders();
200 /*! \fn bool BamMultiReader::Jump(int refID, int position)
201 \brief Performs a random-access jump within current BAM files.
203 This is a convenience method, equivalent to calling SetRegion()
204 with only a left boundary specified.
206 \returns \c true if jump was successful
207 \sa HasIndex(), BamReader::Jump()
210 bool BamMultiReader::Jump(int refID, int position) {
211 return d->Jump(refID, position);
214 /*! \fn bool BamMultiReader::LocateIndexes(const BamIndex::IndexType& preferredType)
215 \brief Looks for index files that match current BAM files.
217 Use this function when you need index files, and perhaps have a
218 preferred index format, but do not depend heavily on which indexes
219 actually get loaded at runtime.
221 For each BAM file, this function will defer to your \a preferredType
222 whenever possible. However, if an index file of \a preferredType can
223 not be found, then it will look for any other index file that matches
226 An example case would look this:
229 BamMultiReader reader;
232 // ensure that all files have an index
233 if ( !reader.LocateIndexes() ) // opens any existing index files that match our BAM files
234 reader.CreateIndexes(); // creates index files for BAM files that still lack one
236 // do interesting stuff
241 If you want precise control over which index files are loaded, use OpenIndexes()
242 with the desired index filenames. If that function returns false, you can use
243 CreateIndexes() to then build index files of the exact requested format.
245 \param preferredType desired index file format, see BamIndex::IndexType for available formats
246 \returns \c true if index files could be found for \b ALL open BAM files
247 \sa BamReader::LocateIndex()
249 bool BamMultiReader::LocateIndexes(const BamIndex::IndexType& preferredType) {
250 return d->LocateIndexes(preferredType);
253 /*! \fn bool BamMultiReader::Open(const std::vector<std::string>& filenames)
254 \brief Opens BAM files.
256 N.B. - Opening BAM files will invalidate any current region set on the multireader.
257 All file pointers will be returned to the beginning of the alignment data.
258 Follow this with Jump() or SetRegion() to establish a region of interest.
260 \param filenames list of BAM filenames to open
261 \returns \c true if BAM files were opened successfully
262 \sa Close(), HasOpenReaders(), OpenFile(), OpenIndexes(), BamReader::Open()
264 bool BamMultiReader::Open(const std::vector<std::string>& filenames) {
265 return d->Open(filenames);
268 /*! \fn bool BamMultiReader::OpenFile(const std::string& filename)
269 \brief Opens a single BAM file.
271 Adds another BAM file to multireader "on-the-fly".
273 N.B. - Opening a BAM file invalidates any current region set on the multireader.
274 All file pointers will be returned to the beginning of the alignment data.
275 Follow this with Jump() or SetRegion() to establish a region of interest.
277 \param filename BAM filename to open
278 \returns \c true if BAM file was opened successfully
279 \sa Close(), HasOpenReaders(), Open(), OpenIndexes(), BamReader::Open()
281 bool BamMultiReader::OpenFile(const std::string& filename) {
282 return d->OpenFile(filename);
285 /*! \fn bool BamMultiReader::OpenIndexes(const std::vector<std::string>& indexFilenames)
286 \brief Opens index files for current BAM files.
288 N.B. - Currently assumes that index filenames match the order (and number) of
289 BAM files passed to Open().
291 \param indexFilenames list of BAM index file names
292 \returns \c true if BAM index file was opened & data loaded successfully
293 \sa LocateIndex(), Open(), SetIndex(), BamReader::OpenIndex()
295 bool BamMultiReader::OpenIndexes(const std::vector<std::string>& indexFilenames) {
296 return d->OpenIndexes(indexFilenames);
299 /*! \fn bool BamMultiReader::Rewind(void)
300 \brief Returns the internal file pointers to the beginning of alignment records.
302 Useful for performing multiple sequential passes through BAM files.
303 Calling this function clears any prior region that may have been set.
305 \returns \c true if rewind operation was successful
306 \sa Jump(), SetRegion(), BamReader::Rewind()
308 bool BamMultiReader::Rewind(void) {
312 /*! \fn void BamMultiReader::SetIndexCacheMode(const BamIndex::IndexCacheMode& mode)
313 \brief Changes the caching behavior of the index data.
315 Default mode is BamIndex::LimitedIndexCaching.
317 \param mode desired cache mode for index, see BamIndex::IndexCacheMode for
318 description of the available cache modes
319 \sa HasIndex(), BamReader::SetIndexCacheMode()
321 void BamMultiReader::SetIndexCacheMode(const BamIndex::IndexCacheMode& mode) {
322 d->SetIndexCacheMode(mode);
325 /*! \fn bool BamMultiReader::SetRegion(const BamRegion& region)
326 \brief Sets a target region of interest
328 Equivalent to calling BamReader::SetRegion() on all open BAM files.
330 \param region desired region-of-interest to activate
331 \returns \c true if ALL readers set the region successfully
332 \sa HasIndexes(), Jump(), BamReader::SetRegion()
334 bool BamMultiReader::SetRegion(const BamRegion& region) {
335 return d->SetRegion(region);
338 /*! \fn bool BamMultiReader::SetRegion(const int& leftRefID,
339 const int& leftPosition,
340 const int& rightRefID,
341 const int& rightPosition)
342 \brief Sets a target region of interest
344 This is an overloaded function.
346 Equivalent to calling BamReader::SetRegion() on all open BAM files.
348 \param leftRefID referenceID of region's left boundary
349 \param leftPosition position of region's left boundary
350 \param rightRefID reference ID of region's right boundary
351 \param rightPosition position of region's right boundary
353 \returns \c true if ALL readers set the region successfully
354 \sa HasIndexes(), Jump(), BamReader::SetRegion()
356 bool BamMultiReader::SetRegion(const int& leftRefID,
357 const int& leftPosition,
358 const int& rightRefID,
359 const int& rightPosition)
361 BamRegion region(leftRefID, leftPosition, rightRefID, rightPosition);
362 return d->SetRegion(region);