|
0
|
1 /*****************************************************************************
|
|
|
2 bedFile.cpp
|
|
|
3
|
|
|
4 (c) 2009 - Aaron Quinlan
|
|
|
5 Hall Laboratory
|
|
|
6 Department of Biochemistry and Molecular Genetics
|
|
|
7 University of Virginia
|
|
|
8 aaronquinlan@gmail.com
|
|
|
9
|
|
|
10 Licensed under the GNU General Public License 2.0 license.
|
|
|
11 ******************************************************************************/
|
|
|
12 #include "bedFile.h"
|
|
|
13
|
|
|
14
|
|
|
15 /************************************************
|
|
|
16 Helper functions
|
|
|
17 *************************************************/
|
|
|
18 void splitBedIntoBlocks(const BED &bed, int lineNum, bedVector &bedBlocks) {
|
|
|
19
|
|
|
20 if (bed.otherFields.size() < 6) {
|
|
|
21 cerr << "Input error: Cannot split into blocks. Found interval with fewer than 12 columns on line " << lineNum << "." << endl;
|
|
|
22 exit(1);
|
|
|
23 }
|
|
|
24
|
|
|
25 int blockCount = atoi(bed.otherFields[3].c_str());
|
|
|
26 if ( blockCount <= 0 ) {
|
|
|
27 cerr << "Input error: found interval having <= 0 blocks on line " << lineNum << "." << endl;
|
|
|
28 exit(1);
|
|
|
29 }
|
|
|
30 else if ( blockCount == 1 ) {
|
|
|
31 //take a short-cut for single blocks
|
|
|
32 bedBlocks.push_back(bed);
|
|
|
33 }
|
|
|
34 else {
|
|
|
35 // get the comma-delimited strings for the BED12 block starts and block ends.
|
|
|
36 string blockSizes(bed.otherFields[4]);
|
|
|
37 string blockStarts(bed.otherFields[5]);
|
|
|
38
|
|
|
39 vector<int> sizes;
|
|
|
40 vector<int> starts;
|
|
|
41 Tokenize(blockSizes, sizes, ",");
|
|
|
42 Tokenize(blockStarts, starts, ",");
|
|
|
43
|
|
|
44 if ( sizes.size() != (size_t) blockCount || starts.size() != (size_t) blockCount ) {
|
|
|
45 cerr << "Input error: found interval with block-counts not matching starts/sizes on line " << lineNum << "." << endl;
|
|
|
46 exit(1);
|
|
|
47 }
|
|
|
48
|
|
|
49 // add each BED block to the bedBlocks vector
|
|
|
50 for (UINT i = 0; i < (UINT) blockCount; ++i) {
|
|
|
51 CHRPOS blockStart = bed.start + starts[i];
|
|
|
52 CHRPOS blockEnd = bed.start + starts[i] + sizes[i];
|
|
|
53 BED currBedBlock(bed.chrom, blockStart, blockEnd, bed.name, bed.score, bed.strand, bed.otherFields);
|
|
|
54 bedBlocks.push_back(currBedBlock);
|
|
|
55 }
|
|
|
56 }
|
|
|
57 }
|
|
|
58
|
|
|
59
|
|
|
60 /***********************************************
|
|
|
61 Sorting comparison functions
|
|
|
62 ************************************************/
|
|
|
63 bool sortByChrom(BED const &a, BED const &b) {
|
|
|
64 if (a.chrom < b.chrom) return true;
|
|
|
65 else return false;
|
|
|
66 };
|
|
|
67
|
|
|
68 bool sortByStart(const BED &a, const BED &b) {
|
|
|
69 if (a.start < b.start) return true;
|
|
|
70 else return false;
|
|
|
71 };
|
|
|
72
|
|
|
73 bool sortBySizeAsc(const BED &a, const BED &b) {
|
|
|
74
|
|
|
75 CHRPOS aLen = a.end - a.start;
|
|
|
76 CHRPOS bLen = b.end - b.start;
|
|
|
77
|
|
|
78 if (aLen < bLen) return true;
|
|
|
79 else return false;
|
|
|
80 };
|
|
|
81
|
|
|
82 bool sortBySizeDesc(const BED &a, const BED &b) {
|
|
|
83
|
|
|
84 CHRPOS aLen = a.end - a.start;
|
|
|
85 CHRPOS bLen = b.end - b.start;
|
|
|
86
|
|
|
87 if (aLen > bLen) return true;
|
|
|
88 else return false;
|
|
|
89 };
|
|
|
90
|
|
|
91 bool sortByScoreAsc(const BED &a, const BED &b) {
|
|
|
92 if (a.score < b.score) return true;
|
|
|
93 else return false;
|
|
|
94 };
|
|
|
95
|
|
|
96 bool sortByScoreDesc(const BED &a, const BED &b) {
|
|
|
97 if (a.score > b.score) return true;
|
|
|
98 else return false;
|
|
|
99 };
|
|
|
100
|
|
|
101 bool byChromThenStart(BED const &a, BED const &b) {
|
|
|
102
|
|
|
103 if (a.chrom < b.chrom) return true;
|
|
|
104 else if (a.chrom > b.chrom) return false;
|
|
|
105
|
|
|
106 if (a.start < b.start) return true;
|
|
|
107 else if (a.start >= b.start) return false;
|
|
|
108
|
|
|
109 return false;
|
|
|
110 };
|
|
|
111
|
|
|
112
|
|
|
113 /*******************************************
|
|
|
114 Class methods
|
|
|
115 *******************************************/
|
|
|
116
|
|
|
117 // Constructor
|
|
|
118 BedFile::BedFile(string &bedFile)
|
|
|
119 : bedFile(bedFile),
|
|
|
120 _isGff(false),
|
|
|
121 _isVcf(false),
|
|
|
122 _typeIsKnown(false),
|
|
|
123 _merged_start(-1),
|
|
|
124 _merged_end(-1),
|
|
|
125 _merged_chrom(""),
|
|
|
126 _prev_start(-1),
|
|
|
127 _prev_chrom("")
|
|
|
128 {}
|
|
|
129
|
|
|
130 // Destructor
|
|
|
131 BedFile::~BedFile(void) {
|
|
|
132 }
|
|
|
133
|
|
|
134
|
|
|
135 void BedFile::Open(void) {
|
|
|
136
|
|
|
137 _bedFields.reserve(12);
|
|
|
138
|
|
|
139 if (bedFile == "stdin" || bedFile == "-") {
|
|
|
140 _bedStream = &cin;
|
|
|
141 }
|
|
|
142 else {
|
|
|
143 _bedStream = new ifstream(bedFile.c_str(), ios::in);
|
|
|
144
|
|
|
145 if( isGzipFile(_bedStream) ) {
|
|
|
146 delete _bedStream;
|
|
|
147 _bedStream = new igzstream(bedFile.c_str(), ios::in);
|
|
|
148 }
|
|
|
149 if ( !(_bedStream->good()) ) {
|
|
|
150 cerr << "Error: The requested bed file (" << bedFile << ") could not be opened. Exiting!" << endl;
|
|
|
151 exit (1);
|
|
|
152 }
|
|
|
153 }
|
|
|
154 }
|
|
|
155
|
|
|
156 // Rewind the pointer back to the beginning of the file
|
|
|
157 void BedFile::Rewind(void) {
|
|
|
158 _bedStream->seekg(0, ios::beg);
|
|
|
159 }
|
|
|
160
|
|
|
161 // Jump to a specific byte in the file
|
|
|
162 void BedFile::Seek(unsigned long offset) {
|
|
|
163 _bedStream->seekg(offset);
|
|
|
164 }
|
|
|
165
|
|
|
166 // Jump to a specific byte in the file
|
|
|
167 bool BedFile::Empty() {
|
|
|
168 return _bedStream->eof();
|
|
|
169 }
|
|
|
170
|
|
|
171
|
|
|
172 // Close the BED file
|
|
|
173 void BedFile::Close(void) {
|
|
|
174 if (bedFile != "stdin" && bedFile != "-") delete _bedStream;
|
|
|
175 }
|
|
|
176
|
|
|
177
|
|
|
178 BedLineStatus BedFile::GetNextBed(BED &bed, int &lineNum, bool forceSorted) {
|
|
|
179
|
|
|
180 // make sure there are still lines to process.
|
|
|
181 // if so, tokenize, validate and return the BED entry.
|
|
|
182 _bedFields.clear();
|
|
|
183 // clear out the previous bed's data
|
|
|
184 if (_bedStream->good()) {
|
|
|
185 // parse the bedStream pointer
|
|
|
186 getline(*_bedStream, _bedLine);
|
|
|
187 lineNum++;
|
|
|
188
|
|
|
189 // split into a string vector.
|
|
|
190 Tokenize(_bedLine, _bedFields);
|
|
|
191
|
|
|
192 // load the BED struct as long as it's a valid BED entry.
|
|
|
193 BedLineStatus status = parseLine(bed, _bedFields, lineNum);
|
|
|
194 if (!forceSorted) {
|
|
|
195 return status;
|
|
|
196 }
|
|
|
197 else if (status == BED_VALID) {
|
|
|
198 if (bed.chrom == _prev_chrom) {
|
|
|
199 if ((int) bed.start >= _prev_start) {
|
|
|
200 _prev_chrom = bed.chrom;
|
|
|
201 _prev_start = bed.start;
|
|
|
202 return status;
|
|
|
203 }
|
|
|
204 else {
|
|
|
205 cerr << "ERROR: input file: (" << bedFile << ") is not sorted by chrom then start" << endl;
|
|
|
206 exit(1);
|
|
|
207 }
|
|
|
208 }
|
|
|
209 else if (bed.chrom > _prev_chrom) {
|
|
|
210 _prev_chrom = bed.chrom;
|
|
|
211 _prev_start = bed.start;
|
|
|
212 return status;
|
|
|
213 }
|
|
|
214 else if (bed.chrom < _prev_chrom) {
|
|
|
215 cerr << "ERROR: input file: (" << bedFile << ") is not sorted by chrom then start" << endl;
|
|
|
216 exit(1);
|
|
|
217 }
|
|
|
218 }
|
|
|
219 }
|
|
|
220
|
|
|
221 // default if file is closed or EOF
|
|
|
222 return BED_INVALID;
|
|
|
223 }
|
|
|
224
|
|
|
225
|
|
|
226 bool BedFile::GetNextMergedBed(BED &merged_bed, int &lineNum) {
|
|
|
227
|
|
|
228 if (_bedStream->good()) {
|
|
|
229 BED bed;
|
|
|
230 BedLineStatus bedStatus;
|
|
|
231 // force sorting; hence third param = true
|
|
|
232 while ((bedStatus = GetNextBed(bed, lineNum, true)) != BED_INVALID) {
|
|
|
233 if (bedStatus == BED_VALID) {
|
|
|
234 if (((int) bed.start - _merged_end > 0) ||
|
|
|
235 (_merged_end < 0) ||
|
|
|
236 (bed.chrom != _merged_chrom))
|
|
|
237 {
|
|
|
238 if (_merged_start >= 0) {
|
|
|
239 merged_bed.chrom = _merged_chrom;
|
|
|
240 merged_bed.start = _merged_start;
|
|
|
241 merged_bed.end = _merged_end;
|
|
|
242
|
|
|
243 _merged_chrom = bed.chrom;
|
|
|
244 _merged_start = bed.start;
|
|
|
245 _merged_end = bed.end;
|
|
|
246
|
|
|
247 return true;
|
|
|
248 }
|
|
|
249 else {
|
|
|
250 _merged_start = bed.start;
|
|
|
251 _merged_chrom = bed.chrom;
|
|
|
252 _merged_end = bed.end;
|
|
|
253 }
|
|
|
254 }
|
|
|
255 else if ((int) bed.end > _merged_end)
|
|
|
256 {
|
|
|
257 _merged_end = bed.end;
|
|
|
258 }
|
|
|
259 }
|
|
|
260 }
|
|
|
261 // handle the last merged block in the file.
|
|
|
262 if (bedStatus == BED_INVALID)
|
|
|
263 {
|
|
|
264 merged_bed.chrom = _merged_chrom;
|
|
|
265 merged_bed.start = _merged_start;
|
|
|
266 merged_bed.end = _merged_end;
|
|
|
267 return true;
|
|
|
268 }
|
|
|
269 }
|
|
|
270 return false;
|
|
|
271 }
|
|
|
272
|
|
|
273
|
|
|
274 void BedFile::FindOverlapsPerBin(string chrom, CHRPOS start, CHRPOS end,
|
|
|
275 string strand, vector<BED> &hits, bool sameStrand, bool diffStrand) {
|
|
|
276
|
|
|
277 BIN startBin, endBin;
|
|
|
278 startBin = (start >> _binFirstShift);
|
|
|
279 endBin = ((end-1) >> _binFirstShift);
|
|
|
280
|
|
|
281 // loop through each bin "level" in the binning hierarchy
|
|
|
282 for (BINLEVEL i = 0; i < _binLevels; ++i) {
|
|
|
283
|
|
|
284 // loop through each bin at this level of the hierarchy
|
|
|
285 BIN offset = _binOffsetsExtended[i];
|
|
|
286 for (BIN j = (startBin+offset); j <= (endBin+offset); ++j) {
|
|
|
287
|
|
|
288 // loop through each feature in this chrom/bin and see if it overlaps
|
|
|
289 // with the feature that was passed in. if so, add the feature to
|
|
|
290 // the list of hits.
|
|
|
291 vector<BED>::const_iterator bedItr = bedMap[chrom][j].begin();
|
|
|
292 vector<BED>::const_iterator bedEnd = bedMap[chrom][j].end();
|
|
|
293
|
|
|
294 for (; bedItr != bedEnd; ++bedItr) {
|
|
|
295 // do we have sufficient overlap?
|
|
|
296 if (overlaps(bedItr->start, bedItr->end, start, end) > 0) {
|
|
|
297
|
|
|
298 bool strands_are_same = (strand == bedItr->strand);
|
|
|
299
|
|
|
300 // test for necessary strandedness
|
|
|
301 if ( (sameStrand == false && diffStrand == false)
|
|
|
302 ||
|
|
|
303 (sameStrand == true && strands_are_same == true)
|
|
|
304 ||
|
|
|
305 (diffStrand == true && strands_are_same == false)
|
|
|
306 )
|
|
|
307 {
|
|
|
308 hits.push_back(*bedItr);
|
|
|
309 }
|
|
|
310 }
|
|
|
311 }
|
|
|
312 }
|
|
|
313 startBin >>= _binNextShift;
|
|
|
314 endBin >>= _binNextShift;
|
|
|
315 }
|
|
|
316 }
|
|
|
317
|
|
|
318
|
|
|
319 bool BedFile::FindOneOrMoreOverlapsPerBin(string chrom, CHRPOS start, CHRPOS end, string strand,
|
|
|
320 bool sameStrand, bool diffStrand, float overlapFraction) {
|
|
|
321
|
|
|
322 BIN startBin, endBin;
|
|
|
323 startBin = (start >> _binFirstShift);
|
|
|
324 endBin = ((end-1) >> _binFirstShift);
|
|
|
325
|
|
|
326 CHRPOS aLength = (end - start);
|
|
|
327
|
|
|
328 // loop through each bin "level" in the binning hierarchy
|
|
|
329 for (BINLEVEL i = 0; i < _binLevels; ++i) {
|
|
|
330
|
|
|
331 // loop through each bin at this level of the hierarchy
|
|
|
332 BIN offset = _binOffsetsExtended[i];
|
|
|
333 for (BIN j = (startBin+offset); j <= (endBin+offset); ++j) {
|
|
|
334
|
|
|
335 // loop through each feature in this chrom/bin and see if it overlaps
|
|
|
336 // with the feature that was passed in. if so, add the feature to
|
|
|
337 // the list of hits.
|
|
|
338 vector<BED>::const_iterator bedItr = bedMap[chrom][j].begin();
|
|
|
339 vector<BED>::const_iterator bedEnd = bedMap[chrom][j].end();
|
|
|
340 for (; bedItr != bedEnd; ++bedItr) {
|
|
|
341
|
|
|
342 CHRPOS s = max(start, bedItr->start);
|
|
|
343 CHRPOS e = min(end, bedItr->end);
|
|
|
344 // the number of overlapping bases b/w a and b
|
|
|
345 int overlapBases = (e - s);
|
|
|
346
|
|
|
347 // do we have sufficient overlap?
|
|
|
348 if ( (float) overlapBases / (float) aLength >= overlapFraction) {
|
|
|
349
|
|
|
350 bool strands_are_same = (strand == bedItr->strand);
|
|
|
351 // test for necessary strandedness
|
|
|
352 if ( (sameStrand == false && diffStrand == false)
|
|
|
353 ||
|
|
|
354 (sameStrand == true && strands_are_same == true)
|
|
|
355 ||
|
|
|
356 (diffStrand == true && strands_are_same == false)
|
|
|
357 )
|
|
|
358 {
|
|
|
359 return true;
|
|
|
360 }
|
|
|
361 }
|
|
|
362 }
|
|
|
363 }
|
|
|
364 startBin >>= _binNextShift;
|
|
|
365 endBin >>= _binNextShift;
|
|
|
366 }
|
|
|
367 return false;
|
|
|
368 }
|
|
|
369
|
|
|
370
|
|
|
371 bool BedFile::FindOneOrMoreReciprocalOverlapsPerBin(string chrom, CHRPOS start, CHRPOS end, string strand,
|
|
|
372 bool sameStrand, bool diffStrand, float overlapFraction) {
|
|
|
373
|
|
|
374 BIN startBin, endBin;
|
|
|
375 startBin = (start >> _binFirstShift);
|
|
|
376 endBin = ((end-1) >> _binFirstShift);
|
|
|
377
|
|
|
378 CHRPOS aLength = (end - start);
|
|
|
379
|
|
|
380 // loop through each bin "level" in the binning hierarchy
|
|
|
381 for (BINLEVEL i = 0; i < _binLevels; ++i) {
|
|
|
382
|
|
|
383 // loop through each bin at this level of the hierarchy
|
|
|
384 BIN offset = _binOffsetsExtended[i];
|
|
|
385 for (BIN j = (startBin+offset); j <= (endBin+offset); ++j) {
|
|
|
386
|
|
|
387 // loop through each feature in this chrom/bin and see if it overlaps
|
|
|
388 // with the feature that was passed in. if so, add the feature to
|
|
|
389 // the list of hits.
|
|
|
390 vector<BED>::const_iterator bedItr = bedMap[chrom][j].begin();
|
|
|
391 vector<BED>::const_iterator bedEnd = bedMap[chrom][j].end();
|
|
|
392 for (; bedItr != bedEnd; ++bedItr) {
|
|
|
393 CHRPOS s = max(start, bedItr->start);
|
|
|
394 CHRPOS e = min(end, bedItr->end);
|
|
|
395
|
|
|
396 // the number of overlapping bases b/w a and b
|
|
|
397 int overlapBases = (e - s);
|
|
|
398
|
|
|
399 // do we have sufficient overlap?
|
|
|
400 if ( (float) overlapBases / (float) aLength >= overlapFraction) {
|
|
|
401 CHRPOS bLength = (bedItr->end - bedItr->start);
|
|
|
402 float bOverlap = ( (float) overlapBases / (float) bLength );
|
|
|
403 bool strands_are_same = (strand == bedItr->strand);
|
|
|
404
|
|
|
405 // test for sufficient reciprocal overlap and strandedness
|
|
|
406 if ( (bOverlap >= overlapFraction) &&
|
|
|
407 ((sameStrand == false && diffStrand == false)
|
|
|
408 ||
|
|
|
409 (sameStrand == true && strands_are_same == true)
|
|
|
410 ||
|
|
|
411 (diffStrand == true && strands_are_same == false))
|
|
|
412 )
|
|
|
413 {
|
|
|
414 return true;
|
|
|
415 }
|
|
|
416 }
|
|
|
417 }
|
|
|
418 }
|
|
|
419 startBin >>= _binNextShift;
|
|
|
420 endBin >>= _binNextShift;
|
|
|
421 }
|
|
|
422 return false;
|
|
|
423 }
|
|
|
424
|
|
|
425
|
|
|
426 void BedFile::countHits(const BED &a, bool sameStrand, bool diffStrand, bool countsOnly) {
|
|
|
427
|
|
|
428 BIN startBin, endBin;
|
|
|
429 startBin = (a.start >> _binFirstShift);
|
|
|
430 endBin = ((a.end-1) >> _binFirstShift);
|
|
|
431
|
|
|
432 // loop through each bin "level" in the binning hierarchy
|
|
|
433 for (BINLEVEL i = 0; i < _binLevels; ++i) {
|
|
|
434
|
|
|
435 // loop through each bin at this level of the hierarchy
|
|
|
436 BIN offset = _binOffsetsExtended[i];
|
|
|
437 for (BIN j = (startBin+offset); j <= (endBin+offset); ++j) {
|
|
|
438
|
|
|
439 // loop through each feature in this chrom/bin and see if it overlaps
|
|
|
440 // with the feature that was passed in. if so, add the feature to
|
|
|
441 // the list of hits.
|
|
|
442 vector<BEDCOV>::iterator bedItr = bedCovMap[a.chrom][j].begin();
|
|
|
443 vector<BEDCOV>::iterator bedEnd = bedCovMap[a.chrom][j].end();
|
|
|
444 for (; bedItr != bedEnd; ++bedItr) {
|
|
|
445
|
|
|
446 bool strands_are_same = (a.strand == bedItr->strand);
|
|
|
447 // skip the hit if not on the same strand (and we care)
|
|
|
448 if ((sameStrand == true && strands_are_same == false) ||
|
|
|
449 (diffStrand == true && strands_are_same == true)
|
|
|
450 )
|
|
|
451 {
|
|
|
452 continue;
|
|
|
453 }
|
|
|
454 else if (overlaps(bedItr->start, bedItr->end, a.start, a.end) > 0) {
|
|
|
455
|
|
|
456 bedItr->count++;
|
|
|
457 if (countsOnly == false) {
|
|
|
458 if (a.zeroLength == false) {
|
|
|
459 bedItr->depthMap[a.start+1].starts++;
|
|
|
460 bedItr->depthMap[a.end].ends++;
|
|
|
461 }
|
|
|
462 else {
|
|
|
463 // correct for the fact that we artificially expanded the zeroLength feature
|
|
|
464 bedItr->depthMap[a.start+2].starts++;
|
|
|
465 bedItr->depthMap[a.end-1].ends++;
|
|
|
466 }
|
|
|
467
|
|
|
468 if (a.start < bedItr->minOverlapStart) {
|
|
|
469 bedItr->minOverlapStart = a.start;
|
|
|
470 }
|
|
|
471 }
|
|
|
472 }
|
|
|
473 }
|
|
|
474 }
|
|
|
475 startBin >>= _binNextShift;
|
|
|
476 endBin >>= _binNextShift;
|
|
|
477 }
|
|
|
478 }
|
|
|
479
|
|
|
480
|
|
|
481 void BedFile::countSplitHits(const vector<BED> &bedBlocks, bool sameStrand, bool diffStrand, bool countsOnly) {
|
|
|
482
|
|
|
483 // set to track the distinct B features that had coverage.
|
|
|
484 // we'll update the counts of coverage for these features by one
|
|
|
485 // at the end of this function to avoid over-counting.
|
|
|
486 set< vector<BEDCOV>::iterator > validHits;
|
|
|
487
|
|
|
488 vector<BED>::const_iterator blockItr = bedBlocks.begin();
|
|
|
489 vector<BED>::const_iterator blockEnd = bedBlocks.end();
|
|
|
490 for (; blockItr != blockEnd; ++blockItr) {
|
|
|
491
|
|
|
492 BIN startBin, endBin;
|
|
|
493 startBin = (blockItr->start >> _binFirstShift);
|
|
|
494 endBin = ((blockItr->end-1) >> _binFirstShift);
|
|
|
495
|
|
|
496 // loop through each bin "level" in the binning hierarchy
|
|
|
497 for (BINLEVEL i = 0; i < _binLevels; ++i) {
|
|
|
498
|
|
|
499 // loop through each bin at this level of the hierarchy
|
|
|
500 BIN offset = _binOffsetsExtended[i];
|
|
|
501 for (BIN j = (startBin+offset); j <= (endBin+offset); ++j) {
|
|
|
502
|
|
|
503 // loop through each feature in this chrom/bin and see if it overlaps
|
|
|
504 // with the feature that was passed in. if so, add the feature to
|
|
|
505 // the list of hits.
|
|
|
506 vector<BEDCOV>::iterator bedItr = bedCovMap[blockItr->chrom][j].begin();
|
|
|
507 vector<BEDCOV>::iterator bedEnd = bedCovMap[blockItr->chrom][j].end();
|
|
|
508 for (; bedItr != bedEnd; ++bedItr) {
|
|
|
509
|
|
|
510 bool strands_are_same = (blockItr->strand == bedItr->strand);
|
|
|
511 // skip the hit if not on the same strand (and we care)
|
|
|
512 if ((sameStrand == true && strands_are_same == false) ||
|
|
|
513 (diffStrand == true && strands_are_same == true)
|
|
|
514 )
|
|
|
515 {
|
|
|
516 continue;
|
|
|
517 }
|
|
|
518 else if (overlaps(bedItr->start, bedItr->end, blockItr->start, blockItr->end) > 0) {
|
|
|
519 if (countsOnly == false) {
|
|
|
520 if (blockItr->zeroLength == false) {
|
|
|
521 bedItr->depthMap[blockItr->start+1].starts++;
|
|
|
522 bedItr->depthMap[blockItr->end].ends++;
|
|
|
523 }
|
|
|
524 else {
|
|
|
525 // correct for the fact that we artificially expanded the zeroLength feature
|
|
|
526 bedItr->depthMap[blockItr->start+2].starts++;
|
|
|
527 bedItr->depthMap[blockItr->end-1].ends++;
|
|
|
528 }
|
|
|
529 }
|
|
|
530
|
|
|
531 validHits.insert(bedItr);
|
|
|
532 if (blockItr->start < bedItr->minOverlapStart)
|
|
|
533 bedItr->minOverlapStart = blockItr->start;
|
|
|
534 }
|
|
|
535 }
|
|
|
536 }
|
|
|
537 startBin >>= _binNextShift;
|
|
|
538 endBin >>= _binNextShift;
|
|
|
539 }
|
|
|
540 }
|
|
|
541 // incrment the count of overlap by one for each B feature that overlapped
|
|
|
542 // the current passed hit. This is necessary to prevent over-counting for
|
|
|
543 // each "split"" of a single read.
|
|
|
544 set< vector<BEDCOV>::iterator >::iterator validHitsItr = validHits.begin();
|
|
|
545 set< vector<BEDCOV>::iterator >::iterator validHitsEnd = validHits.end();
|
|
|
546 for (; validHitsItr != validHitsEnd; ++validHitsItr)
|
|
|
547 // the validHitsItr points to another itr, hence the (*itr)-> dereferencing.
|
|
|
548 // ugly, but that's C++.
|
|
|
549 (*validHitsItr)->count++;
|
|
|
550 }
|
|
|
551
|
|
|
552
|
|
|
553 void BedFile::countListHits(const BED &a, int index, bool sameStrand, bool diffStrand) {
|
|
|
554
|
|
|
555 BIN startBin, endBin;
|
|
|
556 startBin = (a.start >> _binFirstShift);
|
|
|
557 endBin = ((a.end-1) >> _binFirstShift);
|
|
|
558
|
|
|
559 // loop through each bin "level" in the binning hierarchy
|
|
|
560 for (BINLEVEL i = 0; i < _binLevels; ++i) {
|
|
|
561
|
|
|
562 // loop through each bin at this level of the hierarchy
|
|
|
563 BIN offset = _binOffsetsExtended[i];
|
|
|
564 for (BIN j = (startBin+offset); j <= (endBin+offset); ++j) {
|
|
|
565
|
|
|
566 // loop through each feature in this chrom/bin and see if it overlaps
|
|
|
567 // with the feature that was passed in. if so, add the feature to
|
|
|
568 // the list of hits.
|
|
|
569 vector<BEDCOVLIST>::iterator bedItr = bedCovListMap[a.chrom][j].begin();
|
|
|
570 vector<BEDCOVLIST>::iterator bedEnd = bedCovListMap[a.chrom][j].end();
|
|
|
571 for (; bedItr != bedEnd; ++bedItr) {
|
|
|
572
|
|
|
573 bool strands_are_same = (a.strand == bedItr->strand);
|
|
|
574 // skip the hit if not on the same strand (and we care)
|
|
|
575 if ((sameStrand == true && strands_are_same == false) ||
|
|
|
576 (diffStrand == true && strands_are_same == true)
|
|
|
577 )
|
|
|
578 {
|
|
|
579 continue;
|
|
|
580 }
|
|
|
581 else if (overlaps(bedItr->start, bedItr->end, a.start, a.end) > 0) {
|
|
|
582 bedItr->counts[index]++;
|
|
|
583 if (a.zeroLength == false) {
|
|
|
584 bedItr->depthMapList[index][a.start+1].starts++;
|
|
|
585 bedItr->depthMapList[index][a.end].ends++;
|
|
|
586 }
|
|
|
587 else {
|
|
|
588 // correct for the fact that we artificially expanded the zeroLength feature
|
|
|
589 bedItr->depthMapList[index][a.start+2].starts++;
|
|
|
590 bedItr->depthMapList[index][a.end-1].ends++;
|
|
|
591 }
|
|
|
592
|
|
|
593 if (a.start < bedItr->minOverlapStarts[index]) {
|
|
|
594 bedItr->minOverlapStarts[index] = a.start;
|
|
|
595 }
|
|
|
596 }
|
|
|
597 }
|
|
|
598 }
|
|
|
599 startBin >>= _binNextShift;
|
|
|
600 endBin >>= _binNextShift;
|
|
|
601 }
|
|
|
602 }
|
|
|
603
|
|
|
604 void BedFile::setZeroBased(bool zeroBased) { this->isZeroBased = zeroBased; }
|
|
|
605
|
|
|
606 void BedFile::setGff (bool gff) { this->_isGff = gff; }
|
|
|
607
|
|
|
608
|
|
|
609 void BedFile::setVcf (bool vcf) { this->_isVcf = vcf; }
|
|
|
610
|
|
|
611
|
|
|
612 void BedFile::setFileType (FileType type) {
|
|
|
613 _fileType = type;
|
|
|
614 _typeIsKnown = true;
|
|
|
615 }
|
|
|
616
|
|
|
617
|
|
|
618 void BedFile::setBedType (int colNums) {
|
|
|
619 bedType = colNums;
|
|
|
620 }
|
|
|
621
|
|
|
622
|
|
|
623 void BedFile::loadBedFileIntoMap() {
|
|
|
624
|
|
|
625 BED bedEntry, nullBed;
|
|
|
626 int lineNum = 0;
|
|
|
627 BedLineStatus bedStatus;
|
|
|
628
|
|
|
629 Open();
|
|
|
630 while ((bedStatus = GetNextBed(bedEntry, lineNum)) != BED_INVALID) {
|
|
|
631 if (bedStatus == BED_VALID) {
|
|
|
632 BIN bin = getBin(bedEntry.start, bedEntry.end);
|
|
|
633 bedMap[bedEntry.chrom][bin].push_back(bedEntry);
|
|
|
634 bedEntry = nullBed;
|
|
|
635 }
|
|
|
636 }
|
|
|
637 Close();
|
|
|
638 }
|
|
|
639
|
|
|
640
|
|
|
641 void BedFile::loadBedCovFileIntoMap() {
|
|
|
642
|
|
|
643 BED bedEntry, nullBed;
|
|
|
644 int lineNum = 0;
|
|
|
645 BedLineStatus bedStatus;
|
|
|
646
|
|
|
647 Open();
|
|
|
648 while ((bedStatus = GetNextBed(bedEntry, lineNum)) != BED_INVALID) {
|
|
|
649 if (bedStatus == BED_VALID) {
|
|
|
650 BIN bin = getBin(bedEntry.start, bedEntry.end);
|
|
|
651
|
|
|
652 BEDCOV bedCov;
|
|
|
653 bedCov.chrom = bedEntry.chrom;
|
|
|
654 bedCov.start = bedEntry.start;
|
|
|
655 bedCov.end = bedEntry.end;
|
|
|
656 bedCov.name = bedEntry.name;
|
|
|
657 bedCov.score = bedEntry.score;
|
|
|
658 bedCov.strand = bedEntry.strand;
|
|
|
659 bedCov.otherFields = bedEntry.otherFields;
|
|
|
660 bedCov.zeroLength = bedEntry.zeroLength;
|
|
|
661 bedCov.count = 0;
|
|
|
662 bedCov.minOverlapStart = INT_MAX;
|
|
|
663
|
|
|
664 bedCovMap[bedEntry.chrom][bin].push_back(bedCov);
|
|
|
665 bedEntry = nullBed;
|
|
|
666 }
|
|
|
667 }
|
|
|
668 Close();
|
|
|
669 }
|
|
|
670
|
|
|
671 void BedFile::loadBedCovListFileIntoMap() {
|
|
|
672
|
|
|
673 BED bedEntry, nullBed;
|
|
|
674 int lineNum = 0;
|
|
|
675 BedLineStatus bedStatus;
|
|
|
676
|
|
|
677 Open();
|
|
|
678 while ((bedStatus = GetNextBed(bedEntry, lineNum)) != BED_INVALID) {
|
|
|
679 if (bedStatus == BED_VALID) {
|
|
|
680 BIN bin = getBin(bedEntry.start, bedEntry.end);
|
|
|
681
|
|
|
682 BEDCOVLIST bedCovList;
|
|
|
683 bedCovList.chrom = bedEntry.chrom;
|
|
|
684 bedCovList.start = bedEntry.start;
|
|
|
685 bedCovList.end = bedEntry.end;
|
|
|
686 bedCovList.name = bedEntry.name;
|
|
|
687 bedCovList.score = bedEntry.score;
|
|
|
688 bedCovList.strand = bedEntry.strand;
|
|
|
689 bedCovList.otherFields = bedEntry.otherFields;
|
|
|
690 bedCovList.zeroLength = bedEntry.zeroLength;
|
|
|
691
|
|
|
692 bedCovListMap[bedEntry.chrom][bin].push_back(bedCovList);
|
|
|
693 bedEntry = nullBed;
|
|
|
694 }
|
|
|
695 }
|
|
|
696 Close();
|
|
|
697 }
|
|
|
698
|
|
|
699
|
|
|
700 void BedFile::loadBedFileIntoMapNoBin() {
|
|
|
701
|
|
|
702 BED bedEntry, nullBed;
|
|
|
703 int lineNum = 0;
|
|
|
704 BedLineStatus bedStatus;
|
|
|
705
|
|
|
706 Open();
|
|
|
707 while ((bedStatus = this->GetNextBed(bedEntry, lineNum)) != BED_INVALID) {
|
|
|
708 if (bedStatus == BED_VALID) {
|
|
|
709 bedMapNoBin[bedEntry.chrom].push_back(bedEntry);
|
|
|
710 bedEntry = nullBed;
|
|
|
711 }
|
|
|
712 }
|
|
|
713 Close();
|
|
|
714
|
|
|
715 // sort the BED entries for each chromosome
|
|
|
716 // in ascending order of start position
|
|
|
717 for (masterBedMapNoBin::iterator m = this->bedMapNoBin.begin(); m != this->bedMapNoBin.end(); ++m) {
|
|
|
718 sort(m->second.begin(), m->second.end(), sortByStart);
|
|
|
719 }
|
|
|
720 }
|
