libBigWig
bigWig.h
Go to the documentation of this file.
1 #ifndef LIBBIGWIG_H
2 #define LIBBIGWIG_H
3 
4 #include "bigWigIO.h"
5 #include "bwValues.h"
6 #include <inttypes.h>
7 #include <zlib.h>
8 
9 #ifdef __cplusplus
10 extern "C" {
11 #endif
12 
59 #define LIBBIGWIG_VERSION 0.4.7
60 
64 #ifdef NOCURL
65 #define LIBBIGWIG_CURL 0
66 #ifndef CURLTYPE_DEFINED
67 #define CURLTYPE_DEFINED
68 typedef int CURLcode;
69 typedef void CURL;
70 #endif
71 #else
72 #define LIBBIGWIG_CURL 1
73 #endif
74 
78 #define BIGWIG_MAGIC 0x888FFC26
82 #define BIGBED_MAGIC 0x8789F2EB
86 #define CIRTREE_MAGIC 0x78ca8c91
90 #define IDX_MAGIC 0x2468ace0
94 #define DEFAULT_nCHILDREN 64
98 #define DEFAULT_BLOCKSIZE 32768
99 
105  mean = 0,
106  average = 0,
107  stdev = 1,
108  dev = 1,
109  max = 2,
110  min = 3,
111  cov = 4,
112  coverage = 4,
113  sum = 5
114 };
115 
116 //Should hide this from end users
122 typedef struct {
123  uint32_t *level;
124  //There's 4 bytes of padding between these
125  uint64_t *dataOffset;
126  uint64_t *indexOffset;
128 } bwZoomHdr_t;
129 
135 typedef struct {
136  uint16_t version;
137  uint16_t nLevels;
138  uint64_t ctOffset;
139  uint64_t dataOffset;
140  uint64_t indexOffset;
141  uint16_t fieldCount;
142  uint16_t definedFieldCount;
143  uint64_t sqlOffset;
144  uint64_t summaryOffset;
145  uint32_t bufSize;
146  uint64_t extensionOffset;
148  //total Summary
149  uint64_t nBasesCovered;
150  double minVal;
151  double maxVal;
152  double sumData;
153  double sumSquared;
154 } bigWigHdr_t;
155 
156 //Should probably replace this with a hash
160 typedef struct {
161  int64_t nKeys;
162  char **chrom;
163  uint32_t *len;
164 } chromList_t;
165 
166 //TODO remove from bigWig.h
168 typedef struct bwLL bwLL;
169 struct bwLL {
170  bwRTreeNode_t *node;
171  struct bwLL *next;
172 };
173 typedef struct bwZoomBuffer_t bwZoomBuffer_t;
174 struct bwZoomBuffer_t { //each individual entry takes 32 bytes
175  void *p;
176  uint32_t l, m;
177  struct bwZoomBuffer_t *next;
178 };
180 
185 typedef struct {
186  uint64_t nBlocks;
187  uint32_t blockSize;
188  uint64_t nEntries;
189  uint64_t runningWidthSum;
190  uint32_t tid;
191  uint32_t start;
192  uint32_t end;
193  uint32_t span;
194  uint32_t step;
195  uint8_t ltype;
196  uint32_t l;
197  void *p;
200  bwZoomBuffer_t **firstZoomBuffer;
201  bwZoomBuffer_t **lastZoomBuffer;
202  uint64_t *nNodes;
203  uLongf compressPsz;
204  void *compressP;
206 
210 typedef struct {
216  int isWrite;
217  int type;
218 } bigWigFile_t;
219 
223 typedef struct {
224  uint32_t l;
225  uint32_t m;
226  uint32_t *start;
227  uint32_t *end;
228  float *value;
230 
234 typedef struct {
235  uint32_t l;
236  uint32_t m;
237  uint32_t *start;
238  uint32_t *end;
239  char **str;
241 
246 typedef struct {
248  uint32_t tid;
249  uint32_t start;
250  uint32_t end;
251  uint64_t offset;
254  void *blocks;
257  void *data;
259 
267 int bwInit(size_t bufSize);
268 
273 void bwCleanup(void);
274 
282 int bwIsBigWig(const char *fname, CURLcode (*callBack)(CURL*));
283 
291 int bbIsBigBed(const char *fname, CURLcode (*callBack)(CURL*));
292 
301 bigWigFile_t *bwOpen(const char *fname, CURLcode (*callBack)(CURL*), const char* mode);
302 
310 bigWigFile_t *bbOpen(const char *fname, CURLcode (*callBack)(CURL*));
311 
320 char *bbGetSQL(bigWigFile_t *fp);
321 
327 void bwClose(bigWigFile_t *fp);
328 
329 /*******************************************************************************
330 *
331 * The following are in bwStats.c
332 *
333 *******************************************************************************/
334 
342 uint32_t bwGetTid(const bigWigFile_t *fp, const char *chrom);
343 
350 
357 
370 bwOverlappingIntervals_t *bwGetOverlappingIntervals(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end);
371 
384 bbOverlappingEntries_t *bbGetOverlappingEntries(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, int withString);
385 
400 bwOverlapIterator_t *bwOverlappingIntervalsIterator(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, uint32_t blocksPerIteration);
401 
418 bwOverlapIterator_t *bbOverlappingEntriesIterator(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, int withString, uint32_t blocksPerIteration);
419 
428 
434 
448 bwOverlappingIntervals_t *bwGetValues(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, int includeNA);
449 
462 double *bwStats(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, uint32_t nBins, enum bwStatsType type);
463 
476 double *bwStatsFromFull(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, uint32_t nBins, enum bwStatsType type);
477 
478 //Writer functions
479 
487 int bwCreateHdr(bigWigFile_t *fp, int32_t maxZooms);
488 
497 chromList_t *bwCreateChromList(const char* const* chroms, const uint32_t *lengths, int64_t n);
498 
506 int bwWriteHdr(bigWigFile_t *bw);
507 
524 int bwAddIntervals(bigWigFile_t *fp, const char* const* chrom, const uint32_t *start, const uint32_t *end, const float *values, uint32_t n);
525 
538 int bwAppendIntervals(bigWigFile_t *fp, const uint32_t *start, const uint32_t *end, const float *values, uint32_t n);
539 
556 int bwAddIntervalSpans(bigWigFile_t *fp, const char *chrom, const uint32_t *start, uint32_t span, const float *values, uint32_t n);
557 
569 int bwAppendIntervalSpans(bigWigFile_t *fp, const uint32_t *start, const float *values, uint32_t n);
570 
588 int bwAddIntervalSpanSteps(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t span, uint32_t step, const float *values, uint32_t n);
589 
600 int bwAppendIntervalSpanSteps(bigWigFile_t *fp, const float *values, uint32_t n);
601 
602 #ifdef __cplusplus
603 }
604 #endif
605 
606 #endif // LIBBIGWIG_H
bigWigFile_t * bwOpen(const char *fname, CURLcode(*callBack)(CURL *), const char *mode)
Opens a local or remote bigWig file. This will open a local or remote bigWig file....
Definition: bwRead.c:345
int bwCreateHdr(bigWigFile_t *fp, int32_t maxZooms)
Create a largely empty bigWig header Every bigWig file has a header, this creates the template for on...
Definition: bwWrite.c:56
void bbDestroyOverlappingEntries(bbOverlappingEntries_t *o)
Frees space allocated by bbGetOverlappingEntries
Definition: bwValues.c:333
bwOverlappingIntervals_t * bwGetOverlappingIntervals(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end)
Return bigWig entries overlapping an interval. Find all bigWig entries overlapping a range and return...
Definition: bwValues.c:563
double * bwStats(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, uint32_t nBins, enum bwStatsType type)
Determines per-interval bigWig statistics Can determine mean/min/max/coverage/standard deviation of v...
Definition: bwStats.c:530
bwOverlapIterator_t * bwOverlappingIntervalsIterator(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, uint32_t blocksPerIteration)
Creates an iterator over intervals in a bigWig file Iterators can be traversed with bwIteratorNext() ...
Definition: bwValues.c:587
bwOverlapIterator_t * bwIteratorNext(bwOverlapIterator_t *iter)
Traverses to the entries/intervals in the next group of blocks.
Definition: bwValues.c:652
int bwAppendIntervalSpanSteps(bigWigFile_t *fp, const float *values, uint32_t n)
Append to a previous block of fixed-step entries. If you previously used bwAddIntervalSpanSteps(),...
Definition: bwWrite.c:552
double * bwStatsFromFull(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, uint32_t nBins, enum bwStatsType type)
Determines per-interval bigWig statistics Can determine mean/min/max/coverage/standard deviation of v...
Definition: bwStats.c:485
int bwInit(size_t bufSize)
Initializes curl and global variables. This MUST be called before other functions (at least if you wa...
Definition: bwRead.c:38
int bbIsBigBed(const char *fname, CURLcode(*callBack)(CURL *))
Determine is a file is a bigBed file. This function will quickly check either local or remote files t...
Definition: bwRead.c:332
void bwDestroyOverlappingIntervals(bwOverlappingIntervals_t *o)
Frees space allocated by bwGetOverlappingIntervals
Definition: bwValues.c:325
char * bbGetSQL(bigWigFile_t *fp)
Returns a string containing the SQL entry (or NULL). The "auto SQL" field contains the names and valu...
Definition: bwRead.c:315
bigWigFile_t * bbOpen(const char *fname, CURLcode(*callBack)(CURL *))
Opens a local or remote bigBed file. This will open a local or remote bigBed file....
Definition: bwRead.c:397
chromList_t * bwCreateChromList(const char *const *chroms, const uint32_t *lengths, int64_t n)
Take a list of chromosome names and lengths and return a pointer to a chromList_t This MUST be run be...
Definition: bwWrite.c:22
int bwAddIntervals(bigWigFile_t *fp, const char *const *chrom, const uint32_t *start, const uint32_t *end, const float *values, uint32_t n)
Write a new block of bedGraph-like intervals to a bigWig file Adds entries of the form: chromosome st...
Definition: bwWrite.c:374
bwOverlapIterator_t * bbOverlappingEntriesIterator(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, int withString, uint32_t blocksPerIteration)
Creates an iterator over entries in a bigBed file Iterators can be traversed with bwIteratorNext() an...
Definition: bwValues.c:615
uint32_t bwGetTid(const bigWigFile_t *fp, const char *chrom)
Converts between chromosome name and ID.
Definition: bwValues.c:284
void bwIteratorDestroy(bwOverlapIterator_t *iter)
Destroys a bwOverlapIterator_t.
Definition: bwValues.c:643
int bwAppendIntervalSpans(bigWigFile_t *fp, const uint32_t *start, const float *values, uint32_t n)
Append to a previous block of variable-step entries. If you previously used bwAddIntervalSpans(),...
Definition: bwWrite.c:495
bwOverlappingIntervals_t * bwGetValues(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, int includeNA)
Return all per-base bigWig values in a given interval. Given an interval (e.g., chr1:0-100),...
Definition: bwValues.c:713
bwStatsType
Definition: bigWig.h:103
@ coverage
Definition: bigWig.h:112
@ min
Definition: bigWig.h:110
@ sum
Definition: bigWig.h:113
@ doesNotExist
Definition: bigWig.h:104
@ max
Definition: bigWig.h:109
@ mean
Definition: bigWig.h:105
@ dev
Definition: bigWig.h:108
@ stdev
Definition: bigWig.h:107
@ average
Definition: bigWig.h:106
@ cov
Definition: bigWig.h:111
bbOverlappingEntries_t * bbGetOverlappingEntries(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t end, int withString)
Return bigBed entries overlapping an interval. Find all bigBed entries overlapping a range and return...
Definition: bwValues.c:575
int bwAddIntervalSpans(bigWigFile_t *fp, const char *chrom, const uint32_t *start, uint32_t span, const float *values, uint32_t n)
Add a new block of variable-step entries to a bigWig file Adds entries for the form chromosome start ...
Definition: bwWrite.c:462
int bwAppendIntervals(bigWigFile_t *fp, const uint32_t *start, const uint32_t *end, const float *values, uint32_t n)
Append bedGraph-like intervals to a previous block of bedGraph-like intervals in a bigWig file....
Definition: bwWrite.c:434
int bwIsBigWig(const char *fname, CURLcode(*callBack)(CURL *))
Determine if a file is a bigWig file. This function will quickly check either local or remote files t...
Definition: bwRead.c:302
int bwAddIntervalSpanSteps(bigWigFile_t *fp, const char *chrom, uint32_t start, uint32_t span, uint32_t step, const float *values, uint32_t n)
Add a new block of fixed-step entries to a bigWig file Adds entries for the form value to the file....
Definition: bwWrite.c:520
void bwClose(bigWigFile_t *fp)
Closes a bigWigFile_t and frees up allocated memory This closes both bigWig and bigBed files.
Definition: bwRead.c:289
void bwCleanup(void)
The counterpart to bwInit, this cleans up curl.
Definition: bwRead.c:52
int bwWriteHdr(bigWigFile_t *bw)
Write a the header to a bigWig file. You must have already opened the output file,...
Definition: bwWrite.c:187
This structure holds the file pointers and buffers needed for raw access to local and remote files.
Definition: bigWigIO.h:38
Holds interval:str associations.
Definition: bigWig.h:234
uint32_t m
Definition: bigWig.h:236
char ** str
Definition: bigWig.h:239
uint32_t * start
Definition: bigWig.h:237
uint32_t * end
Definition: bigWig.h:238
uint32_t l
Definition: bigWig.h:235
A structure that holds everything needed to access a bigWig file.
Definition: bigWig.h:210
bwRTree_t * idx
Definition: bigWig.h:214
int isWrite
Definition: bigWig.h:216
int type
Definition: bigWig.h:217
URL_t * URL
Definition: bigWig.h:211
bwWriteBuffer_t * writeBuffer
Definition: bigWig.h:215
chromList_t * cl
Definition: bigWig.h:213
bigWigHdr_t * hdr
Definition: bigWig.h:212
The header section of a bigWig file.
Definition: bigWig.h:135
bwZoomHdr_t * zoomHdrs
Definition: bigWig.h:147
double sumData
Definition: bigWig.h:152
uint64_t extensionOffset
Definition: bigWig.h:146
uint16_t definedFieldCount
Definition: bigWig.h:142
uint64_t sqlOffset
Definition: bigWig.h:143
double minVal
Definition: bigWig.h:150
uint16_t fieldCount
Definition: bigWig.h:141
uint64_t indexOffset
Definition: bigWig.h:140
uint64_t ctOffset
Definition: bigWig.h:138
uint16_t version
Definition: bigWig.h:136
uint64_t dataOffset
Definition: bigWig.h:139
uint64_t nBasesCovered
Definition: bigWig.h:149
uint16_t nLevels
Definition: bigWig.h:137
uint64_t summaryOffset
Definition: bigWig.h:144
double sumSquared
Definition: bigWig.h:153
uint32_t bufSize
Definition: bigWig.h:145
double maxVal
Definition: bigWig.h:151
A structure to hold iterations One of intervals and entries should be used to access records from big...
Definition: bigWig.h:246
uint64_t offset
Definition: bigWig.h:251
uint32_t blocksPerIteration
Definition: bigWig.h:252
bbOverlappingEntries_t * entries
Definition: bigWig.h:256
int withString
Definition: bigWig.h:253
bwOverlappingIntervals_t * intervals
Definition: bigWig.h:255
void * data
Definition: bigWig.h:257
bigWigFile_t * bw
Definition: bigWig.h:247
uint32_t start
Definition: bigWig.h:249
uint32_t tid
Definition: bigWig.h:248
uint32_t end
Definition: bigWig.h:250
void * blocks
Definition: bigWig.h:254
Holds interval:value associations.
Definition: bigWig.h:223
uint32_t m
Definition: bigWig.h:225
float * value
Definition: bigWig.h:228
uint32_t * start
Definition: bigWig.h:226
uint32_t l
Definition: bigWig.h:224
uint32_t * end
Definition: bigWig.h:227
A node within an R-tree holding the index for data.
Definition: bwValues.h:20
Definition: bwValues.h:39
This is only needed for writing bigWig files (and won't be created otherwise) This should be removed ...
Definition: bigWig.h:185
bwZoomBuffer_t ** lastZoomBuffer
Definition: bigWig.h:201
uint64_t nEntries
Definition: bigWig.h:188
uLongf compressPsz
Definition: bigWig.h:203
uint32_t tid
Definition: bigWig.h:190
uint8_t ltype
Definition: bigWig.h:195
uint32_t span
Definition: bigWig.h:193
bwZoomBuffer_t ** firstZoomBuffer
Definition: bigWig.h:200
bwLL * firstIndexNode
Definition: bigWig.h:198
uint64_t runningWidthSum
Definition: bigWig.h:189
uint32_t blockSize
Definition: bigWig.h:187
bwLL * currentIndexNode
Definition: bigWig.h:199
uint64_t * nNodes
Definition: bigWig.h:202
uint32_t l
Definition: bigWig.h:196
uint32_t step
Definition: bigWig.h:194
uint32_t start
Definition: bigWig.h:191
uint32_t end
Definition: bigWig.h:192
void * p
Definition: bigWig.h:197
uint64_t nBlocks
Definition: bigWig.h:186
void * compressP
Definition: bigWig.h:204
BigWig files have multiple "zoom" levels, each of which has its own header. This hold those headers.
Definition: bigWig.h:122
uint64_t * dataOffset
Definition: bigWig.h:125
uint32_t * level
Definition: bigWig.h:123
bwRTree_t ** idx
Definition: bigWig.h:127
uint64_t * indexOffset
Definition: bigWig.h:126
Holds the chromosomes and their lengths.
Definition: bigWig.h:160
char ** chrom
Definition: bigWig.h:162
uint32_t * len
Definition: bigWig.h:163
int64_t nKeys
Definition: bigWig.h:161