synced_bcf_reader.c File Reference

#include <stdio.h>
#include <unistd.h>
#include <string.h>
#include <limits.h>
#include <errno.h>
#include <ctype.h>
#include <sys/stat.h>
#include "htslib/synced_bcf_reader.h"
#include "htslib/kseq.h"
#include "htslib/khash_str2int.h"

Classes
struct	region1_t
struct	_region_t

Macros
#define	MAX_CSI_COOR   0x7fffffff

Typedefs
typedef struct _region_t	region_t

Functions
char *	bcf_sr_strerror (int errnum)
int	bcf_sr_set_regions (bcf_srs_t *readers, const char *regions, int is_file)
int	bcf_sr_set_targets (bcf_srs_t *readers, const char *targets, int is_file, int alleles)
int	bcf_sr_add_reader (bcf_srs_t *files, const char *fname)
bcf_srs_t *	bcf_sr_init (void)
void	bcf_sr_destroy (bcf_srs_t *files)
void	bcf_sr_remove_reader (bcf_srs_t *files, int i)
void	debug_buffer (FILE *fp, bcf_sr_t *reader)
void	debug_buffers (FILE *fp, bcf_srs_t *files)
int	_reader_next_line (bcf_srs_t *files)
int	bcf_sr_next_line (bcf_srs_t *files)
int	bcf_sr_seek (bcf_srs_t *readers, const char *seq, int pos)
int	bcf_sr_set_samples (bcf_srs_t *files, const char *fname, int is_file)
bcf_sr_regions_t *	bcf_sr_regions_init (const char *regions, int is_file, int ichr, int ifrom, int ito)
void	bcf_sr_regions_destroy (bcf_sr_regions_t *reg)
int	bcf_sr_regions_seek (bcf_sr_regions_t *reg, const char *seq)
int	bcf_sr_regions_next (bcf_sr_regions_t *reg)
int	bcf_sr_regions_overlap (bcf_sr_regions_t *reg, const char *seq, int start, int end)
void	bcf_sr_regions_flush (bcf_sr_regions_t *reg)

Macro Definition Documentation

#define MAX_CSI_COOR   0x7fffffff

Typedef Documentation

typedef struct _region_t region_t

Function Documentation

int _reader_next_line

(

bcf_srs_t *

files

)

int bcf_sr_add_reader	(	bcf_srs_t *	readers,
		const char *	fname
	)

bcf_sr_add_reader() - open new reader : holder of the open readers : the VCF file

Returns 1 if the call succeeded, or 0 on error.

See also the bcf_srs_t data structure for parameters controlling the reader's logic.

void bcf_sr_destroy

(

bcf_srs_t *

readers

)

Destroy bcf_srs_t struct

bcf_srs_t* bcf_sr_init

(

void

)

Init bcf_srs_t struct

int bcf_sr_next_line

(

bcf_srs_t *

readers

)

bcf_sr_next_line() - the iterator : holder of the open readers

Returns the number of readers which have the current line (bcf_sr_t.buffer[0]) set at this position. Use the bcf_sr_has_line macro to determine which of the readers are set.

void bcf_sr_regions_destroy

(

bcf_sr_regions_t *

reg

)

void bcf_sr_regions_flush

(

bcf_sr_regions_t *

reg

)

bcf_sr_regions_t* bcf_sr_regions_init	(	const char *	regions,
		int	is_file,
		int	ichr,
		int	ifrom,
		int	ito
	)

int bcf_sr_regions_next

(

bcf_sr_regions_t *

reg

)

int bcf_sr_regions_overlap	(	bcf_sr_regions_t *	reg,
		const char *	seq,
		int	start,
		int	end
	)

int bcf_sr_regions_seek	(	bcf_sr_regions_t *	reg,
		const char *	seq
	)

void bcf_sr_remove_reader	(	bcf_srs_t *	files,
		int	i
	)

int bcf_sr_seek	(	bcf_srs_t *	readers,
		const char *	seq,
		int	pos
	)

bcf_sr_seek() - set all readers to selected position : sequence name; NULL to seek to start : 0-based coordinate

int bcf_sr_set_regions	(	bcf_srs_t *	readers,
		const char *	regions,
		int	is_file
	)

int bcf_sr_set_samples	(	bcf_srs_t *	readers,
		const char *	samples,
		int	is_file
	)

bcf_sr_set_samples() - sets active samples : holder of the open readers : this can be one of: file name with one sample per line; or column-separated list of samples; or '-' for a list of samples shared by all files. If first character is the exclamation mark, all but the listed samples are included. : 0: list of samples; 1: file with sample names

Returns 1 if the call succeeded, or 0 on error.

int bcf_sr_set_targets	(	bcf_srs_t *	readers,
		const char *	targets,
		int	is_file,
		int	alleles
	)

bcf_sr_set_targets(), bcf_sr_set_regions() - init targets/regions : holder of the open readers : list of regions, one-based and inclusive. : 0: targets is a comma-separated list of regions (chr,chr:from-to) 1: targets is a tabix indexed file with a list of regions (<chr,pos> or <chr,from,to>)

Returns 0 if the call succeeded, or -1 on error.

Both functions behave the same way, unlisted positions will be skipped by bcf_sr_next_line(). However, there is an important difference: regions use index to jump to desired positions while targets streams the whole files and merely skip unlisted positions.

Moreover, bcf_sr_set_targets() accepts an optional parameter $alleles which is intepreted as a 1-based column index in the tab-delimited file where alleles are listed. This in principle enables to perform the COLLAPSE_* logic also with tab-delimited files. However, the current implementation considers the alleles merely as a suggestion for prioritizing one of possibly duplicate VCF lines. It is up to the caller to examine targets->als if perfect match is sought after. Note that the duplicate positions in targets file are currently not supported. Targets (but not regions) can be prefixed with "^" to request logical complement, for example "^X,Y,MT" indicates that sequences X, Y and MT should be skipped.

char* bcf_sr_strerror

(

int

errnum

)

void debug_buffer	(	FILE *	fp,
		bcf_sr_t *	reader
	)

void debug_buffers	(	FILE *	fp,
		bcf_srs_t *	files
	)