Mercurial > repos > siyuan > prada

comparison pyPRADA_1.2/tools/samtools-0.1.16/bcftools/bcf.h @ 0:acc2ca1a3ba4

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author siyuan
date Thu, 20 Feb 2014 00:44:58 -0500
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:acc2ca1a3ba4
1 /* The MIT License
2
3 Copyright (c) 2010 Broad Institute
4
5 Permission is hereby granted, free of charge, to any person obtaining
6 a copy of this software and associated documentation files (the
7 "Software"), to deal in the Software without restriction, including
8 without limitation the rights to use, copy, modify, merge, publish,
9 distribute, sublicense, and/or sell copies of the Software, and to
10 permit persons to whom the Software is furnished to do so, subject to
11 the following conditions:
12
13 The above copyright notice and this permission notice shall be
14 included in all copies or substantial portions of the Software.
15
16 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
17 EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
18 MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
19 NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
20 BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
21 ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
22 CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
23 SOFTWARE.
24 */
25
26 /* Contact: Heng Li <lh3@live.co.uk> */
27
28 #ifndef BCF_H
29 #define BCF_H
30
31 #include <stdint.h>
32 #include <zlib.h>
33
34 #ifndef BCF_LITE
35 #include "bgzf.h"
36 typedef BGZF *bcfFile;
37 #else
38 typedef gzFile bcfFile;
39 #define bgzf_open(fn, mode) gzopen(fn, mode)
40 #define bgzf_fdopen(fd, mode) gzdopen(fd, mode)
41 #define bgzf_close(fp) gzclose(fp)
42 #define bgzf_read(fp, buf, len) gzread(fp, buf, len)
43 #define bgzf_write(fp, buf, len)
44 #define bgzf_flush(fp)
45 #endif
46
47 /*
48 A member in the structs below is said to "primary" if its content
49 cannot be inferred from other members in any of structs below; a
50 member is said to be "derived" if its content can be derived from
51 other members. For example, bcf1_t::str is primary as this comes from
52 the input data, while bcf1_t::info is derived as it can always be
53 correctly set if we know bcf1_t::str. Derived members are for quick
54 access to the content and must be synchronized with the primary data.
55 */
56
57 typedef struct {
58 uint32_t fmt; // format of the block, set by bcf_str2int().
59 int len; // length of data for each individual
60 void *data; // concatenated data
61 // derived info: fmt, len (<-bcf1_t::fmt)
62 } bcf_ginfo_t;
63
64 typedef struct {
65 int32_t tid, pos; // refID and 0-based position
66 int32_t l_str, m_str; // length and the allocated size of ->str
67 float qual; // SNP quality
68 char *str; // concatenated string of variable length strings in VCF (from col.2 to col.7)
69 char *ref, *alt, *flt, *info, *fmt; // they all point to ->str; no memory allocation
70 int n_gi, m_gi; // number and the allocated size of geno fields
71 bcf_ginfo_t *gi; // array of geno fields
72 int n_alleles, n_smpl; // number of alleles and samples
73 // derived info: ref, alt, flt, info, fmt (<-str), n_gi (<-fmt), n_alleles (<-alt), n_smpl (<-bcf_hdr_t::n_smpl)
74 } bcf1_t;
75
76 typedef struct {
77 int32_t n_ref, n_smpl; // number of reference sequences and samples
78 int32_t l_nm; // length of concatenated sequence names; 0 padded
79 int32_t l_smpl; // length of concatenated sample names; 0 padded
80 int32_t l_txt; // length of header text (lines started with ##)
81 char *name, *sname, *txt; // concatenated sequence names, sample names and header text
82 char **ns, **sns; // array of sequence and sample names; point to name and sname, respectively
83 // derived info: n_ref (<-name), n_smpl (<-sname), ns (<-name), sns (<-sname)
84 } bcf_hdr_t;
85
86 typedef struct {
87 int is_vcf; // if the file in operation is a VCF
88 void *v; // auxillary data structure for VCF
89 bcfFile fp; // file handler for BCF
90 } bcf_t;
91
92 struct __bcf_idx_t;
93 typedef struct __bcf_idx_t bcf_idx_t;
94
95 #ifdef __cplusplus
96 extern "C" {
97 #endif
98
99 // open a BCF file; for BCF file only
100 bcf_t *bcf_open(const char *fn, const char *mode);
101 // close file
102 int bcf_close(bcf_t *b);
103 // read one record from BCF; return -1 on end-of-file, and <-1 for errors
104 int bcf_read(bcf_t *bp, const bcf_hdr_t *h, bcf1_t *b);
105 // call this function if b->str is changed
106 int bcf_sync(bcf1_t *b);
107 // write a BCF record
108 int bcf_write(bcf_t *bp, const bcf_hdr_t *h, const bcf1_t *b);
109 // read the BCF header; BCF only
110 bcf_hdr_t *bcf_hdr_read(bcf_t *b);
111 // write the BCF header
112 int bcf_hdr_write(bcf_t *b, const bcf_hdr_t *h);
113 // set bcf_hdr_t::ns and bcf_hdr_t::sns
114 int bcf_hdr_sync(bcf_hdr_t *b);
115 // destroy the header
116 void bcf_hdr_destroy(bcf_hdr_t *h);
117 // destroy a record
118 int bcf_destroy(bcf1_t *b);
119 // BCF->VCF conversion
120 char *bcf_fmt(const bcf_hdr_t *h, bcf1_t *b);
121 // append more info
122 int bcf_append_info(bcf1_t *b, const char *info, int l);
123 // copy
124 int bcf_cpy(bcf1_t *r, const bcf1_t *b);
125
126 // open a VCF or BCF file if "b" is set in "mode"
127 bcf_t *vcf_open(const char *fn, const char *mode);
128 // close a VCF/BCF file
129 int vcf_close(bcf_t *bp);
130 // read the VCF/BCF header
131 bcf_hdr_t *vcf_hdr_read(bcf_t *bp);
132 // read the sequence dictionary from a separate file; required for VCF->BCF conversion
133 int vcf_dictread(bcf_t *bp, bcf_hdr_t *h, const char *fn);
134 // read a VCF/BCF record; return -1 on end-of-file and <-1 for errors
135 int vcf_read(bcf_t *bp, bcf_hdr_t *h, bcf1_t *b);
136 // write the VCF header
137 int vcf_hdr_write(bcf_t *bp, const bcf_hdr_t *h);
138 // write a VCF record
139 int vcf_write(bcf_t *bp, bcf_hdr_t *h, bcf1_t *b);
140
141 // keep the first n alleles and discard the rest
142 int bcf_shrink_alt(bcf1_t *b, int n);
143 // convert GL to PL
144 int bcf_gl2pl(bcf1_t *b);
145 // if the site is an indel
146 int bcf_is_indel(const bcf1_t *b);
147 bcf_hdr_t *bcf_hdr_subsam(const bcf_hdr_t *h0, int n, char *const* samples, int *list);
148 int bcf_subsam(int n_smpl, int *list, bcf1_t *b);
149 // move GT to the first FORMAT field
150 int bcf_fix_gt(bcf1_t *b);
151 // update PL generated by old samtools
152 int bcf_fix_pl(bcf1_t *b);
153
154 // string hash table
155 void *bcf_build_refhash(bcf_hdr_t *h);
156 void bcf_str2id_destroy(void *_hash);
157 void bcf_str2id_thorough_destroy(void *_hash);
158 int bcf_str2id_add(void *_hash, const char *str);
159 int bcf_str2id(void *_hash, const char *str);
160 void *bcf_str2id_init();
161
162 // indexing related functions
163 int bcf_idx_build(const char *fn);
164 uint64_t bcf_idx_query(const bcf_idx_t *idx, int tid, int beg);
165 int bcf_parse_region(void *str2id, const char *str, int *tid, int *begin, int *end);
166 bcf_idx_t *bcf_idx_load(const char *fn);
167 void bcf_idx_destroy(bcf_idx_t *idx);
168
169 #ifdef __cplusplus
170 }
171 #endif
172
173 static inline uint32_t bcf_str2int(const char *str, int l)
174 {
175 int i;
176 uint32_t x = 0;
177 for (i = 0; i < l && i < 4; ++i) {
178 if (str[i] == 0) return x;
179 x = x<<8 | str[i];
180 }
181 return x;
182 }
183
184 #endif