6 #include "sam_header.h"
10 KHASH_SET_INIT_STR(rg)
12 typedef khash_t(rg) *rghash_t;
14 rghash_t g_rghash = 0;
15 static int g_min_mapQ = 0, g_flag_on = 0, g_flag_off = 0;
16 static char *g_library, *g_rg;
17 static int g_sol2sanger_tbl[128];
19 static void sol2sanger(bam1_t *b)
22 uint8_t *qual = bam1_qual(b);
23 if (g_sol2sanger_tbl[30] == 0) {
24 for (l = 0; l != 128; ++l) {
25 g_sol2sanger_tbl[l] = (int)(10.0 * log(1.0 + pow(10.0, (l - 64 + 33) / 10.0)) / log(10.0) + .499);
26 if (g_sol2sanger_tbl[l] >= 93) g_sol2sanger_tbl[l] = 93;
29 for (l = 0; l < b->core.l_qseq; ++l) {
32 qual[l] = g_sol2sanger_tbl[q];
36 static inline int __g_skip_aln(const bam_header_t *h, const bam1_t *b)
38 if (b->core.qual < g_min_mapQ || ((b->core.flag & g_flag_on) != g_flag_on) || (b->core.flag & g_flag_off))
40 if (g_rg || g_rghash) {
41 uint8_t *s = bam_aux_get(b, "RG");
43 if (g_rg) return (strcmp(g_rg, (char*)(s + 1)) == 0)? 0 : 1;
45 khint_t k = kh_get(rg, g_rghash, (char*)(s + 1));
46 return (k != kh_end(g_rghash))? 0 : 1;
51 const char *p = bam_get_library((bam_header_t*)h, b);
52 return (p && strcmp(p, g_library) == 0)? 0 : 1;
57 // callback function for bam_fetch()
58 static int view_func(const bam1_t *b, void *data)
60 if (!__g_skip_aln(((samfile_t*)data)->header, b))
61 samwrite((samfile_t*)data, b);
65 static int usage(int is_long_help);
67 int main_samview(int argc, char *argv[])
69 int c, is_header = 0, is_header_only = 0, is_bamin = 1, ret = 0, is_uncompressed = 0, is_bamout = 0, slx2sngr = 0;
70 int of_type = BAM_OFDEC, is_long_help = 0;
71 samfile_t *in = 0, *out = 0;
72 char in_mode[5], out_mode[5], *fn_out = 0, *fn_list = 0, *fn_ref = 0, *fn_rg = 0;
74 /* parse command-line options */
75 strcpy(in_mode, "r"); strcpy(out_mode, "w");
76 while ((c = getopt(argc, argv, "Sbt:hHo:q:f:F:ul:r:xX?T:CR:")) >= 0) {
78 case 'C': slx2sngr = 1; break;
79 case 'S': is_bamin = 0; break;
80 case 'b': is_bamout = 1; break;
81 case 't': fn_list = strdup(optarg); is_bamin = 0; break;
82 case 'h': is_header = 1; break;
83 case 'H': is_header_only = 1; break;
84 case 'o': fn_out = strdup(optarg); break;
85 case 'f': g_flag_on = strtol(optarg, 0, 0); break;
86 case 'F': g_flag_off = strtol(optarg, 0, 0); break;
87 case 'q': g_min_mapQ = atoi(optarg); break;
88 case 'u': is_uncompressed = 1; break;
89 case 'l': g_library = strdup(optarg); break;
90 case 'r': g_rg = strdup(optarg); break;
91 case 'R': fn_rg = strdup(optarg); break;
92 case 'x': of_type = BAM_OFHEX; break;
93 case 'X': of_type = BAM_OFSTR; break;
94 case '?': is_long_help = 1; break;
95 case 'T': fn_ref = strdup(optarg); is_bamin = 0; break;
96 default: return usage(is_long_help);
99 if (is_uncompressed) is_bamout = 1;
100 if (is_header_only) is_header = 1;
101 if (is_bamout) strcat(out_mode, "b");
103 if (of_type == BAM_OFHEX) strcat(out_mode, "x");
104 else if (of_type == BAM_OFSTR) strcat(out_mode, "X");
106 if (is_bamin) strcat(in_mode, "b");
107 if (is_header) strcat(out_mode, "h");
108 if (is_uncompressed) strcat(out_mode, "u");
109 if (argc == optind) return usage(is_long_help); // potential memory leak...
111 // read the list of read groups
116 g_rghash = kh_init(rg);
117 fp_rg = fopen(fn_rg, "r");
118 while (!feof(fp_rg) && fscanf(fp_rg, "%s", buf) > 0) // this is not a good style, but bear me...
119 kh_put(rg, g_rghash, strdup(buf), &ret); // we'd better check duplicates...
123 // generate the fn_list if necessary
124 if (fn_list == 0 && fn_ref) fn_list = samfaipath(fn_ref);
125 // open file handlers
126 if ((in = samopen(argv[optind], in_mode, fn_list)) == 0) {
127 fprintf(stderr, "[main_samview] fail to open file for reading.\n");
130 if (in->header == 0) {
131 fprintf(stderr, "[main_samview] fail to read the header.\n");
134 if ((out = samopen(fn_out? fn_out : "-", out_mode, in->header)) == 0) {
135 fprintf(stderr, "[main_samview] fail to open file for writing.\n");
138 if (is_header_only) goto view_end; // no need to print alignments
140 if (argc == optind + 1) { // convert/print the entire file
141 bam1_t *b = bam_init1();
143 while ((r = samread(in, b)) >= 0) { // read one alignment from `in'
144 if (!__g_skip_aln(in->header, b)) {
145 if (slx2sngr) sol2sanger(b);
146 samwrite(out, b); // write the alignment to `out'
149 if (r < -1) fprintf(stderr, "[main_samview] truncated file.\n");
151 } else { // retrieve alignments in specified regions
153 bam_index_t *idx = 0;
154 if (is_bamin) idx = bam_index_load(argv[optind]); // load BAM index
155 if (idx == 0) { // index is unavailable
156 fprintf(stderr, "[main_samview] random alignment retrieval only works for indexed BAM files.\n");
160 for (i = optind + 1; i < argc; ++i) {
162 bam_parse_region(in->header, argv[i], &tid, &beg, &end); // parse a region in the format like `chr2:100-200'
163 if (tid < 0) { // reference name is not found
164 fprintf(stderr, "[main_samview] fail to get the reference name. Continue anyway.\n");
167 bam_fetch(in->x.bam, idx, tid, beg, end, out, view_func); // fetch alignments
169 bam_index_destroy(idx); // destroy the BAM index
173 // close files, free and return
174 free(fn_list); free(fn_ref); free(fn_out); free(g_library); free(g_rg); free(fn_rg);
177 for (k = 0; k < kh_end(g_rghash); ++k)
178 if (kh_exist(g_rghash, k)) free((char*)kh_key(g_rghash, k));
179 kh_destroy(rg, g_rghash);
186 static int usage(int is_long_help)
188 fprintf(stderr, "\n");
189 fprintf(stderr, "Usage: samtools view [options] <in.bam>|<in.sam> [region1 [...]]\n\n");
190 fprintf(stderr, "Options: -b output BAM\n");
191 fprintf(stderr, " -h print header for the SAM output\n");
192 fprintf(stderr, " -H print header only (no alignments)\n");
193 fprintf(stderr, " -S input is SAM\n");
194 fprintf(stderr, " -u uncompressed BAM output (force -b)\n");
195 fprintf(stderr, " -x output FLAG in HEX (samtools-C specific)\n");
196 fprintf(stderr, " -X output FLAG in string (samtools-C specific)\n");
197 fprintf(stderr, " -t FILE list of reference names and lengths (force -S) [null]\n");
198 fprintf(stderr, " -T FILE reference sequence file (force -S) [null]\n");
199 fprintf(stderr, " -o FILE output file name [stdout]\n");
200 fprintf(stderr, " -R FILE list of read groups to be outputted [null]\n");
201 fprintf(stderr, " -f INT required flag, 0 for unset [0]\n");
202 fprintf(stderr, " -F INT filtering flag, 0 for unset [0]\n");
203 fprintf(stderr, " -q INT minimum mapping quality [0]\n");
204 fprintf(stderr, " -l STR only output reads in library STR [null]\n");
205 fprintf(stderr, " -r STR only output reads in read group STR [null]\n");
206 fprintf(stderr, " -? longer help\n");
207 fprintf(stderr, "\n");
209 fprintf(stderr, "Notes:\n\
211 1. By default, this command assumes the file on the command line is in\n\
212 the BAM format and it prints the alignments in SAM. If `-t' is\n\
213 applied, the input file is assumed to be in the SAM format. The\n\
214 file supplied with `-t' is SPACE/TAB delimited with the first two\n\
215 fields of each line consisting of the reference name and the\n\
216 corresponding sequence length. The `.fai' file generated by `faidx'\n\
217 can be used here. This file may be empty if reads are unaligned.\n\
219 2. SAM->BAM conversion: `samtools view -bT ref.fa in.sam.gz'.\n\
221 3. BAM->SAM conversion: `samtools view in.bam'.\n\
223 4. A region should be presented in one of the following formats:\n\
224 `chr1', `chr2:1,000' and `chr3:1000-2,000'. When a region is\n\
225 specified, the input alignment file must be an indexed BAM file.\n\
227 5. Option `-u' is preferred over `-b' when the output is piped to\n\
228 another samtools command.\n\
230 6. In a string FLAG, each character represents one bit with\n\
231 p=0x1 (paired), P=0x2 (properly paired), u=0x4 (unmapped),\n\
232 U=0x8 (mate unmapped), r=0x10 (reverse), R=0x20 (mate reverse)\n\
233 1=0x40 (first), 2=0x80 (second), s=0x100 (not primary), \n\
234 f=0x200 (failure) and d=0x400 (duplicate). Note that `-x' and\n\
235 `-X' are samtools-C specific. Picard and older samtools do not\n\
236 support HEX or string flags.\n\
241 int main_import(int argc, char *argv[])
246 fprintf(stderr, "Usage: bamtk import <in.ref_list> <in.sam> <out.bam>\n");
250 argv2 = calloc(6, sizeof(char*));
251 argv2[0] = "import", argv2[1] = "-o", argv2[2] = argv[3], argv2[3] = "-bt", argv2[4] = argv[1], argv2[5] = argv[2];
252 ret = main_samview(argc2, argv2);