void bwt_index(char *prefix)
{
int block_size = 10000000;
int64_t l_pac;
char *str, *str2, *str3;
str = (char*)calloc(strlen(prefix) + 10, 1);
str2 = (char*)calloc(strlen(prefix) + 10, 1);
str3 = (char*)calloc(strlen(prefix) + 10, 1);
fprintf(stderr, "[bwt_index] Building bwt-index for genome...\n");
{ // for&rev.pac .ann .amb
gzFile fp = gzopen(prefix, "r");
l_pac = bns_fasta2bntseq(fp, prefix, 0);
gzclose(fp);
}
{ // .bwt
strcpy(str, prefix); strcat(str, ".pac");
strcpy(str2, prefix); strcat(str2, ".bwt");
bwt_bwtgen2(str, str2, block_size);
}
{ // update .bwt
bwt_t *bwt;
strcpy(str, prefix); strcat(str, ".bwt");
bwt = bwt_restore_bwt(str);
bwt_bwtupdate_core(bwt);
bwt_dump_bwt(str, bwt);
bwt_destroy(bwt);
}
{ // forward.pac
gzFile fp = gzopen(prefix, "r");
l_pac = bns_fasta2bntseq(fp, prefix, 1);
gzclose(fp);
}
{ // .sa
bwt_t *bwt;
strcpy(str, prefix); strcat(str, ".bwt");
strcpy(str3, prefix); strcat(str3, ".sa");
bwt = bwt_restore_bwt(str);
bwt_cal_sa(bwt, 32);
bwt_dump_sa(str3, bwt);
bwt_destroy(bwt);
}
fprintf(stderr, "[bwt_index] Building done!\n");
free(str); free(str2); free(str3);
}
int bwa_fa2pac(int argc, char *argv[])
{
gzFile fp;
if (argc < 2) {
fprintf(stderr, "Usage: bwa fa2pac <in.fasta> [<out.prefix>]\n");
return 1;
}
fp = xzopen(argv[1], "r");
bns_fasta2bntseq(fp, (argc < 3)? argv[1] : argv[2]);
gzclose(fp);
return 0;
}
int bwa_fa2pac(int argc, char *argv[])
{
int c, for_only = 0;
gzFile fp;
while ((c = getopt(argc, argv, "f")) >= 0) {
switch (c) {
case 'f': for_only = 1; break;
}
}
if (argc == optind) {
fprintf(stderr, "Usage: bwa fa2pac [-f] <in.fasta> [<out.prefix>]\n");
return 1;
}
fp = xzopen(argv[optind], "r");
bns_fasta2bntseq(fp, (optind+1 < argc)? argv[optind+1] : argv[optind], for_only);
err_gzclose(fp);
return 0;
}
int bwa_index(int argc, char *argv[])
{
char *prefix = 0, *str, *str2, *str3;
int c, algo_type = 0, is_color = 0, is_64 = 0;
clock_t t;
int64_t l_pac;
while ((c = getopt(argc, argv, "6ca:p:")) >= 0) {
switch (c) {
case 'a': // if -a is not set, algo_type will be determined later
if (strcmp(optarg, "div") == 0) algo_type = 1;
else if (strcmp(optarg, "bwtsw") == 0) algo_type = 2;
else if (strcmp(optarg, "is") == 0) algo_type = 3;
else err_fatal(__func__, "unknown algorithm: '%s'.", optarg);
break;
case 'p': prefix = strdup(optarg); break;
case 'c': is_color = 1; break;
case '6': is_64 = 1; break;
default: return 1;
}
}
if (optind + 1 > argc) {
fprintf(stderr, "\n");
fprintf(stderr, "Usage: bwa index [-a bwtsw|is] [-c] <in.fasta>\n\n");
fprintf(stderr, "Options: -a STR BWT construction algorithm: bwtsw or is [auto]\n");
fprintf(stderr, " -p STR prefix of the index [same as fasta name]\n");
fprintf(stderr, " -6 index files named as <in.fasta>.64.* instead of <in.fasta>.* \n");
// fprintf(stderr, " -c build color-space index\n");
fprintf(stderr, "\n");
fprintf(stderr, "Warning: `-a bwtsw' does not work for short genomes, while `-a is' and\n");
fprintf(stderr, " `-a div' do not work not for long genomes. Please choose `-a'\n");
fprintf(stderr, " according to the length of the genome.\n\n");
return 1;
}
if (prefix == 0) {
prefix = malloc(strlen(argv[optind]) + 4);
strcpy(prefix, argv[optind]);
if (is_64) strcat(prefix, ".64");
}
str = (char*)calloc(strlen(prefix) + 10, 1);
str2 = (char*)calloc(strlen(prefix) + 10, 1);
str3 = (char*)calloc(strlen(prefix) + 10, 1);
if (is_color == 0) { // nucleotide indexing
gzFile fp = xzopen(argv[optind], "r");
t = clock();
fprintf(stderr, "[bwa_index] Pack FASTA... ");
l_pac = bns_fasta2bntseq(fp, prefix, 0);
fprintf(stderr, "%.2f sec\n", (float)(clock() - t) / CLOCKS_PER_SEC);
gzclose(fp);
} else { // color indexing
gzFile fp = xzopen(argv[optind], "r");
strcat(strcpy(str, prefix), ".nt");
t = clock();
fprintf(stderr, "[bwa_index] Pack nucleotide FASTA... ");
l_pac = bns_fasta2bntseq(fp, str, 0);
fprintf(stderr, "%.2f sec\n", (float)(clock() - t) / CLOCKS_PER_SEC);
gzclose(fp);
{
char *tmp_argv[3];
tmp_argv[0] = argv[0]; tmp_argv[1] = str; tmp_argv[2] = prefix;
t = clock();
fprintf(stderr, "[bwa_index] Convert nucleotide PAC to color PAC... ");
bwa_pac2cspac(3, tmp_argv);
fprintf(stderr, "%.2f sec\n", (float)(clock() - t) / CLOCKS_PER_SEC);
}
}
if (algo_type == 0) algo_type = l_pac > 50000000? 2 : 3; // set the algorithm for generating BWT
{
strcpy(str, prefix); strcat(str, ".pac");
strcpy(str2, prefix); strcat(str2, ".bwt");
t = clock();
fprintf(stderr, "[bwa_index] Construct BWT for the packed sequence...\n");
if (algo_type == 2) bwt_bwtgen(str, str2);
else if (algo_type == 1 || algo_type == 3) {
bwt_t *bwt;
bwt = bwt_pac2bwt(str, algo_type == 3);
bwt_dump_bwt(str2, bwt);
bwt_destroy(bwt);
}
fprintf(stderr, "[bwa_index] %.2f seconds elapse.\n", (float)(clock() - t) / CLOCKS_PER_SEC);
}
{
bwt_t *bwt;
strcpy(str, prefix); strcat(str, ".bwt");
t = clock();
fprintf(stderr, "[bwa_index] Update BWT... ");
bwt = bwt_restore_bwt(str);
bwt_bwtupdate_core(bwt);
bwt_dump_bwt(str, bwt);
bwt_destroy(bwt);
fprintf(stderr, "%.2f sec\n", (float)(clock() - t) / CLOCKS_PER_SEC);
}
{
gzFile fp = xzopen(argv[optind], "r");
t = clock();
fprintf(stderr, "[bwa_index] Pack forward-only FASTA... ");
l_pac = bns_fasta2bntseq(fp, prefix, 1);
fprintf(stderr, "%.2f sec\n", (float)(clock() - t) / CLOCKS_PER_SEC);
gzclose(fp);
}
{
bwt_t *bwt;
strcpy(str, prefix); strcat(str, ".bwt");
strcpy(str3, prefix); strcat(str3, ".sa");
t = clock();
fprintf(stderr, "[bwa_index] Construct SA from BWT and Occ... ");
bwt = bwt_restore_bwt(str);
bwt_cal_sa(bwt, 32);
bwt_dump_sa(str3, bwt);
bwt_destroy(bwt);
fprintf(stderr, "%.2f sec\n", (float)(clock() - t) / CLOCKS_PER_SEC);
}
free(str3); free(str2); free(str); free(prefix);
return 0;
}
