// Convert the given reference nucleotide FASTA and GFF file to a protein FASTA file
int writeIndexProtein(const char * passPrefix, const char * passProName, const char * passAnnName) {
struct CDS currentCDS;
gzFile inputSeqPtr;
FILE * inputAnnPtr, * outputPtr;
char * outputBuffer;
kseq_t * seq;
unsigned long outputSize, currentLine;
// Prepare file handles
inputSeqPtr = xzopen(passPrefix, "r");
inputAnnPtr = fopen(passAnnName, "r");
outputPtr = fopen(passProName, "w");
// Read in 1st sequence data
seq = kseq_init(inputSeqPtr);
kseq_read(seq);
// Iterate through each CDS sequence in the annotation file
currentLine = 0;
while (getNextCDS(inputAnnPtr, ¤tCDS, ¤tLine)) {
convertToAA(seq->seq.s, ¤tCDS, &outputBuffer, &outputSize);
fprintf(outputPtr, ">%d\n%.*s\n", currentLine, outputSize, outputBuffer);
free(outputBuffer);
}
// Close files
fflush(outputPtr);
err_gzclose(inputSeqPtr);
err_fclose(inputAnnPtr);
err_fclose(outputPtr);
return 0;
}
void bns_dump(const bntseq_t *bns, const char *prefix)
{
char str[1024];
FILE *fp;
int i;
{ // dump .ann
strcpy(str, prefix); strcat(str, ".ann");
fp = xopen(str, "w");
err_fprintf(fp, "%lld %d %u\n", (long long)bns->l_pac, bns->n_seqs, bns->seed);
for (i = 0; i != bns->n_seqs; ++i) {
bntann1_t *p = bns->anns + i;
err_fprintf(fp, "%d %s", p->gi, p->name);
if (p->anno[0]) err_fprintf(fp, " %s\n", p->anno);
else err_fprintf(fp, "\n");
err_fprintf(fp, "%lld %d %d\n", (long long)p->offset, p->len, p->n_ambs);
}
err_fflush(fp);
err_fclose(fp);
}
{ // dump .amb
strcpy(str, prefix); strcat(str, ".amb");
fp = xopen(str, "w");
err_fprintf(fp, "%lld %d %u\n", (long long)bns->l_pac, bns->n_seqs, bns->n_holes);
for (i = 0; i != bns->n_holes; ++i) {
bntamb1_t *p = bns->ambs + i;
err_fprintf(fp, "%lld %d %c\n", (long long)p->offset, p->len, p->amb);
}
err_fflush(fp);
err_fclose(fp);
}
}
int main(int argc, char *argv[])
{
int i, ret;
double t_real;
kstring_t pg = {0,0,0};
t_real = realtime();
ksprintf(&pg, "@PG\tID:bwa\tPN:bwa\tVN:%s\tCL:%s", PACKAGE_VERSION, argv[0]);
for (i = 1; i < argc; ++i) ksprintf(&pg, " %s", argv[i]);
bwa_pg = pg.s;
if (argc < 2) return usage();
if (strcmp(argv[1], "fa2pac") == 0) ret = bwa_fa2pac(argc-1, argv+1);
else if (strcmp(argv[1], "pac2bwt") == 0) ret = bwa_pac2bwt(argc-1, argv+1);
else if (strcmp(argv[1], "pac2bwtgen") == 0) ret = bwt_bwtgen_main(argc-1, argv+1);
else if (strcmp(argv[1], "bwtupdate") == 0) ret = bwa_bwtupdate(argc-1, argv+1);
else if (strcmp(argv[1], "bwt2sa") == 0) ret = bwa_bwt2sa(argc-1, argv+1);
else if (strcmp(argv[1], "index") == 0) ret = bwa_index(argc-1, argv+1);
else if (strcmp(argv[1], "aln") == 0) ret = bwa_aln(argc-1, argv+1);
else if (strcmp(argv[1], "samse") == 0) ret = bwa_sai2sam_se(argc-1, argv+1);
else if (strcmp(argv[1], "sampe") == 0) ret = bwa_sai2sam_pe(argc-1, argv+1);
else if (strcmp(argv[1], "bwtsw2") == 0) ret = bwa_bwtsw2(argc-1, argv+1);
else if (strcmp(argv[1], "dbwtsw") == 0) ret = bwa_bwtsw2(argc-1, argv+1);
else if (strcmp(argv[1], "bwasw") == 0) ret = bwa_bwtsw2(argc-1, argv+1);
else if (strcmp(argv[1], "fastmap") == 0) ret = main_fastmap(argc-1, argv+1);
else if (strcmp(argv[1], "mem") == 0) ret = main_mem(argc-1, argv+1);
else if (strcmp(argv[1], "pemerge") == 0) ret = main_pemerge(argc-1, argv+1);
else {
fprintf(stderr, "[main] unrecognized command '%s'\n", argv[1]);
return 1;
}
#ifdef USE_HTSLIB
if (strcmp(argv[1], "mem") != 0) {
err_fflush(stdout);
err_fclose(stdout);
}
#else
err_fflush(stdout);
err_fclose(stdout);
#endif
if (ret == 0) {
fprintf(stderr, "[%s] Version: %s\n", __func__, PACKAGE_VERSION);
fprintf(stderr, "[%s] CMD:", __func__);
for (i = 0; i < argc; ++i)
fprintf(stderr, " %s", argv[i]);
fprintf(stderr, "\n[%s] Real time: %.3f sec; CPU: %.3f sec\n", __func__, realtime() - t_real, cputime());
}
free(bwa_pg);
return ret;
}
int main(int argc, char *argv[])
{
extern char *bwa_pg;
int i, ret;
double t_real;
kstring_t pg = {0,0,0};
t_real = realtime();
ksprintf(&pg, "@PG\tID:biscuit\tPN:biscuit\tVN:%s\tCL:%s", PACKAGE_VERSION, argv[0]);
for (i = 1; i < argc; ++i) ksprintf(&pg, " %s", argv[i]);
bwa_pg = pg.s;
if (argc < 2) return usage();
if (strcmp(argv[1], "index") == 0) ret = main_biscuit_index(argc-1, argv+1);
else if (strcmp(argv[1], "align") == 0) ret = main_align(argc-1, argv+1);
else if (strcmp(argv[1], "pileup") == 0) ret = main_pileup(argc-1, argv+1);
else if (strcmp(argv[1], "somatic") == 0) ret = main_somatic(argc-1, argv+1);
else {
fprintf(stderr, "[main] unrecognized command '%s'\n", argv[1]);
return 1;
}
err_fflush(stdout);
err_fclose(stdout);
if (ret == 0) {
fprintf(stderr, "[%s] Version: %s\n", __func__, PACKAGE_VERSION);
fprintf(stderr, "[%s] CMD:", __func__);
for (i = 0; i < argc; ++i)
fprintf(stderr, " %s", argv[i]);
fprintf(stderr, "\n[%s] Real time: %.3f sec; CPU: %.3f sec\n", __func__, realtime() - t_real, cputime());
}
free(bwa_pg);
return ret;
}
int64_t bns_fasta2bntseq(gzFile fp_fa, const char *prefix, int for_only)
{
extern void seq_reverse(int len, ubyte_t *seq, int is_comp); // in bwaseqio.c
kseq_t *seq;
char name[1024];
bntseq_t *bns;
uint8_t *pac = 0;
int32_t m_seqs, m_holes;
int64_t ret = -1, m_pac, l;
bntamb1_t *q;
FILE *fp;
// initialization
seq = kseq_init(fp_fa);
bns = (bntseq_t*)calloc(1, sizeof(bntseq_t));
bns->seed = 11; // fixed seed for random generator
srand48(bns->seed);
m_seqs = m_holes = 8; m_pac = 0x10000;
bns->anns = (bntann1_t*)calloc(m_seqs, sizeof(bntann1_t));
bns->ambs = (bntamb1_t*)calloc(m_holes, sizeof(bntamb1_t));
pac = calloc(m_pac/4, 1);
q = bns->ambs;
strcpy(name, prefix); strcat(name, ".pac");
fp = xopen(name, "wb");
// read sequences
while (kseq_read(seq) >= 0) pac = add1(seq, bns, pac, &m_pac, &m_seqs, &m_holes, &q);
if (!for_only) { // add the reverse complemented sequence
m_pac = (bns->l_pac * 2 + 3) / 4 * 4;
pac = realloc(pac, m_pac/4);
memset(pac + (bns->l_pac+3)/4, 0, (m_pac - (bns->l_pac+3)/4*4) / 4);
for (l = bns->l_pac - 1; l >= 0; --l, ++bns->l_pac)
_set_pac(pac, bns->l_pac, 3-_get_pac(pac, l));
}
ret = bns->l_pac;
{ // finalize .pac file
ubyte_t ct;
err_fwrite(pac, 1, (bns->l_pac>>2) + ((bns->l_pac&3) == 0? 0 : 1), fp);
// the following codes make the pac file size always (l_pac/4+1+1)
if (bns->l_pac % 4 == 0) {
ct = 0;
err_fwrite(&ct, 1, 1, fp);
}
ct = bns->l_pac % 4;
err_fwrite(&ct, 1, 1, fp);
// close .pac file
err_fflush(fp);
err_fclose(fp);
}
bns_dump(bns, prefix);
bns_destroy(bns);
kseq_destroy(seq);
free(pac);
return ret;
}
int64_t dump_forward_pac(gzFile fp_fa, const char *prefix)
{
extern void seq_reverse(int len, ubyte_t *seq, int is_comp); // in bwaseqio.c
kseq_t *seq;
char name[1024];
bntseq_t *bns;
uint8_t *pac = 0;
int32_t m_seqs, m_holes;
int64_t ret = -1, m_pac;
bntamb1_t *q;
FILE *fp;
// initialization
seq = kseq_init(fp_fa);
bns = (bntseq_t*)calloc(1, sizeof(bntseq_t));
bns->seed = 11; // fixed seed for random generator
srand48(bns->seed);
m_seqs = m_holes = 8; m_pac = 0x10000;
bns->anns = (bntann1_t*)calloc(m_seqs, sizeof(bntann1_t));
bns->ambs = (bntamb1_t*)calloc(m_holes, sizeof(bntamb1_t));
pac = calloc(m_pac/4, 1);
q = bns->ambs;
strcpy(name, prefix); strcat(name, ".bis.pac");
fp = xopen(name, "wb");
// read sequences
while (kseq_read(seq) >= 0) pac = add1(seq, bns, pac, &m_pac, &m_seqs, &m_holes, &q);
ret = bns->l_pac;
{ // finalize .pac file
ubyte_t ct;
err_fwrite(pac, 1, (bns->l_pac>>2) + ((bns->l_pac&3) == 0? 0 : 1), fp);
// the following codes make the pac file size always (l_pac/4+1+1)
if (bns->l_pac % 4 == 0) {
ct = 0;
err_fwrite(&ct, 1, 1, fp);
}
ct = bns->l_pac % 4;
err_fwrite(&ct, 1, 1, fp);
// close .pac file
err_fflush(fp);
err_fclose(fp);
}
/* re-dump forward bis bns, otherwise the .bis.ann and .bis.amb have twice as long pac */
/* strcpy(name, prefix); strcat(name, ".bis"); */
/* bis_bns_dump(bns, prefix); */
bns_destroy(bns);
kseq_destroy(seq);
free(pac);
return ret;
}
void bns_destroy(bntseq_t *bns)
{
if (bns == 0) return;
else {
int i;
if (bns->fp_pac) err_fclose(bns->fp_pac);
free(bns->ambs);
for (i = 0; i < bns->n_seqs; ++i) {
free(bns->anns[i].name);
free(bns->anns[i].anno);
}
free(bns->anns);
free(bns);
}
}
int main(int argc, char *argv[])
{
int i, ret;
double t_real;
t_real = realtime();
if (argc < 2) return usage();
/* if (strcmp(argv[1], "fa2pac") == 0) ret = bwa_fa2pac(argc-1, argv+1);
*/
/* else if (strcmp(argv[1], "pac2bwt") == 0) ret = bwa_pac2bwt(argc-1, argv+1);
* else if (strcmp(argv[1], "pac2bwtgen") == 0) ret = bwt_bwtgen_main(argc-1, argv+1);
* else if (strcmp(argv[1], "bwtupdate") == 0) ret = bwa_bwtupdate(argc-1, argv+1);
* else if (strcmp(argv[1], "bwt2sa") == 0) ret = bwa_bwt2sa(argc-1, argv+1);
*/
else if (strcmp(argv[1], "index") == 0) ret = bwa_index_main(argc-1, argv+1);
else if (strcmp(argv[1], "aln") == 0) ret = bwa_aln(argc-1, argv+1);
/* else if (strcmp(argv[1], "samse") == 0) ret = bwa_sai2sam_se(argc-1, argv+1);
*/
/* else if (strcmp(argv[1], "sw") == 0) ret = bwa_stdsw(argc-1, argv+1);
*
* else if (strcmp(argv[1], "sampe") == 0) ret = bwa_sai2sam_pe(argc-1, argv+1);
* else if (strcmp(argv[1], "pac2cspac") == 0) ret = bwa_pac2cspac(argc-1, argv+1);
* else if (strcmp(argv[1], "stdsw") == 0) ret = bwa_stdsw(argc-1, argv+1);
* else if (strcmp(argv[1], "bwtsw2") == 0) ret = bwa_bwtsw2(argc-1, argv+1);
* else if (strcmp(argv[1], "dbwtsw") == 0) ret = bwa_bwtsw2(argc-1, argv+1);
* else if (strcmp(argv[1], "bwasw") == 0) ret = bwa_bwtsw2(argc-1, argv+1);
* else if (strcmp(argv[1], "fastmap") == 0) ret = main_fastmap(argc-1, argv+1);
*/
else {
fprintf(stderr, "[main] unrecognized command '%s'\n", argv[1]);
return 1;
}
err_fflush(stdout);
err_fclose(stdout);
if (ret == 0) {
fprintf(stderr, "[%s] Version: %s\n", __func__, PACKAGE_VERSION);
fprintf(stderr, "[%s] CMD:", __func__);
for (i = 0; i < argc; ++i)
fprintf(stderr, " %s", argv[i]);
fprintf(stderr, "\n[%s] Real time: %.3f sec; CPU: %.3f sec\n", __func__, realtime() - t_real, cputime());
}
return 0;
}
// Detects ORFs in the given nucleotide FASTA file and converts to a protein FASTA file
int writeReadsProtein(const char * passPrefix, const char * passProName) {
struct CDS * orfList;
gzFile inputSeqPtr;
FILE * outputPtr;
char * outputBuffer;
kseq_t * seq;
unsigned long seqIdx, outputSize, orfCount, orfIdx;
char testHeader[4096];
// Prepare file handles
inputSeqPtr = xzopen(passPrefix, "r");
outputPtr = fopen(passProName, "w");
// Iterate through each read
seqIdx = 0;
seq = kseq_init(inputSeqPtr);
while(kseq_read(seq) >= 0) {
// Search for ORFs
getSequenceORF(seq->seq.s, seq->seq.l, &orfList, &orfCount);
//if (orfCount > 0) testReadHeader(seq->name.s, testHeader);
// Write out the corresponding protein sequence for each ORF
for (orfIdx = 0 ; orfIdx < orfCount ; orfIdx++) {
convertToAA(seq->seq.s, orfList+orfIdx, &outputBuffer, &outputSize);
fprintf(outputPtr, ">%d:%d\n%.*s\n", seqIdx, orfIdx, outputSize, outputBuffer);
free(outputBuffer);
}
seqIdx++;
}
// Close files
fflush(outputPtr);
err_gzclose(inputSeqPtr);
err_fclose(outputPtr);
return 0;
}
bntseq_t *bns_restore_core(const char *ann_filename, const char* amb_filename, const char* pac_filename)
{
char str[8192];
FILE *fp;
const char *fname;
bntseq_t *bns;
long long xx;
int i;
int scanres;
bns = (bntseq_t*)calloc(1, sizeof(bntseq_t));
{ // read .ann
fp = xopen(fname = ann_filename, "r");
scanres = fscanf(fp, "%lld%d%u", &xx, &bns->n_seqs, &bns->seed);
if (scanres != 3) goto badread;
bns->l_pac = xx;
bns->anns = (bntann1_t*)calloc(bns->n_seqs, sizeof(bntann1_t));
for (i = 0; i < bns->n_seqs; ++i) {
bntann1_t *p = bns->anns + i;
char *q = str;
int c;
// read gi and sequence name
scanres = fscanf(fp, "%u%s", &p->gi, str);
if (scanres != 2) goto badread;
p->name = strdup(str);
// read fasta comments
while (q - str < sizeof(str) - 1 && (c = fgetc(fp)) != '\n' && c != EOF) *q++ = c;
while (c != '\n' && c != EOF) c = fgetc(fp);
if (c == EOF) {
scanres = EOF;
goto badread;
}
*q = 0;
if (q - str > 1 && strcmp(str, " (null)") != 0) p->anno = strdup(str + 1); // skip leading space
else p->anno = strdup("");
// read the rest
scanres = fscanf(fp, "%lld%d%d", &xx, &p->len, &p->n_ambs);
if (scanres != 3) goto badread;
p->offset = xx;
}
err_fclose(fp);
}
{ // read .amb
int64_t l_pac;
int32_t n_seqs;
fp = xopen(fname = amb_filename, "r");
scanres = fscanf(fp, "%lld%d%d", &xx, &n_seqs, &bns->n_holes);
if (scanres != 3) goto badread;
l_pac = xx;
xassert(l_pac == bns->l_pac && n_seqs == bns->n_seqs, "inconsistent .ann and .amb files.");
bns->ambs = bns->n_holes? (bntamb1_t*)calloc(bns->n_holes, sizeof(bntamb1_t)) : 0;
for (i = 0; i < bns->n_holes; ++i) {
bntamb1_t *p = bns->ambs + i;
scanres = fscanf(fp, "%lld%d%s", &xx, &p->len, str);
if (scanres != 3) goto badread;
p->offset = xx;
p->amb = str[0];
}
err_fclose(fp);
}
{ // open .pac
bns->fp_pac = xopen(pac_filename, "rb");
}
return bns;
badread:
if (EOF == scanres) {
err_fatal(__func__, "Error reading %s : %s\n", fname, ferror(fp) ? strerror(errno) : "Unexpected end of file");
}
err_fatal(__func__, "Parse error reading %s\n", fname);
}
int64_t bis_bns_fasta2bntseq(gzFile fp_fa, const char *prefix, uint8_t parent) {
extern void seq_reverse(int len, ubyte_t *seq, int is_comp); // in bwaseqio.c
kseq_t *seq;
char name[1024];
bntseq_t *bns;
uint8_t *pac = 0, *_pac = 0;
int32_t m_seqs, m_holes;
int64_t ret = -1, m_pac;
bntamb1_t *q;
FILE *fp;
// initialization
gzseek(fp_fa, 0, SEEK_SET);
seq = kseq_init(fp_fa);
bns = (bntseq_t*)calloc(1, sizeof(bntseq_t));
bns->seed = 11; // fixed seed for random generator
srand48(bns->seed);
m_seqs = m_holes = 8; m_pac = 0x10000;
bns->anns = (bntann1_t*)calloc(m_seqs, sizeof(bntann1_t));
bns->ambs = (bntamb1_t*)calloc(m_holes, sizeof(bntamb1_t));
_pac = calloc(m_pac/4, 1);
q = bns->ambs;
if (parent) {
strcpy(name, prefix); strcat(name, ".par.pac");
} else {
strcpy(name, prefix); strcat(name, ".dau.pac");
}
fp = xopen(name, "wb");
// read sequences
while (kseq_read(seq) >= 0) {
_pac = bis_add1(seq, bns, _pac, &m_pac, &m_seqs, &m_holes, &q);
}
/* kseq_rewind(seq); */
/* gzseek(seq->f->f, 0, SEEK_SET); */
/* fprintf(stderr, "foward end\n"); */
/* fflush(stderr); */
/* while (kseq_read(seq) >= 0) { */
/* if (parent) nt256char_rev_ip(seq->seq.s, seq->seq.l); */
/* pac = bis_add1(seq, bns, pac, &m_pac, &m_seqs, &m_holes, &q, parent, 1); */
/* } */
int64_t l,k;
m_pac = (bns->l_pac*2+3)/4*4; /* in bit */
pac = calloc(m_pac/4,sizeof(uint8_t));
for (l=0; l<bns->l_pac; ++l) {
uint8_t c = _get_pac(_pac,l);
if (parent && c == 1) c = 3;
if (!parent && c == 2) c = 0;
_set_pac(pac, l, c);
}
for (k=bns->l_pac-1; k>=0; --k,++l) {
uint8_t c = 3-_get_pac(_pac,k);
if (parent && c == 1) c = 3;
if (!parent && c == 2) c = 0;
_set_pac(pac, l, c);
}
free(_pac);
/* int64_t l; */
/* fprintf(stderr, "reverse end\n"); */
/* fflush(stderr); */
/* if (!for_only) { // add the reverse complemented sequence */
/* m_pac = (bns->l_pac * 2 + 3) / 4 * 4; */
/* pac = realloc(pac, m_pac/4); */
/* memset(pac + (bns->l_pac+3)/4, 0, (m_pac - (bns->l_pac+3)/4*4) / 4); */
/* for (l = bns->l_pac - 1; l >= 0; --l, ++bns->l_pac) */
/* _set_pac(pac, bns->l_pac, 3-_get_pac(pac, l)); */
/* } */
assert(bns->l_pac<<1 == l);
{ // finalize .pac file
ubyte_t ct;
err_fwrite(pac, 1, (l>>2) + ((l&3) == 0? 0 : 1), fp);
// the following codes make the pac file size always (l_pac/4+1+1)
if (l % 4 == 0) {
ct = 0;
err_fwrite(&ct, 1, 1, fp);
}
ct = l % 4;
err_fwrite(&ct, 1, 1, fp);
// close .pac file
err_fflush(fp);
err_fclose(fp);
}
if (parent) bis_bns_dump(bns, prefix);
bns_destroy(bns);
kseq_destroy(seq);
free(pac);
return l;
}