// 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 bwa_seq_close(bwa_seqio_t *bs)
{
if (bs == 0) return;
if (bs->is_bam) {
if (0 != bam_close(bs->fp)) err_fatal_simple("Error closing bam file");
} else {
err_gzclose(bs->ks->f->f);
kseq_destroy(bs->ks);
}
free(bs);
}
int main(int argc, char *argv[])
{
bwaidx_t *idx;
gzFile fp;
kseq_t *ks;
mem_opt_t *opt;
if (argc < 3) {
fprintf(stderr, "Usage: bwamem-lite <idx.base> <reads.fq>\n");
return 1;
}
idx = bwa_idx_load(argv[1], BWA_IDX_ALL); // load the BWA index
if (NULL == idx) {
fprintf(stderr, "Index load failed.\n");
exit(EXIT_FAILURE);
}
fp = strcmp(argv[2], "-")? gzopen(argv[2], "r") : gzdopen(fileno(stdin), "r");
if (NULL == fp) {
fprintf(stderr, "Couldn't open %s : %s\n",
strcmp(argv[2], "-") ? argv[2] : "stdin",
errno ? strerror(errno) : "Out of memory");
exit(EXIT_FAILURE);
}
ks = kseq_init(fp); // initialize the FASTA/Q parser
opt = mem_opt_init(); // initialize the BWA-MEM parameters to the default values
while (kseq_read(ks) >= 0) { // read one sequence
mem_alnreg_v ar;
int i, k;
ar = mem_align1(opt, idx->bwt, idx->bns, idx->pac, ks->seq.l, ks->seq.s); // get all the hits
for (i = 0; i < ar.n; ++i) { // traverse each hit
mem_aln_t a;
if (ar.a[i].secondary >= 0) continue; // skip secondary alignments
a = mem_reg2aln(opt, idx->bns, idx->pac, ks->seq.l, ks->seq.s, &ar.a[i]); // get forward-strand position and CIGAR
// print alignment
err_printf("%s\t%c\t%s\t%ld\t%d\t", ks->name.s, "+-"[a.is_rev], idx->bns->anns[a.rid].name, (long)a.pos, a.mapq);
for (k = 0; k < a.n_cigar; ++k) // print CIGAR
err_printf("%d%c", a.cigar[k]>>4, "MIDSH"[a.cigar[k]&0xf]);
err_printf("\t%d\n", a.NM); // print edit distance
free(a.cigar); // don't forget to deallocate CIGAR
}
free(ar.a); // and deallocate the hit list
}
free(opt);
kseq_destroy(ks);
err_gzclose(fp);
bwa_idx_destroy(idx);
return 0;
}
void bwa_seq_close(bwa_seqio_t *bs)
{
if (bs == 0) return;
if (bs->is_bam) {
#ifdef USE_HTSLIB
if (0 != sam_close(bs->fp)) err_fatal_simple("Error closing sam/bam file");
bam_hdr_destroy(bs->h);
#else
if (0 != bam_close(bs->fp)) err_fatal_simple("Error closing bam file");
#endif
} else {
err_gzclose(bs->ks->f->f);
kseq_destroy(bs->ks);
}
free(bs);
}
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;
}
// 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;
}
