int sa_sam_writer(void *data) {
sa_wf_batch_t *wf_batch = (sa_wf_batch_t *) data;
sa_mapping_batch_t *mapping_batch = (sa_mapping_batch_t *) wf_batch->mapping_batch;
if (mapping_batch == NULL) {
printf("bam_writer1: error, NULL mapping batch\n");
return 0;
}
/*
for (int i = 0; i < NUM_COUNTERS; i++) {
counters[i] += mapping_batch->counters[i];
}
*/
#ifdef _TIMING
for (int i = 0; i < NUM_TIMING; i++) {
func_times[i] += mapping_batch->func_times[i];
}
#endif
int num_mismatches, num_cigar_ops;
size_t flag, pnext = 0, tlen = 0;
char *cigar_string, *cigar_M_string, *rnext = "*";
fastq_read_t *read;
array_list_t *read_list = mapping_batch->fq_reads;
array_list_t *mapping_list, *mate_list;
FILE *out_file = (FILE *) wf_batch->writer_input->bam_file;
sa_genome3_t *genome = wf_batch->sa_index->genome;
size_t num_reads, num_mappings, num_mate_mappings;
num_reads = mapping_batch->num_reads;
if (mapping_batch->options->pair_mode != SINGLE_END_MODE) {
// PAIR MODE
int len;
char *sequence, *quality;
char *seq, *opt_fields;
alignment_t *alig;
for (size_t i = 0; i < num_reads; i++) {
read = (fastq_read_t *) array_list_get(i, read_list);
// seq = read->sequence;
/*
if (i % 2 == 0) {
mate_list = mapping_batch->mapping_lists[i+1];
num_mate_mappings = array_list_size(mate_list);
} else {
mate_list = mapping_list;
num_mate_mappings = num_mappings;
}
*/
mapping_list = mapping_batch->mapping_lists[i];
num_mappings = array_list_size(mapping_list);
num_total_mappings += num_mappings;
#ifdef _VERBOSE
if (num_mappings > 1) {
num_dup_reads++;
num_total_dup_reads += num_mappings;
}
#endif
if (num_mappings > 0) {
num_mapped_reads++;
if (num_mappings > 1) {
num_multihit_reads++;
}
for (size_t j = 0; j < num_mappings; j++) {
alig = (alignment_t *) array_list_get(j, mapping_list);
/*
// update alignment
alig->secondary_alignment = 0;
if (num_mate_mappings != 1) {
alig->is_mate_mapped = 0;
alig->is_paired_end_mapped = 0;
alig->mate_strand = 0;
}
*/
if (alig->optional_fields) {
opt_fields = (char *) calloc(strlen(alig->optional_fields) + 100, sizeof(char));
sprintf(opt_fields, "NH:i:%i\t%s", num_mappings, alig->optional_fields);
// sprintf(opt_fields, "NH:i:%i\t%s\tXU:i:%i", num_mappings, alig->optional_fields, mapping_batch->status[i]);
} else {
opt_fields = (char *) calloc(100, sizeof(char));
sprintf(opt_fields, "NH:i:%i", num_mappings);
// sprintf(opt_fields, "NH:i:%i\tXU:i:%i", num_mappings, mapping_batch->status[i]);
}
/*
// update alignment
alig->secondary_alignment = 0;
if (num_mate_mappings != 1) {
alig->is_mate_mapped = 0;
alig->is_paired_end_mapped = 0;
alig->mate_strand = 0;
}
*/
flag = 0;
if (alig->is_paired_end) flag += BAM_FPAIRED;
if (alig->is_paired_end_mapped) flag += BAM_FPROPER_PAIR;
if (!alig->is_seq_mapped) flag += BAM_FUNMAP;
if ((!alig->is_mate_mapped) && (alig->is_paired_end)) flag += BAM_FMUNMAP;
if (alig->mate_strand) flag += BAM_FMREVERSE;
if (alig->pair_num == 1) flag += BAM_FREAD1;
if (alig->pair_num == 2) flag += BAM_FREAD2;
if (alig->secondary_alignment) flag += BAM_FSECONDARY;
if (alig->fails_quality_check) flag += BAM_FQCFAIL;
if (alig->pc_optical_duplicate) flag += BAM_FDUP;
if (alig->seq_strand) flag += BAM_FREVERSE;
fprintf(out_file, "%s\t%lu\t%s\t%i\t%i\t%s\t%s\t%i\t%i\t%s\t%s\t%s\n",
read->id,
flag,
genome->chrom_names[alig->chromosome],
alig->position + 1,
(num_mappings > 1 ? 0 : alig->mapq), //60, //(alig->map_quality > 3 ? 0 : alig->map_quality),
alig->cigar,
(alig->chromosome == alig->mate_chromosome ? "=" : genome->chrom_names[alig->mate_chromosome]),
alig->mate_position + 1,
alig->template_length,
alig->sequence,
alig->quality,
opt_fields
);
// free memory
free(opt_fields);
alignment_free(alig);
} // end for num_mappings
} else {
num_unmapped_reads++;
opt_fields = (char *) calloc(100, sizeof(char));
sprintf(opt_fields, "XM:i:%i XU:i:%i", num_mappings, mapping_batch->status[i]);
if (read->adapter) {
len = read->length + abs(read->adapter_length);
sequence = (char *) malloc(len + 1);
quality = (char *) malloc(len + 1);
if (read->adapter_length < 0) {
strcpy(quality, read->adapter_quality);
strcat(quality, read->quality);
} else {
strcpy(quality, read->quality);
strcat(quality, read->adapter_quality);
}
if ((read->adapter_strand == 0 && read->adapter_length < 0) ||
(read->adapter_strand == 1 && read->adapter_length > 0)) {
strcpy(sequence, read->adapter);
strcat(sequence, read->sequence);
} else {
strcpy(sequence, read->sequence);
strcat(sequence, read->adapter);
}
sequence[len] = 0;
quality[len] = 0;
} else {
sequence = read->sequence;
quality = read->quality;
}
fprintf(out_file, "%s\t4\t*\t0\t0\t*\t*\t0\t0\t%s\t%s\t%s\n",
read->id,
sequence,
quality,
opt_fields
);
free(opt_fields);
if (read->adapter) {
free(sequence);
free(quality);
}
}
array_list_free(mapping_list, (void *) NULL);
}
} else {
// SINGLE MODE
int len, mapq;
char *seq;
seed_cal_t *cal;
cigar_t *cigar;
char *sequence, *revcomp, *quality;
for (size_t i = 0; i < num_reads; i++) {
read = (fastq_read_t *) array_list_get(i, read_list);
mapping_list = mapping_batch->mapping_lists[i];
num_mappings = array_list_size(mapping_list);
num_total_mappings += num_mappings;
#ifdef _VERBOSE
if (num_mappings > 1) {
num_dup_reads++;
num_total_dup_reads += num_mappings;
}
#endif
if (num_mappings > 0) {
num_mapped_reads++;
if (num_mappings > 1) {
num_multihit_reads++;
}
for (size_t j = 0; j < num_mappings; j++) {
cal = (seed_cal_t *) array_list_get(j, mapping_list);
if (read->adapter) {
// sequences and cigar
len = read->length + abs(read->adapter_length);
sequence = (char *) malloc(len + 1);
revcomp = (char *) malloc(len + 1);
quality = (char *) malloc(len + 1);
cigar = cigar_new_empty();
if (read->adapter_length < 0) {
strcpy(quality, read->adapter_quality);
strcat(quality, read->quality);
} else {
strcpy(quality, read->quality);
strcat(quality, read->adapter_quality);
}
if ( (cal->strand == 1 &&
((read->adapter_strand == 0 && read->adapter_length > 0) ||
(read->adapter_strand == 1 && read->adapter_length < 0)))
||
(cal->strand == 0 &&
((read->adapter_strand == 0 && read->adapter_length < 0) ||
(read->adapter_strand == 1 && read->adapter_length > 0))) ) {
strcpy(sequence, read->adapter);
strcat(sequence, read->sequence);
strcpy(revcomp, read->adapter_revcomp);
strcat(revcomp, read->revcomp);
cigar_append_op(abs(read->adapter_length), 'S', cigar);
cigar_concat(&cal->cigar, cigar);
} else {
strcpy(sequence, read->sequence);
strcat(sequence, read->adapter);
strcpy(revcomp, read->revcomp);
strcat(revcomp, read->adapter_revcomp);
cigar_concat(&cal->cigar, cigar);
cigar_append_op(read->adapter_length, 'S', cigar);
}
sequence[len] = 0;
revcomp[len] = 0;
quality[len] = 0;
} else {
// sequences and cigar
sequence = read->sequence;
revcomp = read->revcomp;
quality = read->quality;
cigar = &cal->cigar;
}
if (cal->strand) {
flag = 16;
seq = revcomp;
} else {
flag = 0;
seq = sequence;
}
/*
if (i == 0) {
flag += BAM_FSECONDARY;
}
*/
cigar_string = cigar_to_string(cigar);
cigar_M_string = cigar_to_M_string(&num_mismatches, &num_cigar_ops, cigar);
if (num_mappings > 1) {
cal->mapq = 0;
}
fprintf(out_file, "%s\t%i\t%s\t%i\t%i\t%s\t%s\t%lu\t%i\t%s\t%s\tNH:i:%i\tNM:i:%i\n",
read->id,
flag,
genome->chrom_names[cal->chromosome_id],
cal->start + 1,
(num_mappings == 1 ? cal->mapq : 0),
cigar_M_string,
rnext,
pnext,
tlen,
seq,
quality,
num_mappings,
num_mismatches
);
// free memory
free(cigar_M_string);
free(cigar_string);
seed_cal_free(cal);
if (read->adapter) {
free(sequence);
free(revcomp);
free(quality);
cigar_free(cigar);
}
}
} else {
num_unmapped_reads++;
if (read->adapter) {
// sequences and cigar
len = read->length + abs(read->adapter_length);
sequence = (char *) malloc(len + 1);
quality = (char *) malloc(len + 1);
if (read->adapter_length < 0) {
strcpy(quality, read->adapter_quality);
strcat(quality, read->quality);
} else {
strcpy(quality, read->quality);
strcat(quality, read->adapter_quality);
}
if ((read->adapter_strand == 0 && read->adapter_length < 0) ||
(read->adapter_strand == 1 && read->adapter_length > 0)) {
strcpy(sequence, read->adapter);
strcat(sequence, read->sequence);
} else {
strcpy(sequence, read->sequence);
strcat(sequence, read->adapter);
}
sequence[len] = 0;
quality[len] = 0;
} else {
// sequences
sequence = read->sequence;
quality = read->quality;
}
fprintf(out_file, "%s\t4\t*\t0\t0\t*\t*\t0\t0\t%s\t%s\n",
read->id,
sequence,
quality
);
if (read->adapter) {
free(sequence);
free(quality);
}
}
array_list_free(mapping_list, (void *) NULL);
} // end for num_reads
}
// free memory
sa_mapping_batch_free(mapping_batch);
if (wf_batch) sa_wf_batch_free(wf_batch);
return 0;
}
void suffix_mng_create_cals(fastq_read_t *read, int min_area, int strand,
sa_index3_t *sa_index, array_list_t *cal_list,
suffix_mng_t *p) {
if (!p) return;
if (!p->suffix_lists) return;
if (p->num_seeds <= 0) return;
int read_area, chrom;
seed_t *seed;
seed_cal_t *cal;
linked_list_t *seed_list;
claspinfo_t info;
bl_claspinfoInit(&info);
// initialization
info.fragments = (Container *) malloc(sizeof(Container));
bl_containerInit(info.fragments, p->num_seeds, sizeof(slmatch_t));
info.subject = p->subject;
slmatch_t frag;
linked_list_t *suffix_list;
for (unsigned int i = 0; i < p->num_chroms; i++) {
suffix_list = p->suffix_lists[i];
if (suffix_list) {
for (linked_list_item_t *item = suffix_list->first;
item != NULL;
item = item->next) {
seed = item->item;
bl_slmatchInit(&frag, 0);
frag.i = seed->read_start;
frag.j = seed->read_end - seed->read_start + 1;
frag.p = seed->genome_start;
frag.q = seed->genome_end - seed->genome_start + 1;
frag.scr = seed->genome_end - seed->genome_start + 1;
frag.subject = seed->chromosome_id;
bl_containerAdd(info.fragments, &frag);
}
}
}
// sort fragments
qsort(info.fragments->contspace, bl_containerSize(info.fragments),
sizeof(slmatch_t), cmp_slmatch_qsort);
int begin = 0;
for (int i = 1; i <= bl_containerSize(info.fragments); i++){
// end of fragments list or different database sequence
// --> process fragment[begin]...fragment[i-1], write output
// and free chains (less memory consumption with large input files)
if (i == bl_containerSize(info.fragments) ||
((slmatch_t *) bl_containerGet(info.fragments, begin))->subject !=
((slmatch_t *) bl_containerGet(info.fragments, i))->subject){
if (info.chainmode == SOP){
// only use chaining without clustering if no ids are specified
bl_slClusterSop((slmatch_t *) info.fragments->contspace + begin, i - begin,
info.epsilon, info.lambda, info.maxgap);
}
else {
bl_slClusterLin((slmatch_t *) info.fragments->contspace + begin, i - begin,
info.epsilon, info.lambda, info.maxgap);
}
for (int j = begin; j < i; j++) {
slmatch_t *match = (slmatch_t *) bl_containerGet(info.fragments, j);
if (match->chain) {
slchain_t *chain = (slchain_t *) match->chain;
if (chain->scr >= info.minscore &&
bl_containerSize(chain->matches) >= info.minfrag) {
chrom = atoi(*(char **) bl_containerGet(info.subject, chain->subject));
read_area = 0;
seed_list = linked_list_new(COLLECTION_MODE_ASYNCHRONIZED);
for (int k = 0; k < bl_containerSize(chain->matches); k++){
slmatch_t *frag = *(slmatch_t **) bl_containerGet(chain->matches, k);
seed = seed_new(frag->i, frag->i + frag->j - 1, frag->p, frag->p + frag->q - 1);
seed->chromosome_id = chrom;
seed->strand = strand;
read_area += frag->j;
cigar_append_op(frag->j, '=', &seed->cigar);
linked_list_insert_last(seed, seed_list);
}
// extend seeds
cal = seed_cal_new(chrom, strand, chain->p, chain->p + chain->q - 1, seed_list);
cal->read = read;
extend_seeds(cal, sa_index);
seed_cal_update_info(cal);
if (cal->read_area >= min_area) {
array_list_insert(cal, cal_list);
} else {
seed_cal_free(cal);
}
}
bl_slchainDestruct(chain);
free(chain);
match->chain = NULL;
}
} // END OF for (j = begin; j < i; j++)
begin = i;
} // END OF if (i == bl_containerSize(info.fragments) ||
} // END OF for (i = 1; i <= bl_containerSize(info.fragments); i++)
// destruct everything
info.subject = NULL;
bl_claspinfoDestruct(&info);
// finally, clear suffix manager
suffix_mng_clear(p);
}
