static void process_orf(GtRange orf_rng, unsigned int orf_frame,
GtStrand strand, GtFeatureNode *gf,
unsigned long offset, unsigned int min,
unsigned int max, GT_UNUSED GtError *err)
{
gt_assert(gf);
unsigned long tmp;
if ((gt_range_length(&orf_rng) >= min) &&
(gt_range_length(&orf_rng) <= max)) {
switch (strand) {
case GT_STRAND_FORWARD:
orf_rng.start = orf_rng.start + offset;
orf_rng.end = orf_rng.end + offset;
break;
case GT_STRAND_REVERSE:
tmp = orf_rng.start;
orf_rng.start = offset - orf_rng.end;
orf_rng.end = offset - tmp;
break;
default:
exit(GT_EXIT_PROGRAMMING_ERROR);
break;
}
orf_attach_results_to_gff3(gf, orf_rng, orf_frame, strand, err);
}
}
int gt_range_compare_by_length_ptr(const GtRange *range_a,
const GtRange *range_b)
{
GtUword range_a_length, range_b_length;
gt_assert(range_a && range_b);
range_a_length = gt_range_length(range_a);
range_b_length = gt_range_length(range_b);
if (range_a_length == range_b_length)
return 0;
if (range_a_length > range_b_length)
return -1;
return 1;
}
unsigned long gth_seq_con_get_length(GthSeqCon *seq_con, unsigned long seq_num)
{
GtRange range;
gt_assert(seq_con);
range = gth_seq_con_get_range(seq_con, seq_num);
return gt_range_length(&range);
}
GtUword gt_ranges_spanned_length(const GtArray *ranges)
{
GtRange spanned_range;
gt_assert(ranges);
spanned_range.start = ((GtRange*) gt_array_get_first(ranges))->start;
spanned_range.end = ((GtRange*) gt_array_get_last(ranges))->end;
return gt_range_length(&spanned_range);
}
static int get_caption_display_status(GtDiagram *d, const char *gft,
bool *result, GtError *err)
{
bool *status;
gt_assert(d && gft);
status = (bool*) gt_hashmap_get(d->caption_display_status, gft);
if (!status)
{
GtUword threshold = GT_UNDEF_UWORD;
double tmp = GT_UNDEF_DOUBLE;
status = gt_malloc(sizeof (bool));
*status = true;
if (gt_style_get_bool(d->style, "format", "show_block_captions", status,
NULL, err) == GT_STYLE_QUERY_ERROR) {
gt_free(status);
return -1;
}
if (*status)
{
GtStyleQueryStatus rval;
rval = gt_style_get_num(d->style,
gft, "max_capt_show_width",
&tmp, NULL, err);
switch (rval) {
case GT_STYLE_QUERY_ERROR:
gt_free(status);
return -1;
break; /* should never reach this */
case GT_STYLE_QUERY_NOT_SET:
*status = true;
break;
default:
gt_assert(tmp != GT_UNDEF_DOUBLE);
threshold = tmp;
gt_assert(tmp != GT_UNDEF_UWORD);
*status = (gt_range_length(&d->range) <= threshold);
break;
}
*status = (gt_range_length(&d->range) <= threshold);
}
gt_hashmap_add(d->caption_display_status, (void*) gft, status);
}
*result = *status;
return 0;
}
static void compute_type_statistics(GtFeatureNode *fn, GtStatVisitor *sv)
{
GtRange range;
gt_assert(fn && sv);
if (gt_feature_node_has_type(fn, gt_ft_gene)) {
sv->number_of_genes++;
if (gt_feature_node_has_CDS(fn))
sv->number_of_protein_coding_genes++;
if (sv->gene_length_distribution) {
range = gt_genome_node_get_range((GtGenomeNode*) fn);
gt_disc_distri_add(sv->gene_length_distribution, gt_range_length(&range));
}
if (sv->gene_score_distribution) {
gt_disc_distri_add(sv->gene_score_distribution,
gt_feature_node_get_score(fn) * 100.0);
}
}
else if (gt_feature_node_has_type(fn, gt_ft_mRNA)) {
sv->number_of_mRNAs++;
if (gt_feature_node_has_CDS(fn))
sv->number_of_protein_coding_mRNAs++;
}
else if (gt_feature_node_has_type(fn, gt_ft_exon)) {
sv->number_of_exons++;
if (sv->exon_length_distribution) {
range = gt_genome_node_get_range((GtGenomeNode*) fn);
gt_disc_distri_add(sv->exon_length_distribution,
gt_range_length(&range));
}
}
else if (gt_feature_node_has_type(fn, gt_ft_CDS)) {
sv->number_of_CDSs++;
}
else if (gt_feature_node_has_type(fn, gt_ft_intron)) {
if (sv->intron_length_distribution) {
range = gt_genome_node_get_range((GtGenomeNode*) fn);
gt_disc_distri_add(sv->intron_length_distribution,
gt_range_length(&range));
}
}
else if (gt_feature_node_has_type(fn, gt_ft_LTR_retrotransposon)) {
sv->number_of_LTR_retrotransposons++;
}
}
static int stat_visitor_region_node(GtNodeVisitor *nv, GtRegionNode *rn,
GT_UNUSED GtError *err)
{
GtStatVisitor *sv;
GtRange range;
gt_error_check(err);
sv = stat_visitor_cast(nv);
sv->number_of_sequence_regions++;
range = gt_genome_node_get_range((GtGenomeNode*) rn);
sv->total_length_of_sequence_regions += gt_range_length(&range);
return 0;
}
GT_UNUSED static int gt_cluster_matches_gap(GtArray *matches,
GtClusteredSet *cs,
unsigned long max_gap_size,
GtError *err)
{
GtMatchReference *mref;
GtMatchEdgeTable matchedgetab;
GtMatchEdge matchedge;
GtMatch *match;
GtRange range;
unsigned long i,
j,
gap_size = 0,
length = 0;
unsigned long num_of_matches;
int had_err = 0;
num_of_matches = gt_array_size(matches);
if (gt_clustered_set_num_of_elements(cs, err) != num_of_matches) {
had_err = -1;
gt_error_set(err,
"number of matches (%lu) unequals number of elements in"
"clustered set (%lu)",
num_of_matches, gt_clustered_set_num_of_elements(cs, err));
}
if (!had_err) {
matchedgetab.edges = gt_array_new(sizeof (GtMatchEdge));
matchedgetab.num_of_edges = 0;
mref = gt_mirror_and_sort_matches(matches);
for (i = 0; i < (2 * (num_of_matches) - 1); i++) {
for (j = i + 1; j < (2 * num_of_matches); j++) {
match = *(GtMatch**) gt_array_get(matches, mref[i].matchnum);
gt_match_get_range_seq1(match, &range);
length = gt_range_length(&range);
gap_size = mref[j].startpos - mref[i].startpos + length;
if (gap_size > max_gap_size)
break;
if (mref[i].matchnum != mref[j].matchnum) {
STORECLUSTEREDGEG(mref[i].matchnum, mref[j].matchnum, gap_size);
}
}
}
if (gt_cluster_matches(cs, &matchedgetab, err) != 0)
had_err = -1;
gt_array_delete(matchedgetab.edges);
gt_free(mref);
}
return had_err;
}
static int add_exon_or_cds_number(GtFeatureNode *fn, void *data,
GT_UNUSED GtError *err)
{
GtStatVisitor *sv = (GtStatVisitor*) data;
gt_error_check(err);
gt_assert(sv && fn);
if (gt_feature_node_has_type(fn, gt_ft_exon))
sv->exon_number_for_distri++;
else if (gt_feature_node_has_type(fn, gt_ft_CDS)) {
GtRange range = gt_genome_node_get_range((GtGenomeNode*) fn);
sv->cds_length_for_distri += gt_range_length(&range);
}
return 0;
}
int gt_feature_index_get_features_for_range(GtFeatureIndex *feature_index,
GtArray *results,
const char *seqid,
const GtRange *range, GtError *err)
{
int ret;
gt_assert(feature_index && feature_index->c_class && results && seqid &&
range);
gt_assert(gt_range_length(range) > 0);
gt_rwlock_rdlock(feature_index->pvt->lock);
ret = feature_index->c_class->get_features_for_range(feature_index, results,
seqid, range, err);
gt_rwlock_unlock(feature_index->pvt->lock);
return ret;
}
static double gaeval_visitor_coverage_resolve(GtFeatureNode *genemodel,
GtArray *exon_coverage)
{
agn_assert(genemodel && exon_coverage);
agn_assert(gt_feature_node_has_type(genemodel, "mRNA"));
GtUword cum_exon_length =
agn_typecheck_feature_combined_length(genemodel, agn_typecheck_exon);
GtUword i, covered = 0;
for(i = 0; i < gt_array_size(exon_coverage); i++)
{
GtRange *range = gt_array_get(exon_coverage, i);
covered += gt_range_length(range);
}
agn_assert(covered <= cum_exon_length);
return (double)covered / (double)cum_exon_length;
}
GtRange gt_pbs_hit_get_coords(const GtPBSHit *h)
{
GtRange rng;
gt_assert(h && h->end >= h->start);
rng.start = h->start;
rng.end = h->end;
switch (h->strand)
{
case GT_STRAND_FORWARD:
default:
rng.start = h->res->elem->leftLTR_3 + 1 - h->res->opts->radius
+ rng.start;
rng.end = rng.start + (h->end - h->start);
break;
case GT_STRAND_REVERSE:
rng.end = h->res->elem->rightLTR_5 - 1 + h->res->opts->radius - rng.start;
rng.start = rng.end - (h->end - h->start);
break;
}
gt_assert(gt_range_length(&rng) == (h->end - h->start + 1));
return rng;
}
static int extract_join_feature(GtGenomeNode *gn, const char *type,
GtRegionMapping *region_mapping,
GtStr *sequence, bool *reverse_strand,
bool *first_child_of_type_seen, GtPhase *phase,
GtError *err)
{
char *outsequence;
GtFeatureNode *fn;
GtRange range;
int had_err = 0;
gt_error_check(err);
fn = gt_feature_node_cast(gn);
gt_assert(fn);
if (gt_feature_node_has_type(fn, type)) {
if (gt_feature_node_get_strand(fn) == GT_STRAND_REVERSE) {
*reverse_strand = true;
*phase = gt_feature_node_get_phase(fn);
} else {
if (!(*first_child_of_type_seen)) {
*first_child_of_type_seen = true;
*phase = gt_feature_node_get_phase(fn);
} else *phase = GT_PHASE_UNDEFINED;
}
range = gt_genome_node_get_range(gn);
had_err = gt_region_mapping_get_sequence(region_mapping, &outsequence,
gt_genome_node_get_seqid(gn),
range.start, range.end, err);
if (!had_err) {
gt_str_append_cstr_nt(sequence, outsequence, gt_range_length(&range));
gt_free(outsequence);
}
}
return had_err;
}
double gt_coords_convert_point(GtRange viewrange, GtWord pos)
{
return ((double) (((GtWord) pos -(GtWord) viewrange.start)))
/ ((double) gt_range_length(&viewrange));
}
static int gt_extract_feature_sequence_generic(GtStr *sequence,
GtGenomeNode *gn,
const char *type, bool join, GtStr *seqid,
GtStrArray *target_ids,
unsigned int *out_phase_offset,
GtRegionMapping *region_mapping, GtError *err)
{
GtFeatureNode *fn;
GtRange range;
unsigned int phase_offset = 0;
char *outsequence;
const char *target;
int had_err = 0;
gt_error_check(err);
fn = gt_genome_node_cast(gt_feature_node_class(), gn);
gt_assert(fn);
if (seqid)
gt_str_append_str(seqid, gt_genome_node_get_seqid(gn));
if (target_ids &&
(target = gt_feature_node_get_attribute(fn, GT_GFF_TARGET))) {
had_err = gt_gff3_parser_parse_all_target_attributes(target, false,
target_ids, NULL,
NULL, "", 0, err);
}
if (!had_err) {
if (join) {
GtFeatureNodeIterator *fni;
GtFeatureNode *child;
bool reverse_strand = false,
first_child = true,
first_child_of_type_seen = false;
GtPhase phase = GT_PHASE_UNDEFINED;
/* in this case we have to traverse the children */
fni = gt_feature_node_iterator_new_direct(gt_feature_node_cast(gn));
while (!had_err && (child = gt_feature_node_iterator_next(fni))) {
if (first_child) {
if (target_ids &&
(target = gt_feature_node_get_attribute(child, GT_GFF_TARGET))) {
gt_str_array_reset(target_ids);
had_err = gt_gff3_parser_parse_all_target_attributes(target, false,
target_ids,
NULL,
NULL, "", 0,
err);
}
first_child = false;
}
if (!had_err) {
if (extract_join_feature((GtGenomeNode*) child, type, region_mapping,
sequence, &reverse_strand,
&first_child_of_type_seen,
&phase, err)) {
had_err = -1;
}
if (phase != GT_PHASE_UNDEFINED) {
phase_offset = (int) phase;
}
}
}
gt_feature_node_iterator_delete(fni);
gt_assert(phase_offset <= (unsigned int) GT_PHASE_UNDEFINED);
if (!had_err && gt_str_length(sequence)) {
if (reverse_strand) {
had_err = gt_reverse_complement(gt_str_get(sequence),
gt_str_length(sequence), err);
}
}
}
else if (gt_feature_node_get_type(fn) == type) {
GtPhase phase = gt_feature_node_get_phase(fn);
gt_assert(!had_err);
if (phase != GT_PHASE_UNDEFINED)
phase_offset = (unsigned int) phase;
/* otherwise we only have to look at this feature */
range = gt_genome_node_get_range(gn);
gt_assert(range.start); /* 1-based coordinates */
had_err = gt_region_mapping_get_sequence(region_mapping, &outsequence,
gt_genome_node_get_seqid(gn),
range.start, range.end, err);
if (!had_err) {
gt_str_append_cstr_nt(sequence, outsequence, gt_range_length(&range));
gt_free(outsequence);
if (gt_feature_node_get_strand(fn) == GT_STRAND_REVERSE) {
had_err = gt_reverse_complement(gt_str_get(sequence),
gt_str_length(sequence), err);
}
}
}
}
if (out_phase_offset && phase_offset != GT_PHASE_UNDEFINED) {
*out_phase_offset = phase_offset;
}
return had_err;
}
static int select_visitor_feature_node(GtNodeVisitor *nv,
GtFeatureNode *fn,
GT_UNUSED GtError *err)
{
GtSelectVisitor *fv;
bool filter_node = false;
gt_error_check(err);
fv = select_visitor_cast(nv);
fv->current_feature++;
if ((!gt_str_length(fv->seqid) || /* no seqid was specified or seqids are
equal */
!gt_str_cmp(fv->seqid, gt_genome_node_get_seqid((GtGenomeNode*) fn))) &&
(!gt_str_length(fv->source) || /* no source was specified or sources are
equal */
!strcmp(gt_str_get(fv->source), gt_feature_node_get_source(fn)))) {
GtRange range = gt_genome_node_get_range((GtGenomeNode*) fn);
/* enforce maximum gene length */
/* XXX: we (spuriously) assume that genes are always root nodes */
if (fn && gt_feature_node_has_type(fn, gt_ft_gene)) {
if (fv->max_gene_length != GT_UNDEF_ULONG &&
gt_range_length(&range) > fv->max_gene_length) {
filter_node = true;
}
else if (fv->max_gene_num != GT_UNDEF_ULONG &&
fv->gene_num >= fv->max_gene_num) {
filter_node = true;
}
else if (fv->min_gene_score != GT_UNDEF_DOUBLE &&
gt_feature_node_get_score(fn) < fv->min_gene_score) {
filter_node = true;
}
else if (fv->max_gene_score != GT_UNDEF_DOUBLE &&
gt_feature_node_get_score(fn) > fv->max_gene_score) {
filter_node = true;
}
else if (fv->feature_num != GT_UNDEF_ULONG &&
fv->feature_num != fv->current_feature) {
filter_node = true;
}
if (!filter_node)
fv->gene_num++; /* gene passed filter */
}
}
else
filter_node = true;
if (!filter_node)
filter_node = filter_contain_range(fn, fv->contain_range);
if (!filter_node)
filter_node = filter_overlap_range(fn, fv->overlap_range);
if (!filter_node)
filter_node = filter_strand(fn, fv->strand);
if (!filter_node)
filter_node = filter_targetstrand(fn, fv->targetstrand);
if (!filter_node)
filter_node = filter_has_CDS(fn, fv->has_CDS);
if (!filter_node)
filter_node = filter_min_average_ssp(fn, fv->min_average_splice_site_prob);
if (filter_node)
gt_genome_node_delete((GtGenomeNode*) fn);
else
gt_queue_add(fv->node_buffer, fn);
return 0;
}
static int gt_snp_annotator_visitor_prepare_gene(GtSNPAnnotatorVisitor *sav,
GtError *err)
{
GtFeatureNodeIterator *fni,
*mrnafni;
GtFeatureNode *curnode,
*last_mRNA = NULL;
GtStr *mrnaseq,
*seqid;
int had_err = 0;
mrnaseq = gt_str_new();
seqid = gt_genome_node_get_seqid((GtGenomeNode*) sav->gene);
fni = gt_feature_node_iterator_new(sav->gene);
while (!had_err && (curnode = gt_feature_node_iterator_next(fni))) {
if (gt_feature_node_get_type(curnode) == sav->mRNA_type) {
GtFeatureNode *curnode2;
if (last_mRNA) {
char *mrna_charseq = gt_calloc(gt_str_length(mrnaseq)+1, sizeof (char));
(void) strncpy(mrna_charseq, gt_str_get(mrnaseq),
gt_str_length(mrnaseq));
if (gt_feature_node_get_strand(sav->gene) == GT_STRAND_REVERSE) {
had_err = gt_reverse_complement(mrna_charseq, gt_str_length(mrnaseq),
err);
}
if (!had_err) {
gt_hashmap_add(sav->rnaseqs, last_mRNA, mrna_charseq);
last_mRNA = curnode;
gt_str_reset(mrnaseq);
}
} else last_mRNA = curnode;
if (!had_err) {
mrnafni = gt_feature_node_iterator_new(curnode);
while (!had_err && (curnode2 =
gt_feature_node_iterator_next(mrnafni))) {
if (gt_feature_node_get_type(curnode2) == sav->CDS_type) {
char *tmp;
GtRange rng = gt_genome_node_get_range((GtGenomeNode*) curnode2);
had_err = gt_region_mapping_get_sequence(sav->rmap, &tmp, seqid,
rng.start, rng.end, err);
if (!had_err) {
gt_str_append_cstr_nt(mrnaseq, tmp, gt_range_length(&rng));
gt_free(tmp);
}
}
}
gt_feature_node_iterator_delete(mrnafni);
}
}
}
if (!had_err && last_mRNA) {
char *mrna_charseq = gt_calloc(gt_str_length(mrnaseq)+1, sizeof (char));
(void) strncpy(mrna_charseq, gt_str_get(mrnaseq), gt_str_length(mrnaseq));
if (gt_feature_node_get_strand(sav->gene) == GT_STRAND_REVERSE) {
had_err = gt_reverse_complement(mrna_charseq, gt_str_length(mrnaseq),
err);
}
if (!had_err) {
gt_hashmap_add(sav->rnaseqs, last_mRNA, mrna_charseq);
}
}
gt_feature_node_iterator_delete(fni);
gt_str_delete(mrnaseq);
return had_err;
}
static int snp_annotator_visitor_feature_node(GtNodeVisitor *nv,
GtFeatureNode *fn,
GtError *err)
{
int had_err = 0;
GtSNPAnnotatorVisitor *sav;
GtFeatureNodeIterator *fni,
*mrnafni;
GtFeatureNode *curnode,
*curnode2;
GtRange snp_rng;
gt_error_check(err);
sav = snp_annotator_visitor_cast(nv);
/* ignore non-nodes */
if (!fn) return 0;
/* only process SNPs */
if (!(gt_feature_node_get_type(fn) == sav->SNV_type ||
gt_feature_node_get_type(fn) == sav->SNP_type)) {
return 0;
}
fni = gt_feature_node_iterator_new_direct(sav->gene);
snp_rng = gt_genome_node_get_range((GtGenomeNode*) fn);
while (!had_err && (curnode = gt_feature_node_iterator_next(fni))) {
if (gt_feature_node_get_type(curnode) == sav->mRNA_type) {
GtStrand mrna_strand = gt_feature_node_get_strand(curnode);
#ifndef NDEBUG
const char *refstr;
#endif
GtUword mrnasnppos = 0;
mrnafni = gt_feature_node_iterator_new(curnode);
while (!had_err && (curnode2 = gt_feature_node_iterator_next(mrnafni))) {
if (gt_feature_node_get_type(curnode2) == sav->CDS_type) {
GtRange cds_rng = gt_genome_node_get_range((GtGenomeNode*) curnode2);
if (gt_range_overlap(&snp_rng, &cds_rng)) {
char *mRNA,
origchar;
char *variantchars, *variantptr = NULL;
GT_UNUSED char *refchars, *refptr = NULL;
mRNA = (char*) gt_hashmap_get(sav->rnaseqs, curnode);
gt_assert(mRNA);
gt_assert(snp_rng.start >= cds_rng.start);
mrnasnppos += (snp_rng.start - cds_rng.start);
if (mrna_strand == GT_STRAND_REVERSE)
mrnasnppos = strlen(mRNA) - mrnasnppos - 1;
gt_assert(mrnasnppos < strlen(mRNA));
origchar = mRNA[mrnasnppos];
#ifndef NDEBUG
refstr = refptr = gt_cstr_dup(gt_feature_node_get_attribute(fn,
GT_GVF_REFERENCE_SEQ));
if (!had_err && refstr) {
if (gt_feature_node_get_strand(curnode) == GT_STRAND_REVERSE) {
int rval = gt_complement(&origchar, origchar, err);
gt_assert(rval == 0);
}
gt_assert(toupper(origchar) == toupper(refstr[0]));
}
#endif
variantchars = variantptr = gt_cstr_dup(
gt_feature_node_get_attribute(fn, GT_GVF_VARIANT_SEQ));
if (!had_err && variantchars) {
GtUword i = 0;
while (!had_err &&
(*variantchars != ';' && *variantchars != '\0')) {
if (*variantchars != ',' && *variantchars != origchar) {
char variantchar = *variantchars;
#ifndef NDEBUG
char refchar = refstr ? refstr[0] : '-'; /* XXX */
if (!had_err && mrna_strand == GT_STRAND_REVERSE)
had_err = gt_complement(&refchar, refchar, err);
#endif
if (!had_err && mrna_strand == GT_STRAND_REVERSE)
had_err = gt_complement(&variantchar, variantchar, err);
if (!had_err) {
had_err = snp_annotator_classify_snp(sav, curnode, fn,
mrnasnppos,
i++,
variantchar,
#ifndef NDEBUG
refchar,
#endif
err);
}
} else if (*variantchars == origchar) {
i++;
}
variantchars++;
}
gt_free(variantptr);
gt_free(refptr);
}
} else {
mrnasnppos += gt_range_length(&cds_rng);
}
}
}
gt_feature_node_iterator_delete(mrnafni);
}
}
gt_feature_node_iterator_delete(fni);
return had_err;
}
int gt_ltrfileout_stream_next(GtNodeStream *ns, GtGenomeNode **gn, GtError *err)
{
GtLTRdigestFileOutStream *ls;
GtFeatureNode *fn;
GtRange lltr_rng = {GT_UNDEF_UWORD, GT_UNDEF_UWORD},
rltr_rng = {GT_UNDEF_UWORD, GT_UNDEF_UWORD},
ppt_rng = {GT_UNDEF_UWORD, GT_UNDEF_UWORD},
pbs_rng = {GT_UNDEF_UWORD, GT_UNDEF_UWORD};
int had_err;
GtUword i=0;
gt_error_check(err);
ls = gt_ltrdigest_file_out_stream_cast(ns);
/* initialize this element */
memset(&ls->element, 0, sizeof (GtLTRElement));
/* get annotations from parser */
had_err = gt_node_stream_next(ls->in_stream, gn, err);
if (!had_err && *gn)
{
GtFeatureNodeIterator* gni;
GtFeatureNode *mygn;
/* only process feature nodes */
if (!(fn = gt_feature_node_try_cast(*gn)))
return 0;
ls->element.pdomorder = gt_array_new(sizeof (const char*));
/* fill LTRElement structure from GFF3 subgraph */
gni = gt_feature_node_iterator_new(fn);
for (mygn = fn; mygn != NULL; mygn = gt_feature_node_iterator_next(gni))
(void) gt_genome_node_accept((GtGenomeNode*) mygn,
(GtNodeVisitor*) ls->lv,
err);
gt_feature_node_iterator_delete(gni);
}
if (!had_err && ls->element.mainnode != NULL)
{
char desc[GT_MAXFASTAHEADER];
GtFeatureNode *ltr3, *ltr5;
GtStr *sdesc, *sreg, *seq;
/* find sequence in GtEncseq */
sreg = gt_genome_node_get_seqid((GtGenomeNode*) ls->element.mainnode);
sdesc = gt_str_new();
had_err = gt_region_mapping_get_description(ls->rmap, sdesc, sreg, err);
if (!had_err) {
GtRange rng;
ls->element.seqid = gt_calloc((size_t) ls->seqnamelen+1, sizeof (char));
(void) snprintf(ls->element.seqid,
MIN((size_t) gt_str_length(sdesc),
(size_t) ls->seqnamelen)+1,
"%s", gt_str_get(sdesc));
gt_cstr_rep(ls->element.seqid, ' ', '_');
if (gt_str_length(sdesc) > (GtUword) ls->seqnamelen)
ls->element.seqid[ls->seqnamelen] = '\0';
(void) gt_ltrelement_format_description(&ls->element,
ls->seqnamelen,
desc,
(size_t) (GT_MAXFASTAHEADER-1));
gt_str_delete(sdesc);
/* output basic retrotransposon data */
lltr_rng = gt_genome_node_get_range((GtGenomeNode*) ls->element.leftLTR);
rltr_rng = gt_genome_node_get_range((GtGenomeNode*) ls->element.rightLTR);
rng = gt_genome_node_get_range((GtGenomeNode*) ls->element.mainnode);
gt_file_xprintf(ls->tabout_file,
GT_WU"\t"GT_WU"\t"GT_WU"\t%s\t"GT_WU"\t"GT_WU"\t"GT_WU"\t"
GT_WU"\t"GT_WU"\t"GT_WU"\t",
rng.start, rng.end, gt_ltrelement_length(&ls->element),
ls->element.seqid, lltr_rng.start, lltr_rng.end,
gt_ltrelement_leftltrlen(&ls->element), rltr_rng.start,
rltr_rng.end, gt_ltrelement_rightltrlen(&ls->element));
}
seq = gt_str_new();
/* output TSDs */
if (!had_err && ls->element.leftTSD != NULL)
{
GtRange tsd_rng;
tsd_rng = gt_genome_node_get_range((GtGenomeNode*) ls->element.leftTSD);
had_err = gt_extract_feature_sequence(seq,
(GtGenomeNode*) ls->element.leftTSD,
gt_symbol(gt_ft_target_site_duplication),
false,
NULL, NULL, ls->rmap, err);
if (!had_err) {
gt_file_xprintf(ls->tabout_file,
""GT_WU"\t"GT_WU"\t%s\t",
tsd_rng.start,
tsd_rng.end,
gt_str_get(seq));
}
gt_str_reset(seq);
} else gt_file_xprintf(ls->tabout_file, "\t\t\t");
if (!had_err && ls->element.rightTSD != NULL)
{
GtRange tsd_rng;
tsd_rng = gt_genome_node_get_range((GtGenomeNode*) ls->element.rightTSD);
had_err = gt_extract_feature_sequence(seq,
(GtGenomeNode*) ls->element.rightTSD,
gt_symbol(gt_ft_target_site_duplication),
false,
NULL, NULL, ls->rmap, err);
if (!had_err) {
gt_file_xprintf(ls->tabout_file,
""GT_WU"\t"GT_WU"\t%s\t",
tsd_rng.start,
tsd_rng.end,
gt_str_get(seq));
}
gt_str_reset(seq);
} else gt_file_xprintf(ls->tabout_file, "\t\t\t");
/* output PPT */
if (!had_err && ls->element.ppt != NULL)
{
GtStrand ppt_strand = gt_feature_node_get_strand(ls->element.ppt);
ppt_rng = gt_genome_node_get_range((GtGenomeNode*) ls->element.ppt);
had_err = gt_extract_feature_sequence(seq,
(GtGenomeNode*) ls->element.ppt,
gt_symbol(gt_ft_RR_tract), false,
NULL, NULL, ls->rmap, err);
if (!had_err) {
gt_fasta_show_entry(desc, gt_str_get(seq), gt_range_length(&ppt_rng),
GT_FSWIDTH, ls->pptout_file);
gt_file_xprintf(ls->tabout_file,
""GT_WU"\t"GT_WU"\t%s\t%c\t%d\t",
ppt_rng.start,
ppt_rng.end,
gt_str_get(seq),
GT_STRAND_CHARS[ppt_strand],
(ppt_strand == GT_STRAND_FORWARD ?
abs((int) (rltr_rng.start - ppt_rng.end)) :
abs((int) (lltr_rng.end - ppt_rng.start))));
}
gt_str_reset(seq);
} else gt_file_xprintf(ls->tabout_file, "\t\t\t\t\t");
/* output PBS */
if (!had_err && ls->element.pbs != NULL)
{
GtStrand pbs_strand;
pbs_strand = gt_feature_node_get_strand(ls->element.pbs);
pbs_rng = gt_genome_node_get_range((GtGenomeNode*) ls->element.pbs);
had_err = gt_extract_feature_sequence(seq,
(GtGenomeNode*) ls->element.pbs,
gt_symbol(gt_ft_primer_binding_site),
false, NULL, NULL, ls->rmap, err);
if (!had_err) {
gt_fasta_show_entry(desc, gt_str_get(seq), gt_range_length(&pbs_rng),
GT_FSWIDTH, ls->pbsout_file);
gt_file_xprintf(ls->tabout_file,
""GT_WU"\t"GT_WU"\t%c\t%s\t%s\t%s\t%s\t%s\t",
pbs_rng.start,
pbs_rng.end,
GT_STRAND_CHARS[pbs_strand],
gt_feature_node_get_attribute(ls->element.pbs, "trna"),
gt_str_get(seq),
gt_feature_node_get_attribute(ls->element.pbs,
"pbsoffset"),
gt_feature_node_get_attribute(ls->element.pbs,
"trnaoffset"),
gt_feature_node_get_attribute(ls->element.pbs,
"edist"));
}
gt_str_reset(seq);
} else gt_file_xprintf(ls->tabout_file, "\t\t\t\t\t\t\t\t");
/* output protein domains */
if (!had_err && ls->element.pdoms != NULL)
{
GtStr *pdomorderstr = gt_str_new();
for (i=0; !had_err && i<gt_array_size(ls->element.pdomorder); i++)
{
const char* key = *(const char**) gt_array_get(ls->element.pdomorder,
i);
GtArray *entry = (GtArray*) gt_hashmap_get(ls->element.pdoms, key);
had_err = write_pdom(ls, entry, key, ls->rmap, desc, err);
}
if (GT_STRAND_REVERSE == gt_feature_node_get_strand(ls->element.mainnode))
gt_array_reverse(ls->element.pdomorder);
for (i=0 ;!had_err && i<gt_array_size(ls->element.pdomorder); i++)
{
const char* name = *(const char**) gt_array_get(ls->element.pdomorder,
i);
gt_str_append_cstr(pdomorderstr, name);
if (i != gt_array_size(ls->element.pdomorder)-1)
gt_str_append_cstr(pdomorderstr, "/");
}
gt_file_xprintf(ls->tabout_file, "%s", gt_str_get(pdomorderstr));
gt_str_delete(pdomorderstr);
}
/* output LTRs (we just expect them to exist) */
switch (gt_feature_node_get_strand(ls->element.mainnode))
{
case GT_STRAND_REVERSE:
ltr5 = ls->element.rightLTR;
ltr3 = ls->element.leftLTR;
break;
case GT_STRAND_FORWARD:
default:
ltr5 = ls->element.leftLTR;
ltr3 = ls->element.rightLTR;
break;
}
if (!had_err) {
had_err = gt_extract_feature_sequence(seq, (GtGenomeNode*) ltr5,
gt_symbol(gt_ft_long_terminal_repeat),
false,
NULL, NULL, ls->rmap, err);
}
if (!had_err) {
gt_fasta_show_entry(desc, gt_str_get(seq), gt_str_length(seq),
GT_FSWIDTH, ls->ltr5out_file);
gt_str_reset(seq);
}
if (!had_err) {
had_err = gt_extract_feature_sequence(seq, (GtGenomeNode*) ltr3,
gt_symbol(gt_ft_long_terminal_repeat),
false,
NULL, NULL, ls->rmap, err);
}
if (!had_err) {
gt_fasta_show_entry(desc, gt_str_get(seq), gt_str_length(seq),
GT_FSWIDTH, ls->ltr3out_file);
gt_str_reset(seq);
}
/* output complete oriented element */
if (!had_err) {
had_err = gt_extract_feature_sequence(seq,
(GtGenomeNode*) ls->element.mainnode,
gt_symbol(gt_ft_LTR_retrotransposon),
false,
NULL, NULL, ls->rmap, err);
}
if (!had_err) {
gt_fasta_show_entry(desc,gt_str_get(seq), gt_str_length(seq),
GT_FSWIDTH, ls->elemout_file);
gt_str_reset(seq);
}
gt_file_xprintf(ls->tabout_file, "\n");
gt_str_delete(seq);
}
gt_hashmap_delete(ls->element.pdoms);
gt_array_delete(ls->element.pdomorder);
gt_free(ls->element.seqid);
return had_err;
}
static int write_pdom(GtLTRdigestFileOutStream *ls, GtArray *pdoms,
const char *pdomname, GT_UNUSED GtRegionMapping *rmap,
char *desc, GtError *err)
{
int had_err = 0;
GtFile *seqfile = NULL,
*alifile = NULL,
*aafile = NULL;
GtUword i = 0,
seq_length = 0;
GtStr *pdom_seq,
*pdom_aaseq;
gt_error_check(err);
pdom_seq = gt_str_new();
pdom_aaseq = gt_str_new();
/* get protein domain output file */
seqfile = (GtFile*) gt_hashmap_get(ls->pdomout_files, pdomname);
if (seqfile == NULL)
{
/* no file opened for this domain yet, do it */
char buffer[GT_MAXFILENAMELEN];
(void) snprintf(buffer, (size_t) (GT_MAXFILENAMELEN-1), "%s_pdom_%s.fas",
ls->fileprefix, pdomname);
seqfile = gt_file_xopen(buffer, "w+");
gt_hashmap_add(ls->pdomout_files, gt_cstr_dup(pdomname), seqfile);
}
/* get protein alignment output file */
if (ls->write_pdom_alignments)
{
alifile = (GtFile*) gt_hashmap_get(ls->pdomali_files, pdomname);
if (alifile == NULL)
{
/* no file opened for this domain yet, do it */
char buffer[GT_MAXFILENAMELEN];
(void) snprintf(buffer, (size_t) (GT_MAXFILENAMELEN-1), "%s_pdom_%s.ali",
ls->fileprefix, pdomname);
alifile = gt_file_xopen(buffer, "w+");
gt_hashmap_add(ls->pdomali_files, gt_cstr_dup(pdomname), alifile);
}
}
/* get amino acid sequence output file */
if (ls->write_pdom_aaseqs)
{
aafile = (GtFile*) gt_hashmap_get(ls->pdomaa_files, pdomname);
if (aafile == NULL)
{
/* no file opened for this domain yet, do it */
char buffer[GT_MAXFILENAMELEN];
(void) snprintf(buffer, (size_t) (GT_MAXFILENAMELEN-1),
"%s_pdom_%s_aa.fas",
ls->fileprefix, pdomname);
aafile = gt_file_xopen(buffer, "w+");
gt_hashmap_add(ls->pdomaa_files, gt_cstr_dup(pdomname), aafile);
}
}
if (gt_array_size(pdoms) > 1UL)
{
for (i=1UL; i<gt_array_size(pdoms); i++)
{
gt_assert(gt_genome_node_cmp(*(GtGenomeNode**)gt_array_get(pdoms, i),
*(GtGenomeNode**)gt_array_get(pdoms, i-1))
>= 0);
}
if (gt_feature_node_get_strand(*(GtFeatureNode**) gt_array_get(pdoms, 0UL))
== GT_STRAND_REVERSE)
{
gt_array_reverse(pdoms);
}
}
/* output protein domain data */
for (i=0;i<gt_array_size(pdoms);i++)
{
GtRange pdom_rng;
GtStr *ali,
*aaseq;
GtFeatureNode *fn;
int rval;
fn = *(GtFeatureNode**) gt_array_get(pdoms, i);
ali = gt_genome_node_get_user_data((GtGenomeNode*) fn, "pdom_alignment");
aaseq = gt_genome_node_get_user_data((GtGenomeNode*) fn, "pdom_aaseq");
pdom_rng = gt_genome_node_get_range((GtGenomeNode*) fn);
rval = gt_extract_feature_sequence(pdom_seq, (GtGenomeNode*) fn,
gt_symbol(gt_ft_protein_match), false,
NULL, NULL, rmap, err);
if (rval)
{
had_err = -1;
break;
}
if (ls->write_pdom_alignments && ali)
{
char buf[BUFSIZ];
/* write away alignment */
(void) snprintf(buf, BUFSIZ-1, "Protein domain alignment in translated "
"sequence for candidate\n'%s':\n\n",
desc);
gt_file_xwrite(alifile, buf, (size_t) strlen(buf) * sizeof (char));
gt_file_xwrite(alifile, gt_str_get(ali),
(size_t) gt_str_length(ali) * sizeof (char));
gt_file_xwrite(alifile, "---\n\n", 5 * sizeof (char));
}
if (ls->write_pdom_aaseqs && aaseq)
{
/* append amino acid sequence */
gt_str_append_str(pdom_aaseq, aaseq);
}
gt_genome_node_release_user_data((GtGenomeNode*) fn, "pdom_alignment");
gt_genome_node_release_user_data((GtGenomeNode*) fn, "pdom_aaseq");
seq_length += gt_range_length(&pdom_rng);
}
if (!had_err)
{
gt_fasta_show_entry(desc,
gt_str_get(pdom_seq),
seq_length,
GT_FSWIDTH,
seqfile);
if (ls->write_pdom_aaseqs)
{
gt_fasta_show_entry(desc,
gt_str_get(pdom_aaseq),
gt_str_length(pdom_aaseq),
GT_FSWIDTH,
aafile);
}
}
gt_str_delete(pdom_seq);
gt_str_delete(pdom_aaseq);
return had_err;
}
static int process_node(GtDiagram *d, GtFeatureNode *node,
GtFeatureNode *parent, GtError *err)
{
GtRange elem_range;
bool *collapse;
GtShouldGroupByParent *group;
const char *feature_type = NULL,
*parent_gft = NULL;
double tmp;
GtStyleQueryStatus rval;
GtUword max_show_width = GT_UNDEF_UWORD,
par_max_show_width = GT_UNDEF_UWORD;
gt_assert(d && node);
gt_log_log(">> getting '%s'", gt_feature_node_get_type(node));
/* skip pseudonodes */
if (gt_feature_node_is_pseudo(node))
return 0;
feature_type = gt_feature_node_get_type(node);
gt_assert(feature_type);
/* discard elements that do not overlap with visible range */
elem_range = gt_genome_node_get_range((GtGenomeNode*) node);
if (!gt_range_overlap(&d->range, &elem_range))
return 0;
/* get maximal view widths in nucleotides to show this type */
rval = gt_style_get_num(d->style, feature_type, "max_show_width", &tmp, NULL,
err);
switch (rval) {
case GT_STYLE_QUERY_OK:
max_show_width = tmp;
break;
case GT_STYLE_QUERY_ERROR:
return -1;
break; /* should never be reached */
default:
/* do not change default value */
break;
}
/* for non-root nodes, get maximal view with to show parent */
if (parent)
{
if (!gt_feature_node_is_pseudo(parent))
{
parent_gft = gt_feature_node_get_type(parent);
rval = gt_style_get_num(d->style,
parent_gft, "max_show_width",
&tmp, NULL, err);
switch (rval) {
case GT_STYLE_QUERY_OK:
par_max_show_width = tmp;
break;
case GT_STYLE_QUERY_ERROR:
return -1;
break; /* should never be reached */
default:
/* do not change default value */
break;
}
} else
par_max_show_width = GT_UNDEF_UWORD;
}
/* check if this type is to be displayed at all */
if (max_show_width != GT_UNDEF_UWORD &&
gt_range_length(&d->range) > max_show_width)
{
return 0;
}
/* disregard parent node if it is configured not to be shown */
if (parent
&& par_max_show_width != GT_UNDEF_UWORD
&& gt_range_length(&d->range) > par_max_show_width)
{
parent = NULL;
}
/* check if this is a collapsing type, cache result */
if ((collapse = (bool*) gt_hashmap_get(d->collapsingtypes,
feature_type)) == NULL)
{
collapse = gt_malloc(sizeof (bool));
*collapse = false;
if (gt_style_get_bool(d->style, feature_type, "collapse_to_parent",
collapse, NULL, err) == GT_STYLE_QUERY_ERROR) {
gt_free(collapse);
return -1;
}
gt_hashmap_add(d->collapsingtypes, (void*) feature_type, collapse);
}
/* check if type should be grouped by parent, cache result */
if ((group = (GtShouldGroupByParent*) gt_hashmap_get(d->groupedtypes,
feature_type)) == NULL)
{
bool tmp;
group = gt_malloc(sizeof (GtShouldGroupByParent));
rval = gt_style_get_bool(d->style, feature_type, "group_by_parent",
&tmp, NULL, err);
switch (rval) {
case GT_STYLE_QUERY_OK:
if (tmp)
*group = GT_GROUP_BY_PARENT;
else
*group = GT_DO_NOT_GROUP_BY_PARENT;
break;
case GT_STYLE_QUERY_NOT_SET:
*group = GT_UNDEFINED_GROUPING;
break;
case GT_STYLE_QUERY_ERROR:
gt_free(group);
return -1;
break; /* should never be reached */
}
gt_hashmap_add(d->groupedtypes, (void*) feature_type, group);
}
/* decide where to place this feature: */
if (*collapse)
{
/* user has specified collapsing to parent for this type */
if (parent && !gt_feature_node_is_pseudo(parent)) {
/* collapsing child nodes are added to upwards blocks,
but never collapse into pseudo nodes */
add_recursive(d, node, parent, node);
} else {
/* if no parent or only pseudo-parent, do not collapse */
if (add_to_current(d, node, parent, err) < 0) {
return -1;
}
}
}
else /* (!*collapse) */
{
if (parent) {
bool do_not_overlap = false;
do_not_overlap = gt_feature_node_direct_children_do_not_overlap_st(parent,
node);
if (*group == GT_GROUP_BY_PARENT
|| (do_not_overlap && *group == GT_UNDEFINED_GROUPING))
{
if (gt_feature_node_is_pseudo(parent)
&& gt_feature_node_is_multi(node))
{
if (add_to_rep(d, node, parent, err) < 0) {
return -1;
}
} else if
(gt_feature_node_number_of_children(parent) > 1)
{
if (add_to_parent(d, node, parent, err) < 0) {
return -1;
}
} else {
if (add_to_current(d, node, parent, err) < 0) {
return -1;
}
}
} else {
if (gt_feature_node_is_pseudo(parent)
&& gt_feature_node_is_multi(node))
{
if (add_to_rep(d, node, parent, err) < 0) {
return -1;
}
} else {
if (add_to_current(d, node, parent, err) < 0) {
return -1;
}
}
}
} else {
/* root nodes always get their own block */
if (add_to_current(d, node, parent, err) < 0) {
return -1;
}
}
}
/* we can now assume that this node (or its representative)
has been processed into the reverse lookup structure */
#ifndef NDEBUG
if (gt_feature_node_is_multi(node))
{
GtFeatureNode *rep;
rep = gt_feature_node_get_multi_representative((GtFeatureNode*) node);
gt_assert(gt_hashmap_get(d->nodeinfo, rep));
}
else
gt_assert(gt_hashmap_get(d->nodeinfo, node));
#endif
return 0;
}
static int calc_spliced_alignments(GthSACollection *sa_collection,
GthChainCollection *chain_collection,
GthCallInfo *call_info,
GthInput *input,
GthStat *stat,
GtUword gen_file_num,
GtUword ref_file_num,
bool directmatches,
GthMatchInfo *match_info,
GthDNACompletePathMatrixJT
dna_complete_path_matrix_jt,
GthProteinCompletePathMatrixJT
protein_complete_path_matrix_jt)
{
const unsigned char *ref_seq_tran, *ref_seq_orig, *ref_seq_tran_rc = NULL,
*ref_seq_orig_rc = NULL;
GtUword chainctr, gen_offset = GT_UNDEF_UWORD, gen_total_length,
ref_total_length;
GtFile *outfp = call_info->out->outfp;
GtRange gen_seq_bounds, gen_seq_bounds_rc;
bool refseqisdna;
GthChain *chain;
GtRange range;
GthSA *saA;
int rval;
gt_assert(sa_collection && chain_collection);
refseqisdna = gth_input_ref_file_is_dna(input, ref_file_num);
for (chainctr = 0;
chainctr < gth_chain_collection_size(chain_collection);
chainctr++) {
chain = gth_chain_collection_get(chain_collection, chainctr);
if (++match_info->call_number > call_info->firstalshown &&
call_info->firstalshown > 0) {
if (!(call_info->out->xmlout || call_info->out->gff3out))
gt_file_xfputc('\n', outfp);
else if (call_info->out->xmlout)
gt_file_xprintf(outfp, "<!--\n");
if (!call_info->out->gff3out) {
gt_file_xprintf(outfp, "Maximal matching %s count (%u) reached.\n",
refseqisdna ? "EST" : "protein",
call_info->firstalshown);
gt_file_xprintf(outfp, "Only the first %u matches will be "
"displayed.\n", call_info->firstalshown);
}
if (!(call_info->out->xmlout || call_info->out->gff3out))
gt_file_xfputc('\n', outfp);
else if (call_info->out->xmlout)
gt_file_xprintf(outfp, "-->\n");
match_info->max_call_number_reached = true;
break; /* break out of loop */
}
/* compute considered genomic regions if not set by -frompos */
if (!gth_input_use_substring_spec(input)) {
gen_seq_bounds = gth_input_get_genomic_range(input, chain->gen_file_num,
chain->gen_seq_num);
gen_total_length = gt_range_length(&gen_seq_bounds);
gen_offset = gen_seq_bounds.start;
gen_seq_bounds_rc = gen_seq_bounds;
}
else {
/* genomic multiseq contains exactly one sequence */
gt_assert(gth_input_num_of_gen_seqs(input, chain->gen_file_num) == 1);
gen_total_length = gth_input_genomic_file_total_length(input,
chain
->gen_file_num);
gen_seq_bounds.start = gth_input_genomic_substring_from(input);
gen_seq_bounds.end = gth_input_genomic_substring_to(input);
gen_offset = 0;
gen_seq_bounds_rc.start = gen_total_length - 1 - gen_seq_bounds.end;
gen_seq_bounds_rc.end = gen_total_length - 1 - gen_seq_bounds.start;
}
/* "retrieving" the reference sequence */
range = gth_input_get_reference_range(input, chain->ref_file_num,
chain->ref_seq_num);
ref_seq_tran = gth_input_current_ref_seq_tran(input) + range.start;
ref_seq_orig = gth_input_current_ref_seq_orig(input) + range.start;
if (refseqisdna) {
ref_seq_tran_rc = gth_input_current_ref_seq_tran_rc(input) + range.start;
ref_seq_orig_rc = gth_input_current_ref_seq_orig_rc(input) + range.start;
}
ref_total_length = range.end - range.start + 1;
/* check if protein sequences have a stop amino acid */
if (!refseqisdna && !match_info->stop_amino_acid_warning &&
ref_seq_orig[ref_total_length - 1] != GT_STOP_AMINO) {
GtStr *ref_id = gt_str_new();
gth_input_save_ref_id(input, ref_id, chain->ref_file_num,
chain->ref_seq_num);
gt_warning("protein sequence '%s' (#" GT_WU " in file %s) does not end "
"with a stop amino acid ('%c'). If it is not a protein "
"fragment you should add a stop amino acid to improve the "
"prediction. For example with `gt seqtransform "
"-addstopaminos` (see http://genometools.org for details).",
gt_str_get(ref_id), chain->ref_seq_num,
gth_input_get_reference_filename(input, chain->ref_file_num),
GT_STOP_AMINO);
match_info->stop_amino_acid_warning = true;
gt_str_delete(ref_id);
}
/* allocating space for alignment */
saA = gth_sa_new_and_set(directmatches, true, input, chain->gen_file_num,
chain->gen_seq_num, chain->ref_file_num,
chain->ref_seq_num, match_info->call_number,
gen_total_length, gen_offset, ref_total_length);
/* extend the DP borders to the left and to the right */
gth_chain_extend_borders(chain, &gen_seq_bounds, &gen_seq_bounds_rc,
gen_total_length, gen_offset);
/* From here on the dp positions always refer to the forward strand of the
genomic DNA. */
/* call the Dynamic Programming */
if (refseqisdna) {
rval = call_dna_DP(directmatches, call_info, input, stat,
sa_collection, saA, gen_file_num, ref_file_num,
gen_total_length, gen_offset, &gen_seq_bounds,
&gen_seq_bounds_rc, ref_total_length, range.start,
chainctr, gth_chain_collection_size(chain_collection),
match_info, ref_seq_tran, ref_seq_orig,
ref_seq_tran_rc, ref_seq_orig_rc, chain,
dna_complete_path_matrix_jt,
protein_complete_path_matrix_jt);
}
else {
rval = call_protein_DP(directmatches, call_info, input,
stat, sa_collection, saA, gen_file_num,
ref_file_num, gen_total_length, gen_offset,
&gen_seq_bounds, &gen_seq_bounds_rc,
ref_total_length, range.start, chainctr,
gth_chain_collection_size(chain_collection),
match_info, ref_seq_tran, ref_seq_orig, chain,
dna_complete_path_matrix_jt,
protein_complete_path_matrix_jt);
}
/* check return value */
if (rval == GTH_ERROR_DP_PARAMETER_ALLOCATION_FAILED) {
/* statistics bookkeeping */
gth_stat_increment_numoffailedDPparameterallocations(stat);
gth_stat_increment_numofundeterminedSAs(stat);
/* free space */
gth_sa_delete(saA);
match_info->call_number--;
continue; /* continue with the next DP range */
}
else if (rval)
return -1;
}
if (!call_info->out->xmlout && !call_info->out->gff3out && !directmatches &&
!match_info->significant_match_found &&
match_info->call_number <= call_info->firstalshown) {
show_no_match_line(gth_input_get_alphatype(input, ref_file_num), outfp);
}
return 0;
}
static int gt_ltrdigest_pdom_visitor_feature_node(GtNodeVisitor *nv,
GtFeatureNode *fn,
GtError *err)
{
GtLTRdigestPdomVisitor *lv;
GtFeatureNodeIterator *fni;
GtFeatureNode *curnode = NULL;
int had_err = 0;
GtRange rng;
GtUword i;
lv = gt_ltrdigest_pdom_visitor_cast(nv);
gt_assert(lv);
gt_error_check(err);
/* traverse annotation subgraph and find LTR element */
fni = gt_feature_node_iterator_new(fn);
while (!had_err && (curnode = gt_feature_node_iterator_next(fni))) {
if (strcmp(gt_feature_node_get_type(curnode), lv->root_type) == 0) {
lv->ltr_retrotrans = curnode;
}
}
gt_feature_node_iterator_delete(fni);
if (!had_err && lv->ltr_retrotrans != NULL) {
GtCodonIterator *ci;
GtTranslator *tr;
GtTranslatorStatus status;
GtUword seqlen;
char translated, *rev_seq;
#ifndef _WIN32
FILE *instream;
GtHMMERParseStatus *pstatus;
#endif
unsigned int frame;
GtStr *seq;
seq = gt_str_new();
rng = gt_genome_node_get_range((GtGenomeNode*) lv->ltr_retrotrans);
lv->leftLTR_5 = rng.start - 1;
lv->rightLTR_3 = rng.end - 1;
seqlen = gt_range_length(&rng);
had_err = gt_extract_feature_sequence(seq,
(GtGenomeNode*) lv->ltr_retrotrans,
lv->root_type,
false, NULL, NULL, lv->rmap, err);
if (!had_err) {
for (i = 0UL; i < 3UL; i++) {
gt_str_reset(lv->fwd[i]);
gt_str_reset(lv->rev[i]);
}
/* create translations */
ci = gt_codon_iterator_simple_new(gt_str_get(seq), seqlen, NULL);
gt_assert(ci);
tr = gt_translator_new(ci);
status = gt_translator_next(tr, &translated, &frame, err);
while (status == GT_TRANSLATOR_OK && translated) {
gt_str_append_char(lv->fwd[frame], translated);
status = gt_translator_next(tr, &translated, &frame, NULL);
}
if (status == GT_TRANSLATOR_ERROR) had_err = -1;
if (!had_err) {
rev_seq = gt_malloc((size_t) seqlen * sizeof (char));
strncpy(rev_seq, gt_str_get(seq), (size_t) seqlen * sizeof (char));
(void) gt_reverse_complement(rev_seq, seqlen, NULL);
gt_codon_iterator_delete(ci);
ci = gt_codon_iterator_simple_new(rev_seq, seqlen, NULL);
gt_translator_set_codon_iterator(tr, ci);
status = gt_translator_next(tr, &translated, &frame, err);
while (status == GT_TRANSLATOR_OK && translated) {
gt_str_append_char(lv->rev[frame], translated);
status = gt_translator_next(tr, &translated, &frame, NULL);
}
if (status == GT_TRANSLATOR_ERROR) had_err = -1;
gt_free(rev_seq);
}
gt_codon_iterator_delete(ci);
gt_translator_delete(tr);
}
/* run HMMER and handle results */
if (!had_err) {
#ifndef _WIN32
int pid, pc[2], cp[2];
GT_UNUSED int rval;
(void) signal(SIGCHLD, SIG_IGN); /* XXX: for now, ignore child's
exit status */
rval = pipe(pc);
gt_assert(rval == 0);
rval = pipe(cp);
gt_assert(rval == 0);
switch ((pid = (int) fork())) {
case -1:
perror("Can't fork");
exit(1); /* XXX: error handling */
case 0: /* child */
(void) close(1); /* close current stdout. */
rval = dup(cp[1]); /* make stdout go to write end of pipe. */
(void) close(0); /* close current stdin. */
rval = dup(pc[0]); /* make stdin come from read end of pipe. */
(void) close(pc[0]);
(void) close(pc[1]);
(void) close(cp[0]);
(void) close(cp[1]);
(void) execvp("hmmscan", lv->args); /* XXX: read path from env */
perror("couldn't execute hmmscan!");
exit(1);
default: /* parent */
for (i = 0UL; i < 3UL; i++) {
char buf[5];
GT_UNUSED ssize_t written;
(void) sprintf(buf, ">"GT_WU"%c\n", i, '+');
written = write(pc[1], buf, 4 * sizeof (char));
written = write(pc[1], gt_str_get(lv->fwd[i]),
(size_t) gt_str_length(lv->fwd[i]) * sizeof (char));
written = write(pc[1], "\n", 1 * sizeof (char));
(void) sprintf(buf, ">"GT_WU"%c\n", i, '-');
written = write(pc[1], buf, 4 * sizeof (char));
written = write(pc[1], gt_str_get(lv->rev[i]),
(size_t) gt_str_length(lv->rev[i]) * sizeof (char));
written = write(pc[1], "\n", 1 * sizeof (char));
}
(void) close(pc[0]);
(void) close(pc[1]);
(void) close(cp[1]);
instream = fdopen(cp[0], "r");
pstatus = gt_hmmer_parse_status_new();
had_err = gt_ltrdigest_pdom_visitor_parse_output(lv, pstatus,
instream, err);
(void) fclose(instream);
if (!had_err)
had_err = gt_ltrdigest_pdom_visitor_process_hits(lv, pstatus, err);
gt_hmmer_parse_status_delete(pstatus);
}
#else
/* XXX */
gt_error_set(err, "HMMER call not implemented on Windows\n");
had_err = -1;
#endif
}
gt_str_delete(seq);
}
if (!had_err)
had_err = gt_ltrdigest_pdom_visitor_choose_strand(lv);
return had_err;
}
/* Renders a ruler with dynamic scale labeling and optional grid. */
int gt_canvas_cairo_draw_ruler(GtCanvas *canvas, GtRange viewrange,
GtError *err)
{
double step, minorstep, vmajor, vminor,
theight = gt_graphics_get_text_height(canvas->pvt->g);
long base_length, tick;
GtColor rulercol, gridcol;
GtStr *left_str, *right_str, *unit;
char str[BUFSIZ];
GtStyleQueryStatus rval;
bool showgrid = true;
gt_assert(canvas);
if (gt_style_get_bool(canvas->pvt->sty, "format", "show_grid", &showgrid,
NULL, err) == GT_STYLE_QUERY_ERROR) {
return -1;
}
if (gt_style_get_num(canvas->pvt->sty, "format", "ruler_font_size",
&theight, NULL, err) == GT_STYLE_QUERY_ERROR) {
return -1;
}
/* get unit value from style, default: empty */
unit = gt_str_new();
if (gt_style_get_str(canvas->pvt->sty,
"format", "unit",
unit, NULL, err) == GT_STYLE_QUERY_ERROR) {
gt_str_delete(unit);
return -1;
}
/* get additional description texts from style */
left_str = gt_str_new();
rval = gt_style_get_str(canvas->pvt->sty, "format",
"ruler_left_text", left_str, NULL, err);
switch (rval) {
case GT_STYLE_QUERY_NOT_SET:
gt_str_append_cstr(left_str, FIVE_PRIME_STRING);
break;
case GT_STYLE_QUERY_ERROR:
gt_str_delete(unit);
gt_str_delete(left_str);
return -1;
break; /* shouldn't reach this */
default:
break;
}
right_str = gt_str_new();
rval = gt_style_get_str(canvas->pvt->sty, "format",
"ruler_right_text", right_str, NULL, err);
switch (rval) {
case GT_STYLE_QUERY_NOT_SET:
gt_str_append_cstr(right_str, THREE_PRIME_STRING);
break;
case GT_STYLE_QUERY_ERROR:
gt_str_delete(unit);
gt_str_delete(left_str);
gt_str_delete(right_str);
return -1;
break; /* shouldn't reach this */
default:
break;
}
/* reset font to default */
gt_graphics_set_font(canvas->pvt->g,
"Sans",
SLANT_NORMAL,
WEIGHT_NORMAL,
theight);
rulercol.red = rulercol.green = rulercol.blue = RULER_GREY;
rulercol.alpha = 1.0;
gridcol.red = gridcol.green = gridcol.blue = GRID_GREY;
gridcol.alpha = 1.0;
/* determine range and step of the scale */
base_length = gt_range_length(&viewrange);
/* determine tick steps */
step = pow(10,ceil(log10(base_length))-1);
minorstep = step/10.0;
/* calculate starting positions */
vminor = (double) (floor(viewrange.start / minorstep))*minorstep;
vmajor = (double) (floor(viewrange.start / step))*step;
/* draw major ticks */
for (tick = vmajor; tick <= viewrange.end; tick += step)
{
double drawtick = (gt_coords_convert_point(viewrange, tick)
* (canvas->pvt->width-2*canvas->pvt->margins))
+ canvas->pvt->margins;
if (tick < viewrange.start) continue;
gt_graphics_draw_vertical_line(canvas->pvt->g,
drawtick,
canvas->pvt->y + 30,
rulercol,
10,
1.0);
gt_format_ruler_label(str, tick, gt_str_get(unit), BUFSIZ);
gt_graphics_draw_text_centered(canvas->pvt->g,
drawtick,
canvas->pvt->y + 20,
str);
}
/* draw minor ticks */
if (minorstep >= 1)
{
for (tick = vminor; tick <= viewrange.end; tick += minorstep)
{
double drawtick;
if (tick < viewrange.start)
continue;
drawtick = (gt_coords_convert_point(viewrange, tick)
* (canvas->pvt->width-2*canvas->pvt->margins))
+ canvas->pvt->margins;
if (showgrid)
{
gt_graphics_draw_vertical_line(canvas->pvt->g,
drawtick,
canvas->pvt->y + 40,
gridcol,
canvas->pvt->height - 40 - 15,
1.0);
}
gt_graphics_draw_vertical_line(canvas->pvt->g,
drawtick,
canvas->pvt->y + 35,
rulercol,
5,
1.0);
}
}
/* draw ruler line */
gt_graphics_draw_horizontal_line(canvas->pvt->g,
canvas->pvt->margins,
canvas->pvt->y + 40,
rulercol,
canvas->pvt->width - 2
* canvas->pvt->margins,
1.25);
gt_graphics_draw_text_right(canvas->pvt->g,
canvas->pvt->margins - 10,
canvas->pvt->y + 39 + (theight/2),
gt_str_get(left_str));
gt_graphics_draw_text(canvas->pvt->g,
canvas->pvt->width - canvas->pvt->margins + 10,
canvas->pvt->y + 39 + (theight/2),
gt_str_get(right_str));
gt_str_delete(unit);
gt_str_delete(left_str);
gt_str_delete(right_str);
return 0;
}
static int run_orffinder(GtRegionMapping *rmap,
GtFeatureNode *gf,
unsigned long start,
GT_UNUSED unsigned long end,
unsigned int min,
unsigned int max,
bool all,
GtError *err)
{
int had_err = 0, i;
unsigned long sum;
GtCodonIterator* ci = NULL;
GtTranslator* translator = NULL;
GtORFIterator* orfi = NULL;
GtORFIteratorStatus state;
GtRange orf_rng, tmp_orf_rng[3];
GtStr *seq;
unsigned int orf_frame;
/* forward strand */
seq = gt_str_new();
had_err = gt_extract_feature_sequence(seq,
(GtGenomeNode*) gf,
gt_feature_node_get_type(gf),
false, NULL, NULL, rmap, err);
ci = gt_codon_iterator_simple_new(gt_str_get(seq), gt_str_length(seq), err);
gt_assert(ci);
translator = gt_translator_new(ci);
gt_assert(translator);
orfi = gt_orf_iterator_new(ci, translator);
gt_assert(orfi);
for (i = 0; i < 3; i++) {
tmp_orf_rng[i].start = GT_UNDEF_ULONG;
tmp_orf_rng[i].end = GT_UNDEF_ULONG;
}
while ((state = gt_orf_iterator_next(orfi, &orf_rng, &orf_frame,
err)) == GT_ORF_ITERATOR_OK) {
if (all) {
process_orf(orf_rng, orf_frame, GT_STRAND_FORWARD, gf,
start, min, max, err);
} else {
if (gt_range_length(&orf_rng) >
gt_range_length(&tmp_orf_rng[orf_frame])) {
tmp_orf_rng[orf_frame].start = orf_rng.start;
tmp_orf_rng[orf_frame].end = orf_rng.end;
}
}
}
if (state == GT_ORF_ITERATOR_ERROR)
had_err = -1;
if (!had_err) {
if (!all) {
for (i = 0; i < 3; i++) {
if (tmp_orf_rng[i].start != GT_UNDEF_ULONG) {
process_orf(tmp_orf_rng[i], (unsigned int) i, GT_STRAND_FORWARD, gf,
start, min, max, err);
}
}
}
gt_codon_iterator_delete(ci);
gt_translator_delete(translator);
gt_orf_iterator_delete(orfi);
orfi = NULL;
ci = NULL;
translator = NULL;
for (i = 0; i < 3; i++) {
tmp_orf_rng[i].start = GT_UNDEF_ULONG;
tmp_orf_rng[i].end = GT_UNDEF_ULONG;
}
/* reverse strand */
if (!had_err) {
GT_UNUSED int rval = 0;
unsigned long length = gt_str_length(seq);
char *strp = (char*) gt_str_get_mem(seq);
rval = gt_reverse_complement(strp, gt_str_length(seq), err);
gt_assert(!rval); /* XXX */
ci = gt_codon_iterator_simple_new(gt_str_get(seq), gt_str_length(seq),
err);
gt_assert(ci);
translator = gt_translator_new(ci);
gt_assert(translator);
orfi = gt_orf_iterator_new(ci, translator);
gt_assert(orfi);
sum = start + length - 1;
while ((state = gt_orf_iterator_next(orfi, &orf_rng, &orf_frame,
err)) == GT_ORF_ITERATOR_OK) {
if (all) {
process_orf(orf_rng, orf_frame, GT_STRAND_REVERSE, gf,
sum, min, max, err);
} else {
if (gt_range_length(&orf_rng) >
gt_range_length(&tmp_orf_rng[orf_frame])) {
tmp_orf_rng[orf_frame].start = orf_rng.start;
tmp_orf_rng[orf_frame].end = orf_rng.end;
}
}
}
if (state == GT_ORF_ITERATOR_ERROR)
had_err = -1;
if (!had_err) {
if (!all) {
for (i = 0; i < 3; i++) {
if (tmp_orf_rng[i].start != GT_UNDEF_ULONG) {
process_orf(tmp_orf_rng[i], (unsigned int) i, GT_STRAND_REVERSE,
gf, sum, min, max, err);
}
}
}
}
}
gt_str_delete(seq);
gt_codon_iterator_delete(ci);
gt_translator_delete(translator);
gt_orf_iterator_delete(orfi);
}
return had_err;
}
static int cluster_sequences(GtArray *matches,
GtClusteredSet *cs,
GtHashmap *seqdesc2seqnum,
unsigned int psmall,
unsigned int plarge,
GtEncseq *encseq,
GtError *err)
{
GtMatch *match;
GtMatchEdgeTable matchedgetab;
GtMatchEdge matchedge;
GtRange rng_seq1,
rng_seq2;
int had_err = 0;
unsigned long i,
lsmall,
llarge,
matchlen1,
matchlen2,
num_of_seq,
seqnum1 = 0,
seqnum2 = 0;
const char *seqid;
num_of_seq = gt_encseq_num_of_sequences(encseq);
gt_assert(matches && cs && seqdesc2seqnum && encseq);
if (gt_clustered_set_num_of_elements(cs, err) != num_of_seq) {
had_err = -1;
gt_error_set(err,
"number of sequences (%lu) unequals number of elements in"
" clustered set (%lu)",
num_of_seq, gt_clustered_set_num_of_elements(cs, err));
}
if (!had_err) {
matchedgetab.edges = gt_array_new(sizeof (GtMatchEdge));
matchedgetab.num_of_edges = 0;
for (i = 0; i < gt_array_size(matches); i++) {
match = *(GtMatch**) gt_array_get(matches, i);
gt_match_get_range_seq1(match, &rng_seq1);
gt_match_get_range_seq2(match, &rng_seq2);
matchlen1 = gt_range_length(&rng_seq1);
matchlen2 = gt_range_length(&rng_seq2);
seqid = gt_match_get_seqid1(match);
if (gt_hashmap_get(seqdesc2seqnum, (void*) seqid) != NULL)
seqnum1 = ((unsigned long) gt_hashmap_get(seqdesc2seqnum, seqid)) - 1;
else {
had_err = -1;
gt_error_set(err, "key %s not found", seqid);
}
seqid = gt_match_get_seqid2(match);
if (!had_err && gt_hashmap_get(seqdesc2seqnum, (void*) seqid) != NULL)
seqnum2 = ((unsigned long) gt_hashmap_get(seqdesc2seqnum, seqid)) - 1;
else {
had_err = -1;
gt_error_set(err, "key %s not found", seqid);
}
if (!had_err) {
if (gt_encseq_seqlength(encseq, seqnum1) >
gt_encseq_seqlength(encseq, seqnum2)) {
llarge = gt_encseq_seqlength(encseq, seqnum1);
lsmall = gt_encseq_seqlength(encseq, seqnum2);
} else {
lsmall = gt_encseq_seqlength(encseq, seqnum1);
llarge = gt_encseq_seqlength(encseq, seqnum2);
}
if (((llarge * plarge)/100 <= matchlen1) &&
((lsmall * psmall)/100 <= matchlen1) &&
((llarge * plarge)/100 <= matchlen2) &&
((lsmall * psmall)/100 <= matchlen2)) {
if (seqnum1 != seqnum2) {
matchedge.matchnum0 = seqnum1;
matchedge.matchnum1 = seqnum2;
gt_array_add(matchedgetab.edges, matchedge);
matchedgetab.num_of_edges++;
}
}
}
}
}
if (!had_err)
if (gt_cluster_matches(cs, &matchedgetab, err) != 0)
had_err = -1;
if (!had_err)
gt_array_delete(matchedgetab.edges);
return had_err;
}
