static int select_visitor_region_node(GtNodeVisitor *nv, GtRegionNode *rn,
GT_UNUSED GtError *err)
{
GtSelectVisitor *select_visitor;
gt_error_check(err);
select_visitor = select_visitor_cast(nv);
if (!gt_str_length(select_visitor->seqid) || /* no seqid was specified */
!gt_str_cmp(select_visitor->seqid, /* or seqids are equal */
gt_genome_node_get_seqid((GtGenomeNode*) rn))) {
if (select_visitor->contain_range.start != GT_UNDEF_ULONG) {
GtRange range = gt_genome_node_get_range((GtGenomeNode*) rn);
if (gt_range_overlap(&range, &select_visitor->contain_range)) {
/* an overlapping contain range was defined -> update range */
range.start = MAX(range.start, select_visitor->contain_range.start);
range.end = MIN(range.end, select_visitor->contain_range.end);
gt_genome_node_set_range((GtGenomeNode*) rn, &range);
gt_queue_add(select_visitor->node_buffer, rn);
}
else /* contain range does not overlap with <rn> range -> delete <rn> */
gt_genome_node_delete((GtGenomeNode*) rn);
}
else
gt_queue_add(select_visitor->node_buffer, rn);
}
else
gt_genome_node_delete((GtGenomeNode*) rn);
return 0;
}
static int gt_sort_stream_next(GtNodeStream *ns, GtGenomeNode **gn,
GtError *err)
{
GtSortStream *sort_stream;
GtGenomeNode *node, *eofn;
int had_err = 0;
gt_error_check(err);
sort_stream = gt_sort_stream_cast(ns);
if (!sort_stream->sorted) {
while (!(had_err = gt_node_stream_next(sort_stream->in_stream, &node,
err)) && node) {
if ((eofn = gt_eof_node_try_cast(node)))
gt_genome_node_delete(eofn); /* get rid of EOF nodes */
else
gt_array_add(sort_stream->nodes, node);
}
if (!had_err) {
gt_genome_nodes_sort_stable(sort_stream->nodes);
sort_stream->sorted = true;
}
}
if (!had_err) {
gt_assert(sort_stream->sorted);
if (sort_stream->idx < gt_array_size(sort_stream->nodes)) {
*gn = *(GtGenomeNode**) gt_array_get(sort_stream->nodes,
sort_stream->idx);
sort_stream->idx++;
/* join region nodes with the same sequence ID */
if (gt_region_node_try_cast(*gn)) {
GtRange range_a, range_b;
while (sort_stream->idx < gt_array_size(sort_stream->nodes)) {
node = *(GtGenomeNode**) gt_array_get(sort_stream->nodes,
sort_stream->idx);
if (!gt_region_node_try_cast(node) ||
gt_str_cmp(gt_genome_node_get_seqid(*gn),
gt_genome_node_get_seqid(node))) {
/* the next node is not a region node with the same ID */
break;
}
range_a = gt_genome_node_get_range(*gn);
range_b = gt_genome_node_get_range(node);
range_a = gt_range_join(&range_a, &range_b);
gt_genome_node_set_range(*gn, &range_a);
gt_genome_node_delete(node);
sort_stream->idx++;
}
}
return 0;
}
}
if (!had_err) {
gt_array_reset(sort_stream->nodes);
*gn = NULL;
}
return had_err;
}
void gt_region_node_consolidate(GtRegionNode *rn_a, GtRegionNode *rn_b)
{
GtRange range_a, range_b;
gt_assert(rn_a);
gt_assert(rn_b);
gt_assert(!gt_str_cmp(gt_genome_node_get_seqid((GtGenomeNode*) rn_a),
gt_genome_node_get_seqid((GtGenomeNode*) rn_b)));
range_a = gt_genome_node_get_range((GtGenomeNode*) rn_a);
range_b = gt_genome_node_get_range((GtGenomeNode*) rn_b);
range_a = gt_range_join(&range_a, &range_b);
gt_genome_node_set_range((GtGenomeNode*) rn_a, &range_a);
}
int gt_condenseq_output_to_gff3(const GtCondenseq *condenseq,
GtError *err)
{
int had_err = 0;
GtUword idx,
name_len,
seqnum = 0, seqstart = 0, seqend = 0,
desclen;
GtStr *filename = NULL,
*id = gt_str_new_cstr("U"),
*name = gt_str_new_cstr("unique"),
*parent_unique = gt_str_new_cstr("U"),
*seqid = gt_str_new(),
*source = gt_str_new_cstr("Condenseq");
GtFile *outfile = NULL;
GtGFF3Visitor *gffv = NULL;
GtNodeVisitor *nodev = NULL;
GtFeatureNode *fnode = NULL;
GtGenomeNode *node = NULL;
GtRange range;
gt_assert(condenseq != NULL);
filename = gt_str_new_cstr(gt_condenseq_basefilename(condenseq));
name_len = gt_str_length(name);
gt_str_append_cstr(filename, ".gff3");
outfile = gt_file_new(gt_str_get(filename), "w", err);
nodev = gt_gff3_visitor_new(outfile);
gffv = (GtGFF3Visitor *) nodev;
gt_gff3_visitor_retain_id_attributes(gffv);
node = gt_feature_node_new(seqid, "experimental_feature", (GtUword) 1,
(GtUword) 1, GT_STRAND_BOTH);
fnode = (GtFeatureNode*) node;
gt_feature_node_set_source(fnode, source);
for (idx = 0; !had_err && idx < condenseq->udb_nelems; ++idx) {
GtCondenseqUnique uq = condenseq->uniques[idx];
if (seqend <= uq.orig_startpos) {
const char *desc;
gt_genome_node_delete(node);
seqnum = gt_condenseq_pos2seqnum(condenseq, uq.orig_startpos);
seqstart = gt_condenseq_seqstartpos(condenseq, seqnum);
seqend = seqstart + condenseq_seqlength_help(condenseq, seqnum, seqstart);
desc = gt_condenseq_description(condenseq, &desclen, seqnum);
gt_str_reset(seqid);
gt_str_append_cstr_nt(seqid, desc, desclen);
node = gt_feature_node_new(seqid, "experimental_feature", (GtUword) 1,
(GtUword) 1, GT_STRAND_BOTH);
fnode = (GtFeatureNode*) node;
gt_feature_node_set_source(fnode, source);
}
gt_str_set_length(name, name_len);
gt_str_append_uword(name, idx);
gt_str_set_length(id, (GtUword) 1);
gt_str_append_uword(id, idx);
gt_feature_node_set_attribute(fnode, "Name", gt_str_get(name));
gt_feature_node_set_attribute(fnode, "ID", gt_str_get(id));
/* 1 Based coordinates! */
range.start = uq.orig_startpos + 1 - seqstart;
range.end = uq.orig_startpos + uq.len - seqstart;
gt_genome_node_set_range(node, &range);
had_err = gt_genome_node_accept(node, nodev, err);
}
gt_str_reset(name);
gt_str_append_cstr(name, "link");
gt_str_reset(id);
gt_str_append_cstr(id, "L");
name_len = gt_str_length(name);
seqend = 0;
for (idx = 0; !had_err && idx < condenseq->ldb_nelems; ++idx) {
GtCondenseqLink link = condenseq->links[idx];
if (seqend <= link.orig_startpos) {
const char *desc;
gt_genome_node_delete(node);
seqnum = gt_condenseq_pos2seqnum(condenseq, link.orig_startpos);
seqstart = gt_condenseq_seqstartpos(condenseq, seqnum);
seqend = seqstart + condenseq_seqlength_help(condenseq, seqnum, seqstart);
desc = gt_condenseq_description(condenseq, &desclen, seqnum);
gt_str_reset(seqid);
gt_str_append_cstr_nt(seqid, desc, desclen);
node = gt_feature_node_new(seqid, "experimental_feature", (GtUword) 1,
(GtUword) 1, GT_STRAND_BOTH);
fnode = (GtFeatureNode*) node;
gt_feature_node_set_source(fnode, source);
}
gt_str_set_length(name, name_len);
gt_str_append_uword(name, idx);
gt_str_set_length(id, (GtUword) 1);
gt_str_append_uword(id, idx);
gt_feature_node_set_attribute(fnode, "Name", gt_str_get(name));
gt_feature_node_set_attribute(fnode, "ID", gt_str_get(id));
gt_str_set_length(parent_unique, (GtUword) 1);
gt_str_append_uword(parent_unique, link.unique_id);
gt_feature_node_set_attribute(fnode, "Derives_from",
gt_str_get(parent_unique));
/* 1 Based coordinates! */
range.start = link.orig_startpos + 1 - seqstart;
range.end = link.orig_startpos + link.len - seqstart;
gt_genome_node_set_range(node, &range);
had_err = gt_genome_node_accept(node, nodev, err);
}
gt_file_delete(outfile);
gt_genome_node_delete(node);
gt_node_visitor_delete(nodev);
gt_str_delete(filename);
gt_str_delete(id);
gt_str_delete(name);
gt_str_delete(parent_unique);
gt_str_delete(seqid);
gt_str_delete(source);
return had_err;
}
static void make_sequence_region(GtHashmap *sequence_regions,
GtStr *sequenceid,
GthRegionFactory *srf,
GthInput *input,
GtUword filenum,
GtUword seqnum)
{
GtUword offset_is_defined = false;
GtRange range, descrange;
GtGenomeNode *sr = NULL;
gt_assert(sequence_regions && sequenceid && srf && input);
if (gth_input_use_substring_spec(input)) {
range.start = gth_input_genomic_substring_from(input);
range.end = gth_input_genomic_substring_to(input);
}
else {
range = gth_input_get_relative_genomic_range(input, filenum, seqnum);
}
if (srf->use_desc_ranges) {
GtStr *description = gt_str_new();
gth_input_get_genomic_description(input, description, filenum, seqnum);
if (!gt_parse_description_range(gt_str_get(description), &descrange))
offset_is_defined = true;
gt_str_delete(description);
}
if (offset_is_defined)
range = gt_range_offset(&range, descrange.start);
else
range = gt_range_offset(&range, 1); /* 1-based */
if (!gt_str_length(sequenceid) ||
(gt_cstr_table_get(srf->used_seqids, gt_str_get(sequenceid)) &&
!offset_is_defined)) {
/* sequenceid is empty or exists already (and no offset has been parsed)
-> make one up */
GtStr *seqid;
char *base;
base = gt_basename(gth_input_get_genomic_filename(input, filenum));
seqid = gt_str_new_cstr(base);
gt_free(base);
gt_str_append_char(seqid, '|');
gt_str_append_uword(seqid, seqnum + 1); /* 1-based */
seqid_store_add(srf->seqid_store, filenum, seqnum, seqid, GT_UNDEF_UWORD);
gt_assert(!gt_cstr_table_get(srf->used_seqids, gt_str_get(seqid)));
gt_cstr_table_add(srf->used_seqids, gt_str_get(seqid));
sr = gt_region_node_new(seqid_store_get(srf->seqid_store, filenum, seqnum),
range.start, range.end);
gt_hashmap_add(sequence_regions,
(void*) gt_cstr_table_get(srf->used_seqids,
gt_str_get(seqid)),
sr);
gt_str_delete(seqid);
}
else {
/* sequenceid does not exists already (or an offset has been parsed)
-> use this one */
if (!gt_cstr_table_get(srf->used_seqids, gt_str_get(sequenceid))) {
/* no sequence region with this id exists -> create one */
gt_cstr_table_add(srf->used_seqids, gt_str_get(sequenceid));
seqid_store_add(srf->seqid_store, filenum, seqnum, sequenceid,
offset_is_defined ? descrange.start : GT_UNDEF_UWORD);
sr = gt_region_node_new(seqid_store_get(srf->seqid_store, filenum,
seqnum), range.start, range.end);
gt_hashmap_add(sequence_regions,
(void*) gt_cstr_table_get(srf->used_seqids,
gt_str_get(sequenceid)),
sr);
}
else {
GtRange prev_range, new_range;
/* sequence region with this id exists already -> modify range */
sr = gt_hashmap_get(sequence_regions, gt_str_get(sequenceid));
gt_assert(sr);
prev_range = gt_genome_node_get_range(sr);
new_range = gt_range_join(&prev_range, &range);
gt_genome_node_set_range(sr, &new_range);
seqid_store_add(srf->seqid_store, filenum, seqnum, sequenceid,
offset_is_defined ? descrange.start : GT_UNDEF_UWORD);
}
}
gt_assert(sr);
}
static int add_ids_visitor_feature_node(GtNodeVisitor *nv, GtFeatureNode *fn,
GtError *err)
{
AutomaticSequenceRegion *auto_sr;
GtAddIDsVisitor *aiv;
const char *seqid;
bool is_circular;
aiv = add_ids_visitor_cast(nv);
seqid = gt_str_get(gt_genome_node_get_seqid((GtGenomeNode*) fn));
if (aiv->ensure_sorting && !gt_cstr_table_get(aiv->defined_seqids, seqid)) {
gt_error_set(err, "the file %s is not sorted (seqid \"%s\" on line %u has "
"not been previously introduced with a \"%s\" line)",
gt_genome_node_get_filename((GtGenomeNode*) fn), seqid,
gt_genome_node_get_line_number((GtGenomeNode*) fn),
GT_GFF_SEQUENCE_REGION);
return -1;
}
if (!gt_cstr_table_get(aiv->defined_seqids, seqid)) {
GtFeatureNodeIterator *fni;
GtFeatureNode *node;
GtRange range = gt_genome_node_get_range((GtGenomeNode*) fn);
is_circular = gt_feature_node_get_attribute(fn, GT_GFF_IS_CIRCULAR)
? true : false;
if (!is_circular) {
fni = gt_feature_node_iterator_new(fn);
while ((node = gt_feature_node_iterator_next(fni))) {
GtRange node_range = gt_genome_node_get_range((GtGenomeNode*) node);
range = gt_range_join(&range, &node_range);
}
gt_feature_node_iterator_delete(fni);
}
/* sequence region has not been previously introduced -> check if one has
already been created automatically */
auto_sr = gt_hashmap_get(aiv->undefined_sequence_regions, seqid);
if (!auto_sr) {
GtStr *seqid_str;
/* sequence region has not been createad automatically -> do it now */
gt_warning("seqid \"%s\" on line %u in file \"%s\" has not been "
"previously introduced with a \"%s\" line, create such a line "
"automatically", seqid,
gt_genome_node_get_line_number((GtGenomeNode*) fn),
gt_genome_node_get_filename((GtGenomeNode*) fn),
GT_GFF_SEQUENCE_REGION);
auto_sr = automatic_sequence_region_new(is_circular);
seqid_str = gt_genome_node_get_seqid((GtGenomeNode*) fn);
auto_sr->sequence_region = gt_region_node_new(seqid_str, range.start,
range.end);
gt_hashmap_add(aiv->undefined_sequence_regions, gt_str_get(seqid_str),
auto_sr);
}
else {
if (auto_sr->is_circular) {
gt_assert(!is_circular); /* XXX */
}
else if (is_circular) {
gt_assert(!auto_sr->is_circular); /* XXX */
auto_sr->is_circular = true;
gt_genome_node_set_range(auto_sr->sequence_region, &range);
}
else {
GtRange joined_range,
sr_range = gt_genome_node_get_range(auto_sr->sequence_region);
/* update the range of the sequence region */
joined_range = gt_range_join(&range, &sr_range);
gt_genome_node_set_range(auto_sr->sequence_region, &joined_range);
}
}
gt_array_add(auto_sr->feature_nodes, fn);
}
else
gt_queue_add(aiv->node_buffer, fn);
return 0;
}
static int construct_mRNAs(GT_UNUSED void *key, void *value, void *data,
GtError *err)
{
ConstructionInfo *cinfo = (ConstructionInfo*) data;
GtArray *gt_genome_node_array = (GtArray*) value,
*mRNAs = (GtArray*) cinfo->mRNAs;
GtGenomeNode *mRNA_node, *first_node, *gn;
const char *tname;
GtStrand mRNA_strand;
GtRange mRNA_range;
GtStr *mRNA_seqid;
GtUword i;
int had_err = 0;
gt_error_check(err);
gt_assert(key && value && data);
/* at least one node in array */
gt_assert(gt_array_size(gt_genome_node_array));
/* determine the range and the strand of the mRNA */
first_node = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, 0);
mRNA_range = gt_genome_node_get_range(first_node);
mRNA_strand = gt_feature_node_get_strand((GtFeatureNode*) first_node);
mRNA_seqid = gt_genome_node_get_seqid(first_node);
/* TODO: support discontinuous start/stop codons */
for (i = 0; !had_err && i < gt_array_size(gt_genome_node_array); i++) {
gn = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, i);
if (gt_feature_node_get_attribute((GtFeatureNode*) gn,
GTF_PARSER_STOP_CODON_FLAG)) {
GtUword j;
GtRange stop_codon_rng = gt_genome_node_get_range(gn);
bool found_cds = false;
for (j = 0; !had_err && j < gt_array_size(gt_genome_node_array); j++) {
GtGenomeNode* gn2;
GtRange this_rng;
const char *this_type;
gn2 = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, j);
if (gn == gn2) continue;
this_rng = gt_genome_node_get_range(gn2);
this_type = gt_feature_node_get_type((GtFeatureNode*) gn2);
if (this_type == gt_symbol(gt_ft_CDS)) {
if (gt_range_contains(&this_rng, &stop_codon_rng)) {
if (cinfo->tidy) {
gt_warning("stop codon on line %u in file %s is contained in "
"CDS in line %u",
gt_genome_node_get_line_number(gn),
gt_genome_node_get_filename(gn),
gt_genome_node_get_line_number(gn2));
found_cds = true;
} else {
gt_error_set(err, "stop codon on line %u in file %s is "
"contained in CDS in line %u",
gt_genome_node_get_line_number(gn),
gt_genome_node_get_filename(gn),
gt_genome_node_get_line_number(gn2));
had_err = -1;
}
break;
}
if (this_rng.end + 1 == stop_codon_rng.start) {
this_rng.end = stop_codon_rng.end;
gt_genome_node_set_range(gn2, &this_rng);
found_cds = true;
break;
}
if (this_rng.start == stop_codon_rng.end + 1) {
this_rng.start = stop_codon_rng.start;
gt_genome_node_set_range(gn2, &this_rng);
found_cds = true;
break;
}
}
}
if (!found_cds) {
if (!had_err) {
if (cinfo->tidy) {
gt_warning("found stop codon on line %u in file %s with no "
"flanking CDS, ignoring it",
gt_genome_node_get_line_number(gn),
gt_genome_node_get_filename(gn));
} else {
gt_error_set(err, "found stop codon on line %u in file %s with no "
"flanking CDS",
gt_genome_node_get_line_number(gn),
gt_genome_node_get_filename(gn));
had_err = -1;
break;
}
}
} else {
gt_array_rem(gt_genome_node_array, i);
gt_genome_node_delete(gn);
}
}
}
for (i = 1; !had_err && i < gt_array_size(gt_genome_node_array); i++) {
GtRange range;
GtStrand strand;
gn = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, i);
range = gt_genome_node_get_range(gn);
mRNA_range = gt_range_join(&mRNA_range, &range);
strand = gt_feature_node_get_strand((GtFeatureNode*) gn);
if (strand != mRNA_strand) {
gt_error_set(err, "feature %s on line %u has strand %c, but the "
"parent transcript has strand %c",
(const char*) key,
gt_genome_node_get_line_number(gn),
GT_STRAND_CHARS[strand],
GT_STRAND_CHARS[mRNA_strand]);
had_err = -1;
break;
} else {
mRNA_strand = gt_strand_join(mRNA_strand, strand);
}
if (!had_err && gt_str_cmp(mRNA_seqid, gt_genome_node_get_seqid(gn))) {
gt_error_set(err, "The features on lines %u and %u refer to different "
"genomic sequences (``seqname''), although they have the same "
"gene IDs (``gene_id'') which must be globally unique",
gt_genome_node_get_line_number(first_node),
gt_genome_node_get_line_number(gn));
had_err = -1;
break;
}
}
if (!had_err) {
mRNA_node = gt_feature_node_new(mRNA_seqid, gt_ft_mRNA, mRNA_range.start,
mRNA_range.end, mRNA_strand);
gt_feature_node_add_attribute(((GtFeatureNode*) mRNA_node), "ID", key);
gt_feature_node_add_attribute(((GtFeatureNode*) mRNA_node), "transcript_id",
key);
if ((tname = gt_hashmap_get(cinfo->transcript_id_to_name_mapping,
(const char*) key)) && strlen(tname) > 0) {
gt_feature_node_add_attribute((GtFeatureNode*) mRNA_node, GT_GFF_NAME,
tname);
}
/* register children */
for (i = 0; i < gt_array_size(gt_genome_node_array); i++) {
gn = *(GtGenomeNode**) gt_array_get(gt_genome_node_array, i);
gt_feature_node_add_child((GtFeatureNode*) mRNA_node,
(GtFeatureNode*) gt_genome_node_ref(gn));
}
/* store the mRNA */
gt_array_add(mRNAs, mRNA_node);
}
return had_err;
}
