PyObject *pysplicing_from_strvector(const splicing_strvector_t *v) {
int i, n=splicing_strvector_size(v);
PyObject *o=PyTuple_New(n);
for (i=0; i<n; i++) {
PyObject *it=PyString_FromString(splicing_strvector_get(v, i));
PyTuple_SetItem(o, i, it);
}
return o;
}
int splicing_i_gff_constitutive_exons_all(const splicing_gff_t *gff,
splicing_exonset_t *exons,
int min_length) {
size_t g, nogenes;
splicing_vector_int_t events;
SPLICING_CHECK(splicing_gff_nogenes(gff, &nogenes));
SPLICING_CHECK(splicing_exonset_init(exons, 0));
SPLICING_FINALLY(splicing_exonset_destroy, exons);
SPLICING_CHECK(splicing_vector_int_init(&events, 0));
SPLICING_FINALLY(splicing_vector_int_destroy, &events);
for (g=0; g<nogenes; g++) {
const char *seqid=
splicing_strvector_get(&gff->seqids, VECTOR(gff->seqid)[g]);
int noex, idx, noEvents, i;
size_t noiso;
int start=VECTOR(gff->genes)[g];
int end= g+1 < nogenes ? VECTOR(gff->genes)[g+1] : gff->n;
splicing_gff_noiso_one(gff, g, &noiso);
/* Collect and sort all events */
splicing_vector_int_clear(&events);
for (idx=start+1; idx<end; idx++) {
if (VECTOR(gff->type)[idx] == SPLICING_TYPE_EXON) {
SPLICING_CHECK(splicing_vector_int_push_back2
(&events, VECTOR(gff->start)[idx],
-VECTOR(gff->end)[idx]));
}
}
noEvents=splicing_vector_int_size(&events);
splicing_qsort(VECTOR(events), noEvents, sizeof(int),
splicing_i_const_cmp);
/* Now go over the sorted events and extract the constitutive exons */
for (noex=0, i=0; i<noEvents; i++) {
int ev=VECTOR(events)[i];
if (ev > 0) { noex++; }
if (ev < 0) {
int prev=VECTOR(events)[i-1];
if (noex == noiso && (-ev)-prev+1 >= min_length) {
/* constitutive exon */
SPLICING_CHECK(splicing_exonset_append(exons, seqid, prev, -ev));
}
noex--;
}
}
}
splicing_vector_int_destroy(&events);
SPLICING_FINALLY_CLEAN(2); /* + exons */
return 0;
}
int splicing_i_gff_constitutive_exons_full(const splicing_gff_t *gff,
splicing_exonset_t *exons,
int min_length) {
size_t g, nogenes;
splicing_vector_int_t events;
SPLICING_CHECK(splicing_gff_nogenes(gff, &nogenes));
SPLICING_CHECK(splicing_exonset_init(exons, 1));
SPLICING_FINALLY(splicing_exonset_destroy, exons);
SPLICING_CHECK(splicing_vector_int_init(&events, 0));
SPLICING_FINALLY(splicing_vector_int_destroy, &events);
for (g=0; g<nogenes; g++) {
const char *seqid=
splicing_strvector_get(&gff->seqids, VECTOR(gff->seqid)[g]);
int i, idx, noExons, noSame;
size_t noiso;
int start=VECTOR(gff->genes)[g];
int end= g+1 < nogenes ? VECTOR(gff->genes)[g+1] : gff->n;
splicing_gff_noiso_one(gff, g, &noiso);
/* Collect and sort all events */
splicing_vector_int_clear(&events);
for (idx=start+1; idx<end; idx++) {
if (VECTOR(gff->type)[idx] == SPLICING_TYPE_EXON) {
SPLICING_CHECK(splicing_vector_int_push_back2
(&events, VECTOR(gff->start)[idx],
VECTOR(gff->end)[idx]));
}
}
noExons=splicing_vector_int_size(&events)/2;
splicing_qsort(VECTOR(events), noExons, sizeof(int)*2,
splicing_i_const_cmp2);
/* Now go over them and check how many times each exon appears */
for (noSame=1, i=2; i<noExons*2; i+=2) {
int start=VECTOR(events)[i];
int end=VECTOR(events)[i+1];
if (start == VECTOR(events)[i-2] && VECTOR(events)[i-1] == end) {
noSame++;
} else {
noSame=1;
}
if (noSame == noiso) {
SPLICING_CHECK(splicing_exonset_append(exons, seqid, start, end));
}
}
}
splicing_vector_int_destroy(&events);
SPLICING_FINALLY_CLEAN(2); /* + exons */
return 0;
}
int splicing_gff_write(FILE *output, const splicing_gff_t *gff) {
size_t i, n=gff->n;
size_t gene=-1;
const char *types[] = { "gene", "mRNA", "exon", "CDS", "start_codon",
"stop_codon" };
const char *strands[] = { "+", "-", "." };
for (i=0; i<n; i++) {
if (VECTOR(gff->type)[i] == SPLICING_TYPE_GENE) { gene++; }
fprintf(output, "%s\t%s\t%s\t%li\t%li\t",
splicing_strvector_get(&gff->seqids, VECTOR(gff->seqid)[gene]),
splicing_strvector_get(&gff->sources, VECTOR(gff->source)[gene]),
types[VECTOR(gff->type)[i]], (long) VECTOR(gff->start)[i],
(long) VECTOR(gff->end)[i]);
if (VECTOR(gff->score)[i] == SPLICING_NA_REAL) {
fprintf(output, "%s\t", ".");
} else {
fprintf(output, "%g\t", VECTOR(gff->score)[i]);
}
fprintf(output, "%s\t", strands[gene]);
if (VECTOR(gff->phase)[i] == SPLICING_NA_INTEGER) {
fprintf(output, "%s\t", ".");
} else {
fprintf(output, "%i\t", VECTOR(gff->phase)[i]);
}
fprintf(output, "ID=%s", splicing_strvector_get(&gff->ID, i));
if (VECTOR(gff->parent)[i] >= 0) {
fprintf(output, ";Parent=%s",
splicing_strvector_get(&gff->ID, VECTOR(gff->parent)[i]));
}
fputs("\n", output);
}
return 0;
}
int splicing_gff_append(splicing_gff_t *gff, const char *seqid,
const char *source, splicing_type_t type, int start,
int end, double score, splicing_strand_t strand,
int phase, const char *ID, const char *parent) {
if (type == SPLICING_TYPE_GENE) {
gff->nogenes++;
SPLICING_CHECK(splicing_vector_int_push_back(&gff->genes, gff->n));
SPLICING_CHECK(splicing_vector_int_push_back(&gff->strand, strand));
} else if (type == SPLICING_TYPE_MRNA) {
gff->notranscripts++;
SPLICING_CHECK(splicing_vector_int_push_back(&gff->transcripts, gff->n));
}
if (type == SPLICING_TYPE_GENE) {
/* Seqid */
if (!strcmp(seqid, gff->last_seqid)) {
int last=splicing_vector_int_tail(&gff->seqid);
SPLICING_CHECK(splicing_vector_int_push_back(&gff->seqid, last));
} else {
size_t idx;
int seen=splicing_strvector_search(&gff->seqids, seqid, &idx);
if (seen) {
SPLICING_CHECK(splicing_vector_int_push_back(&gff->seqid, idx));
gff->last_seqid=splicing_strvector_get(&gff->seqids, idx);
} else {
size_t size=splicing_strvector_size(&gff->seqids);
SPLICING_CHECK(splicing_strvector_append(&gff->seqids, seqid));
SPLICING_CHECK(splicing_vector_int_push_back(&gff->seqid, size));
gff->last_source=splicing_strvector_get(&gff->seqids, size);
}
}
/* Source */
if (!strcmp(source, gff->last_source)) {
int last=splicing_vector_int_tail(&gff->source);
SPLICING_CHECK(splicing_vector_int_push_back(&gff->source, last));
} else {
size_t idx;
int seen=splicing_strvector_search(&gff->sources, source, &idx);
if (seen) {
SPLICING_CHECK(splicing_vector_int_push_back(&gff->source, idx));
gff->last_source=splicing_strvector_get(&gff->sources, idx);
} else {
size_t size=splicing_strvector_size(&gff->sources);
SPLICING_CHECK(splicing_strvector_append(&gff->sources, source));
SPLICING_CHECK(splicing_vector_int_push_back(&gff->source, size));
gff->last_source=splicing_strvector_get(&gff->sources, size);
}
}
}
/* Parent */
if (!parent || !parent[0]) {
SPLICING_CHECK(splicing_vector_int_push_back(&gff->parent, -1));
} else if (!strcmp(parent, gff->last_gene_id)) {
SPLICING_CHECK(splicing_vector_int_push_back(&gff->parent,
gff->last_gene_no));
} else if (!strcmp(parent, gff->last_mrna_id)) {
SPLICING_CHECK(splicing_vector_int_push_back(&gff->parent,
gff->last_mrna_no));
} else {
size_t idx;
int seen=splicing_strvector_search(&gff->ID, parent, &idx);
if (!seen) {
SPLICING_WARNING("Unknown parent ID, invalid GFF file");
SPLICING_CHECK(splicing_vector_int_push_back(&gff->parent, -1));
} else {
SPLICING_CHECK(splicing_vector_int_push_back(&gff->parent, idx));
}
}
SPLICING_CHECK(splicing_vector_int_push_back(&gff->type, type));
SPLICING_CHECK(splicing_vector_int_push_back(&gff->start, start));
SPLICING_CHECK(splicing_vector_int_push_back(&gff->end, end));
SPLICING_CHECK(splicing_vector_push_back(&gff->score, score));
SPLICING_CHECK(splicing_vector_int_push_back(&gff->phase, phase));
SPLICING_CHECK(splicing_strvector_append(&gff->ID, ID));
/* Update last gene/mrna */
if (type == SPLICING_TYPE_GENE) {
gff->last_gene_id = splicing_strvector_get(&gff->ID, gff->n);
gff->last_gene_no = gff->n;
} else if (type == SPLICING_TYPE_MRNA) {
gff->last_mrna_id = splicing_strvector_get(&gff->ID, gff->n);
gff->last_mrna_no = gff->n;
}
gff->n += 1;
return 0;
}