int splicing_create_gene(const splicing_vector_int_t *exons,
const splicing_vector_int_t *isoforms,
const char *id, const char *seqid,
const char *source, splicing_strand_t strand,
splicing_gff_t *extend) {
size_t i=0;
size_t exlen=splicing_vector_int_size(exons);
size_t isolen=splicing_vector_int_size(isoforms);
size_t genestart=splicing_vector_int_min(exons);
size_t geneend=splicing_vector_int_max(exons);
char buffer[5000], buffer2[5000];
int noiso=0;
/* TODO: error checks */
/* Gene */
SPLICING_CHECK(splicing_gff_append(extend, seqid, source,
SPLICING_TYPE_GENE,
genestart, geneend,
/*score=*/ SPLICING_NA_REAL,
strand, /*phase=*/ SPLICING_NA_INTEGER,
id, /*parent=*/ 0));
while (i<isolen) {
size_t mmin=VECTOR(*exons)[ 2*VECTOR(*isoforms)[i] ];
size_t mmax=VECTOR(*exons)[ 2*VECTOR(*isoforms)[i] + 1 ];
size_t j, exon=0;
for (j=i+1; VECTOR(*isoforms)[j] >= 0; j++) {
size_t m1=VECTOR(*exons)[ 2*VECTOR(*isoforms)[j] ];
size_t m2=VECTOR(*exons)[ 2*VECTOR(*isoforms)[j] + 1 ];
if (m1 < mmin) { mmin = m1; }
if (m2 > mmax) { mmax = m2; }
}
snprintf(buffer, sizeof(buffer)/sizeof(char)-sizeof(char),
"%s-isoform-%i", id, noiso);
SPLICING_CHECK(splicing_gff_append(extend, seqid, source,
SPLICING_TYPE_MRNA, mmin, mmax,
/*score=*/ SPLICING_NA_REAL, strand,
/*phase=*/ SPLICING_NA_INTEGER,
buffer, /*parent=*/ id));
for (; VECTOR(*isoforms)[i] >= 0; i++) {
snprintf(buffer2, sizeof(buffer2)/sizeof(char)-sizeof(char),
"%s-isoform-%i-exon-%i", id, (int) noiso, (int) exon++);
SPLICING_CHECK(splicing_gff_append(extend, seqid, source,
SPLICING_TYPE_EXON,
VECTOR(*exons)[ 2*VECTOR(*isoforms)[i] ],
VECTOR(*exons)[ 2*VECTOR(*isoforms)[i] + 1 ],
/*score=*/ SPLICING_NA_REAL, strand,
/*phase=*/ SPLICING_NA_INTEGER, buffer2,
/*parent=*/ buffer));
}
noiso++;
i++;
}
return 0;
}
int splicing_gff_read(FILE *input, splicing_gff_t *gff) {
int eof=!EOF;
char seqid[200];
char source[200];
char type[200];
int start, end, phase;
double score;
char strand[10];
char attributes[5000];
size_t len;
splicing_type_t realtype;
splicing_strand_t realstrand=SPLICING_STRAND_UNKNOWN;
char *ID, *parent;
do {
eof = eof || splicing_io_skip_newline_and_comments(input);
eof = eof || splicing_io_get_string(input, seqid,
sizeof(seqid)/sizeof(char), &len,
/*delim=*/ '\t', /*newline=*/ 0);
eof = eof || splicing_io_get_string(input, source,
sizeof(source)/sizeof(char), &len,
/*delim=*/ '\t', /*newline=*/ 0);
eof = eof || splicing_io_get_string(input, type,
sizeof(type)/sizeof(char), &len,
/*delim=*/ '\t', /*newline=*/ 0);
eof = eof || splicing_io_get_integer(input, &start, /*delim=*/ '\t');
eof = eof || splicing_io_get_integer(input, &end, /*delim=*/ '\t');
eof = eof || splicing_io_get_real_na(input, &score, /*delim=*/ '\t',
/*nachar=*/ '.');
eof = eof || splicing_io_get_string(input, strand,
sizeof(strand)/sizeof(char), &len,
/*delim=*/ '\t', /*newline=*/ 0);
eof = eof || splicing_io_get_integer_na(input, &phase, /*delim=*/ '\t',
/*nachar=*/ '.');
eof = eof || splicing_io_get_string(input, attributes,
sizeof(attributes)/sizeof(char), &len,
/*delim=*/ '\n', /*newline=*/ 1);
if (eof) {
SPLICING_ERROR("Corrupt GFF file", SPLICING_PARSEERROR);
}
/* TODO: do not hardcode these names
TODO: order them according to their frequency */
if (!strcmp(type, "gene")) {
realtype = SPLICING_TYPE_GENE;
} else if (!strcmp(type, "mRNA")) {
realtype = SPLICING_TYPE_MRNA;
} else if (!strcmp(type, "exon")) {
realtype = SPLICING_TYPE_EXON;
} else if (!strcmp(type, "CDS")) {
realtype = SPLICING_TYPE_CDS;
} else if (!strcmp(type, "start_codon")) {
realtype = SPLICING_TYPE_START_CODON;
} else if (!strcmp(type, "stop_codon")) {
realtype = SPLICING_TYPE_STOP_CODON;
} else {
SPLICING_ERROR("Invalid GFF file", SPLICING_PARSEERROR);
}
if (!strcmp(strand, "+")) {
realstrand=SPLICING_STRAND_PLUS;
} else if (!strcmp(strand, "-")) {
realstrand=SPLICING_STRAND_MINUS;
} else if (!strcmp(strand, ".")) {
realstrand=SPLICING_STRAND_UNKNOWN;
}
/* Parsing the attributes field */
SPLICING_CHECK(splicing_io_parse_attributes(attributes, &ID, &parent));
SPLICING_CHECK(splicing_gff_append(gff, seqid, source, realtype, start,
end, score, realstrand, phase, ID,
parent));
eof = splicing_io_skip_newline_and_comments(input);
} while (!eof);
return 0;
}
static PyObject* pysplicing_to_gff(PyObject *self, PyObject *args) {
PyObject *pygff, *entries;
size_t i, noRec;
splicing_gff_t *cgff;
PyObject *IDkey, *Parentkey;
if (!PyArg_ParseTuple(args, "O", &pygff)) { return NULL; }
if (!PyObject_HasAttrString(pygff, "_GFFDatabase__entries")) {
splicingmodule_handle_splicing_error();
return NULL;
}
entries=PyObject_GetAttrString(pygff, "_GFFDatabase__entries");
noRec=PySequence_Size(entries);
IDkey=PyString_FromString("ID");
Parentkey=PyString_FromString("Parent");
cgff=malloc(sizeof(splicing_gff_t));
if (!cgff) { splicingmodule_handle_splicing_error(); return NULL; }
splicing_gff_init(cgff, noRec);
for (i=0; i<noRec; i++) {
PyObject *rec=0, *seqid=0, *source=0, *type=0, *start=0, *end=0,
*score=0, *strand=0, *phase=0, *attributes=0, *ID=0, *Parent=0;
char *Cseqid, *Csource, *CID, *Cparent=0, *Ctype;
splicing_type_t Ctype2;
int Cstart, Cend, Cphase;
double Cscore;
splicing_strand_t Cstrand;
rec=PySequence_GetItem(entries, i);
seqid=PyObject_GetAttrString(rec, "seqid");
source=PyObject_GetAttrString(rec, "source");
type=PyObject_GetAttrString(rec, "type");
start=PyObject_GetAttrString(rec, "start");
end=PyObject_GetAttrString(rec, "end");
score=PyObject_GetAttrString(rec, "score");
strand=PyObject_GetAttrString(rec, "strand");
phase=PyObject_GetAttrString(rec, "phase");
attributes=PyObject_GetAttrString(rec, "attributes");
ID=PyDict_GetItem(attributes, IDkey);
Parent=PyDict_GetItem(attributes, Parentkey);
Cseqid=PyString_AsString(seqid);
Csource=PyString_AsString(source);
Ctype=PyString_AsString(type);
if (!strcmp(Ctype, "gene")) {
Ctype2=SPLICING_TYPE_GENE;
} else if (!strcmp(Ctype, "mRNA")) {
Ctype2=SPLICING_TYPE_MRNA;
} else if (!strcmp(Ctype, "exon")) {
Ctype2=SPLICING_TYPE_EXON;
} else if (!strcmp(Ctype, "CDS")) {
Ctype2=SPLICING_TYPE_CDS;
} else if (!strcmp(Ctype, "start_codon")) {
Ctype2=SPLICING_TYPE_START_CODON;
} else if (!strcmp(Ctype, "stop_codon")) {
Ctype2=SPLICING_TYPE_STOP_CODON;
} /* TODO: else error? */
Cstart=PyInt_AsLong(start);
Cend=PyInt_AsLong(end);
Cscore=PyFloat_AsDouble(score);
Cstrand=PyInt_AsLong(strand);
Cphase=PyInt_AsLong(phase);
CID=PyString_AsString(ID);
if (Parent) { Cparent=PyString_AsString(Parent); }
SPLICING_PYCHECK(splicing_gff_append(cgff, Cseqid, Csource, Ctype2,
Cstart, Cend, Cscore, Cstrand,
Cphase, CID, Cparent));
Py_DECREF(rec); Py_DECREF(seqid); Py_DECREF(source); Py_DECREF(type);
Py_DECREF(start); Py_DECREF(end); Py_DECREF(score); Py_DECREF(strand);
Py_DECREF(phase); Py_DECREF(attributes);
}
Py_DECREF(entries);
Py_DECREF(IDkey);
Py_DECREF(Parentkey);
return PyCObject_FromVoidPtr(cgff, splicing_gff_destroy2);
}
