Vector *HomologyAdaptor_listStableIdsFromSpecies(HomologyAdaptor *ha, char *sp) {
StatementHandle *sth;
ResultRow *row;
Vector *genes;
char qStr[1024];
char *species;
species = StrUtil_copyString(&species,sp,0);
species = StrUtil_strReplChr(species,'_',' ');
sprintf(qStr,
"select distinct grm.member_stable_id "
" from gene_relationship_member grm,"
" genome_db gd "
" where gd.genome_db_id = grm.genome_db_id "
" and gd.name = '%s'", species);
sth = ha->prepare((BaseAdaptor *)ha, qStr, strlen(qStr));
sth->execute(sth);
genes = Vector_new();
while ((row = sth->fetchRow(sth))) {
char *tmpStr;
Vector_addElement(genes,StrUtil_copyString(&tmpStr, row->getStringAt(row,0),0));
}
sth->finish(sth);
free(species);
return genes;
}
IDType SyntenyRegionAdaptor_store(SyntenyRegionAdaptor *sra, IDType clusterId, SyntenyRegion *region) {
IDType regionId = 0;
StatementHandle *sth;
char qStr[1024];
if (!region) {
fprintf(stderr, "Error: region is not a SyntenyRegion\n");
} else {
sprintf(qStr, "insert into synteny_region (synteny_cluster_id,dnafrag_id,seq_start,seq_end)"
" VALUES (" IDFMTSTR ", " IDFMTSTR ", %d, %d)",
clusterId,
SyntenyRegion_getDNAFragId(region),
SyntenyRegion_getSeqStart(region),
SyntenyRegion_getSeqEnd(region));
sth = sra->prepare((BaseAdaptor *)sra, qStr, strlen(qStr));
sth->execute(sth);
regionId = sth->getInsertId(sth);
SyntenyRegion_setDbID(region, regionId);
SyntenyRegion_setAdaptor(region, sra);
}
return regionId;
}
Chromosome *ChromosomeAdaptor_fetchByChrName(ChromosomeAdaptor *ca, char *chrName) {
Chromosome *chromosome;
char qStr[256];
StatementHandle *sth;
ResultRow *row;
if (StringHash_contains(ca->chrNameCache,chrName)) {
chromosome = StringHash_getValue(ca->chrNameCache, chrName);
} else {
sprintf(qStr,"SELECT chromosome_id, name, length"
" FROM chromosome"
" WHERE name = '%s'", chrName);
sth = ca->prepare((BaseAdaptor *)ca,qStr,strlen(qStr));
sth->execute(sth);
row = sth->fetchRow(sth);
if( row == NULL ) {
sth->finish(sth);
fprintf(stderr, "ERROR: Do not recognise chromosome %s\n",chrName);
return NULL;
}
chromosome = ChromosomeAdaptor_chromosomeFromRow(ca, row);
sth->finish(sth);
}
return chromosome;
}
char *GenomicAlignAdaptor_alignmentTypeByMethodLinkId(GenomicAlignAdaptor *gaa, IDType methodLinkId) {
StatementHandle *sth;
ResultRow *row;
char qStr[512];
char *alignmentType = NULL;
int ok = 1;
if (!methodLinkId) {
fprintf(stderr, "Error: methodLinkId has to be defined");
ok = 0;
}
if (ok) {
sprintf(qStr,"SELECT type FROM method_link WHERE method_link_id = " IDFMTSTR, methodLinkId);
sth = gaa->prepare((BaseAdaptor *)gaa, qStr, strlen(qStr));
sth->execute(sth);
if ((row = sth->fetchRow(sth))) {
alignmentType = StrUtil_copyString(&alignmentType, row->getStringAt(row,0), 0);
} else {
fprintf(stderr,"Error: No alignmentType for " IDFMTSTR "\n",methodLinkId);
ok = 0;
}
}
if (ok) {
sth->finish(sth);
}
// NIY switch to using passed in string
return alignmentType;
}
IDType GenomicAlignAdaptor_methodLinkIdByAlignmentType(GenomicAlignAdaptor *gaa, char *alignmentType) {
StatementHandle *sth;
ResultRow *row;
IDType methodLinkId = 0;
char qStr[512];
int ok = 1;
if (!alignmentType) {
fprintf(stderr, "Error: alignment_type has to be defined\n");
ok = 0;
}
if (ok) {
sprintf(qStr, "SELECT method_link_id FROM method_link WHERE type = '%s'", alignmentType);
sth = gaa->prepare((BaseAdaptor *)gaa, qStr, strlen(qStr));
sth->execute(sth);
if ((row = sth->fetchRow(sth))) {
methodLinkId = row->getLongLongAt(row,0);
} else {
fprintf(stderr,"Error: No methodLinkId for %s\n",alignmentType);
ok = 0;
}
}
if (ok) {
sth->finish(sth);
}
return methodLinkId;
}
// Removed the circular stuff
void AttributeAdaptor_doStoreAllByTypeAndTableAndID(AttributeAdaptor *ata, char *type, char *table, IDType objectId, Vector *attributes) {
int ok = 1;
char qStr[1024];
sprintf(qStr, "INSERT into %s_attrib SET %s_id = %"IDFMTSTR", attrib_type_id = %"IDFMTSTR", value = '%%s'", table, type);
StatementHandle *sth = ata->prepare((BaseAdaptor *)ata,qStr,strlen(qStr));
int i;
for (i=0; i<Vector_getNumElement(attributes); i++) {
Attribute *attrib = Vector_getElementAt(attributes, i);
if (attrib == NULL ) {
fprintf(stderr, "Reference to list of Bio::EnsEMBL::Attribute objects argument expected.\n");
ok = 0;
break;
}
Class_assertType(CLASS_ATTRIBUTE, attrib->objectType);
IDType atId = AttributeAdaptor_storeType(ata, attrib);
sth->execute(sth, objectId, atId, Attribute_getValue(attrib));
}
if (ok) {
sth->finish(sth);
}
return;
}
Vector *IntronSupportingEvidenceAdaptor_listLinkedTranscriptIds(IntronSupportingEvidenceAdaptor *isea, IntronSupportingEvidence *ise) {
char qStr[1024];
sprintf(qStr,"SELECT transcript_id from transcript_intron_supporting_evidence "
"WHERE intron_supporting_evidence_id = "IDFMTSTR, IntronSupportingEvidence_getDbID(ise));
StatementHandle *sth = isea->prepare((BaseAdaptor *)isea,qStr,strlen(qStr));
sth->execute(sth);
Vector *idVec = Vector_new();
ResultRow *row;
while ((row = sth->fetchRow(sth))) {
IDType id = row->getLongLongAt(row, 0);
IDType *idP;
if ((idP = calloc(1,sizeof(IDType))) == NULL) {
fprintf(stderr, "Failed allocating space for a id\n");
exit(1);
}
*idP = id;
Vector_addElement(idVec, idP);
}
sth->finish(sth);
Vector_setFreeFunc(idVec, free);
return idVec;
}
int BaseAdaptor_genericCount(BaseAdaptor *ba, char *constraint) {
char *qStr = NULL;
if ((qStr = (char *)calloc(655500,sizeof(char))) == NULL) {
fprintf(stderr,"Failed allocating qStr\n");
return 0;
}
qStr[0] = '\0';
BaseAdaptor_generateSql(ba, constraint, countCols, qStr);
StatementHandle *sth = ba->prepare(ba,qStr,strlen(qStr));
sth->execute(sth);
if (sth->numRows(sth) != 1) {
fprintf(stderr, "genericCount didn't return a row - bye!\n");
return 0;
}
ResultRow *row = sth->fetchRow(sth);
int count = row->getLongAt(row, 0);
free(qStr);
return count;
}
Chromosome *ChromosomeAdaptor_fetchByDbID(ChromosomeAdaptor *ca, IDType dbID) {
Chromosome *chromosome;
char qStr[256];
StatementHandle *sth;
ResultRow *row;
if (IDHash_contains(ca->chrCache,dbID)) {
chromosome = IDHash_getValue(ca->chrCache, dbID);
} else {
sprintf(qStr,"SELECT chromosome_id, name, length"
" FROM chromosome"
" WHERE chromosome_id = "
IDFMTSTR, dbID);
sth = ca->prepare((BaseAdaptor *)ca,qStr,strlen(qStr));
sth->execute(sth);
row = sth->fetchRow(sth);
if( row == NULL ) {
sth->finish(sth);
return NULL;
}
chromosome = ChromosomeAdaptor_chromosomeFromRow(ca, row);
sth->finish(sth);
}
return chromosome;
}
DBEntry *DBEntryAdaptor_fetchByDbID(DBEntryAdaptor *dbea, IDType dbID) {
char qStr[512];
StatementHandle *sth;
ResultRow *row;
DBEntry *dbe = NULL;
sprintf(qStr,
"SELECT xref.xref_id, xref.dbprimary_acc, xref.display_label,"
" xref.version, xref.description,"
" exDB.db_name, exDB.db_release, es.synonym"
" FROM (xref, external_db exDB)"
" LEFT JOIN external_synonym es on es.xref_id = xref.xref_id"
" WHERE xref.xref_id = " IDFMTSTR
" AND xref.external_db_id = exDB.external_db_id",
dbID);
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
// Why???? my %duplicate;
while ((row = sth->fetchRow(sth))){
if (!row->col(row,0)) {
fprintf(stderr,"WARNING: Got xref with no refID\n");
return NULL;
}
if (!dbe) {
dbe = DBEntry_new();
DBEntry_setAdaptor(dbe,(BaseAdaptor *)dbea);
DBEntry_setDbID(dbe, dbID);
DBEntry_setPrimaryId(dbe, row->getStringAt(row,1));
DBEntry_setDisplayId(dbe, row->getStringAt(row,2));
DBEntry_setVersion(dbe, row->getStringAt(row,3));
DBEntry_setRelease(dbe, row->getStringAt(row,6));
DBEntry_setDbName(dbe, row->getStringAt(row,5));
if (row->col(row,4)) DBEntry_setDescription(dbe, row->getStringAt(row,4));
}
if (row->col(row,7)) DBEntry_addSynonym(dbe, row->getStringAt(row,7));
}
sth->finish(sth);
return dbe;
}
int DBEntryAdaptor_fetchAllByGene(DBEntryAdaptor *dbea, Gene *gene) {
char qStr[512];
StatementHandle *sth;
ResultRow *row;
sprintf(qStr,
"SELECT t.transcript_id, t.canonical_translation_id"
" FROM transcript t"
" WHERE t.gene_id = " IDFMTSTR,
Gene_getDbID(gene));
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
while ((row = sth->fetchRow(sth))) {
IDType transcriptId = row->getLongLongAt(row,0);
int i;
Vector *transLinks;
if (row->col(row,1)) {
IDType translationId = row->getLongLongAt(row,1);
Vector *translatLinks = DBEntryAdaptor_fetchByObjectType(dbea, translationId,"Translation");
for (i=0;i<Vector_getNumElement(translatLinks); i++) {
Gene_addDBLink(gene,Vector_getElementAt(translatLinks,i));
}
Vector_free(translatLinks);
}
transLinks = DBEntryAdaptor_fetchByObjectType(dbea, transcriptId,"Transcript");
for (i=0;i<Vector_getNumElement(transLinks); i++) {
Gene_addDBLink(gene, Vector_getElementAt(transLinks,i));
}
Vector_free(transLinks);
}
/* NIY This is wrong so I'm not going to implement it!
if($gene->stable_id){
my $genelinks = $self->_fetch_by_object_type( $gene->stable_id, 'Gene' );
foreach my $genelink ( @$genelinks ) {
$gene->add_DBLink( $genelink );
}
}
*/
return 1;
}
int DBEntryAdaptor_fetchAllByTranscript(DBEntryAdaptor *dbea, Transcript *trans) {
char qStr[512];
StatementHandle *sth;
ResultRow *row;
Vector *transLinks;
int i;
sprintf(qStr,
"SELECT t.canonical_translation_id"
" FROM transcript t"
" WHERE t.transcript_id = " IDFMTSTR,
Transcript_getDbID(trans));
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
//
// Did this to be consistent with fetch_by_Gene, but don't like
// it (filling in the object). I think returning the array would
// be better. Oh well. EB
//
while ((row = sth->fetchRow(sth))) {
IDType translationId = row->getLongLongAt(row,0);
Vector *translatLinks = DBEntryAdaptor_fetchByObjectType(dbea, translationId,"Translation");
for (i=0;i<Vector_getNumElement(translatLinks); i++) {
Transcript_addDBLink(trans,Vector_getElementAt(translatLinks,i));
}
Vector_free(translatLinks);
}
sth->finish(sth);
transLinks = DBEntryAdaptor_fetchByObjectType(dbea, Transcript_getDbID(trans),"Transcript");
fprintf(stderr,"transLinks\n");
for (i=0;i<Vector_getNumElement(transLinks); i++) {
Transcript_addDBLink(trans,Vector_getElementAt(transLinks,i));
}
Vector_free(transLinks);
return 1;
}
SyntenyRegion *SyntenyRegionAdaptor_fetchByDbID(SyntenyRegionAdaptor *sra, IDType dbID) {
StatementHandle *sth;
ResultRow *row;
char qStr[1024];
SyntenyRegion *sr = NULL;
int ok = 1;
if (!dbID) {
fprintf(stderr,"Error: SyntenyRegionAdaptor_fetchByDbID with no dbID!\n");
ok = 0;
}
if (ok) {
sprintf(qStr,
"select synteny_cluster_id,"
" dnafrag_id,"
" seq_start,"
" seq_end"
" from synteny_region "
" where synteny_region_id = " IDFMTSTR,
dbID);
sth = sra->prepare((BaseAdaptor *)sra, qStr, strlen(qStr));
sth->execute(sth);
if (!(row = sth->fetchRow(sth))) {
fprintf(stderr,"Error: No such dbID " IDFMTSTR "\n",dbID);
ok = 0;
}
}
if (ok) {
sr = SyntenyRegionAdaptor_newRegionFromArray(sra,
dbID,
row->getLongLongAt(row,0),
row->getLongLongAt(row,1),
row->getIntAt(row,2),
row->getIntAt(row,3));
sth->finish(sth);
}
return sr;
}
/*
=head2 store_transcript_linkage
Arg[1] : Bio::EnsEMBL::IntronSupportingEvidence Evidence to link
Arg[2] : Bio::EnsEMBL::Transcript Transcript to link
Arg[3] : Integer an optional ID to give if the Transcript's own ID is possibly incorrect
Example : $isea->store_transcript_linkage($ise, $transcript);
$isea->store_transcript_linkage($ise, $transcript, $tid);
Description : Links a Transcript to a portion of Intron evidence
Returntype : None
Exceptions : Thrown if the given object is not a Transcript, if the
transcript is not stored, if the supporting evidence is not
stored and for any DB exception.
=cut
*/
void IntronSupportingEvidenceAdaptor_storeTranscriptLinkage(IntronSupportingEvidenceAdaptor *isea,
IntronSupportingEvidence *sf,
Transcript *transcript,
IDType transcriptId) {
if (sf == NULL || transcript == NULL) {
fprintf(stderr,"sf or transcript is NULL in IntronSupportingEvidenceAdaptor_storeTranscriptLinkage\n");
exit(1);
}
Class_assertType(CLASS_INTRONSUPPORTINGEVIDENCE, sf->objectType);
Class_assertType(CLASS_TRANSCRIPT, transcript->objectType);
if (! IntronSupportingEvidence_isStored(sf, isea->dba)) {
fprintf(stderr,"Cannot perform the link. The IntronSupportingEvidence must be persisted first\n");
exit(1);
}
// Moved up so can use in sprintf
Intron *intron = IntronSupportingEvidence_getIntron(sf, transcript);
Exon *prevExon = Intron_getPrevExon(intron);
Exon *nextExon = Intron_getNextExon(intron);
if (!transcriptId) {
transcriptId = Transcript_getDbID(transcript);
}
char qStr[1024];
sprintf(qStr, "INSERT IGNORE INTO transcript_intron_supporting_evidence "
"(transcript_id, intron_supporting_evidence_id, previous_exon_id, next_exon_id) "
"VALUES ("IDFMTSTR","IDFMTSTR","IDFMTSTR","IDFMTSTR")",
transcriptId, IntronSupportingEvidence_getDbID(sf), Exon_getDbID(prevExon), Exon_getDbID(nextExon));
StatementHandle *sth = isea->prepare((BaseAdaptor *)isea,qStr,strlen(qStr));
sth->execute(sth);
sth->finish(sth);
Intron_free(intron);
return;
}
IDType HomologyAdaptor_getRelationship(HomologyAdaptor *ha, char *qStr) {
StatementHandle *sth;
ResultRow *row;
IDType relId;
sth = ha->prepare((BaseAdaptor *)ha, qStr, strlen(qStr));
sth->execute(sth);
if ((row = sth->fetchRow(sth))) {
relId = row->getLongLongAt(row,0);
} else {
fprintf(stderr, "Error: No relationship found by %s\n",qStr);
relId = 0;
}
sth->finish(sth);
return relId;
}
/*
=head2 fetch_all_by_Transcript
Arg[1] : Bio::EnsEMBL::Transcript Transcript to search with
Example : my $ises = $isea->fetch_all_by_Transcript($transcript);
Description : Uses the given Transcript to search for all instances of
IntronSupportingEvidence linked to the transcript in the
database
Returntype : ArrayRef of IntronSupportingEvidence objects
Exceptions : Thrown if arguments are not as stated and for DB errors
=cut
*/
Vector *IntronSupportingEvidenceAdaptor_fetchAllByTranscript(IntronSupportingEvidenceAdaptor *isea, Transcript *transcript) {
char qStr[1024];
sprintf(qStr,"SELECT intron_supporting_evidence_id "
"FROM transcript_intron_supporting_evidence "
"WHERE transcript_id = "IDFMTSTR, Transcript_getDbID(transcript));
StatementHandle *sth = isea->prepare((BaseAdaptor *)isea,qStr,strlen(qStr));
sth->execute(sth);
Vector *idVec = Vector_new();
ResultRow *row;
while ((row = sth->fetchRow(sth))) {
IDType id = row->getLongLongAt(row, 0);
IDType *idP;
if ((idP = calloc(1,sizeof(IDType))) == NULL) {
fprintf(stderr, "Failed allocating space for a id\n");
exit(1);
}
*idP = id;
Vector_addElement(idVec, idP);
}
sth->finish(sth);
Vector *out;
if (Vector_getNumElement(idVec) > 0) {
out = IntronSupportingEvidenceAdaptor_fetchAllByDbIDList(isea, idVec, NULL);
} else {
out = Vector_new();
}
// Free ids vector
Vector_setFreeFunc(idVec, free);
Vector_free(idVec);
return out;
}
IDType DBEntryAdaptor_exists(DBEntryAdaptor *dbea, DBEntry *dbe) {
char qStr[512];
StatementHandle *sth;
ResultRow *row;
IDType dbID;
if (!dbe || !Class_isDescendent(CLASS_DBENTRY, dbe->objectType)) {
fprintf(stderr,"Error: arg must be a DBEntry\n");
exit(1);
}
// NIY Was dbe->external_db instead of dbe->dbname - are they different?
// NIY mysql_quote strings
sprintf(qStr,
"SELECT x.xref_id "
" FROM xref x, external_db xdb"
" WHERE x.external_db_id = xdb.external_db_id"
" AND x.display_label = '%s' "
" AND xdb.db_name = '%s'",
DBEntry_getDisplayId(dbe),
DBEntry_getDbName(dbe));
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
row = sth->fetchRow(sth);
if( row == NULL ) {
sth->finish(sth);
return 0;
}
dbID = row->getLongLongAt(row,0);
sth->finish(sth);
return dbID;
}
/*
=head2 generic_fetch
Arg [1] : (optional) string $constraint
An SQL query constraint (i.e. part of the WHERE clause)
Arg [2] : (optional) Bio::EnsEMBL::AssemblyMapper $mapper
A mapper object used to remap features
as they are retrieved from the database
Arg [3] : (optional) Bio::EnsEMBL::Slice $slice
A slice that features should be remapped to
Example : $fts = $a->generic_fetch('contig_id in (1234, 1235)');
Description: Performs a database fetch and returns feature objects in
contig coordinates.
Returntype : listref of Bio::EnsEMBL::SeqFeature in contig coordinates
Exceptions : Thrown if there is an issue with querying the data
Caller : BaseFeatureAdaptor, ProxyDnaAlignFeatureAdaptor::generic_fetch
Status : Stable
=cut
*/
Vector *BaseAdaptor_genericFetch(BaseAdaptor *ba, char *constraint, AssemblyMapper *mapper, Slice *slice) {
Vector *res = NULL;
char *qStr = NULL;
if ((qStr = (char *)calloc(655500,sizeof(char))) == NULL) {
fprintf(stderr,"Failed allocating qStr\n");
return res;
}
qStr[0] = '\0';
BaseAdaptor_generateSql(ba, constraint, NULL, qStr);
StatementHandle *sth = ba->prepare((BaseAdaptor *)ba,qStr,strlen(qStr));
sth->execute(sth);
res = ba->objectsFromStatementHandle(ba, sth, mapper, slice);
sth->finish(sth);
free(qStr);
return res;
}
// Switch to passing two element array to fill in as an argument, so don't have to allocate it
IDType *IntronSupportingEvidenceAdaptor_fetchFlankingExonIds(IntronSupportingEvidenceAdaptor *isea, IntronSupportingEvidence *ise, Transcript *transcript, IDType *flanks) {
char qStr[1024];
sprintf(qStr,"SELECT previous_exon_id, next_exon_id "
"FROM transcript_intron_supporting_evidence "
"WHERE transcript_id ="IDFMTSTR" and intron_supporting_evidence_id ="IDFMTSTR,
Transcript_getDbID(transcript), IntronSupportingEvidence_getDbID(ise));
StatementHandle *sth = isea->prepare((BaseAdaptor *)isea,qStr,strlen(qStr));
sth->execute(sth);
if (sth->numRows(sth) == 0) {
return NULL;
}
ResultRow *row = sth->fetchRow(sth);
flanks[0] = row->getLongLongAt(row, 0);
flanks[1] = row->getLongLongAt(row, 1);
sth->finish(sth);
return flanks;
}
int HomologyAdaptor_getRelationships(HomologyAdaptor *ha, char *qStr, IDType **idsP) {
StatementHandle *sth;
ResultRow *row;
int nAlloced = 2;
int nId = 0;
int ok = 1;
sth = ha->prepare((BaseAdaptor *)ha, qStr, strlen(qStr));
sth->execute(sth);
if ((*idsP = (IDType *)calloc(nAlloced,sizeof(IDType))) == NULL) {
fprintf(stderr,"Error: Failed allocating idsP\n");
ok = 0;
}
if (ok) {
while ((row = sth->fetchRow(sth))) {
if (nId == nAlloced) {
nAlloced = (nAlloced *2);
if ((*idsP = (IDType *)realloc(*idsP,nAlloced*sizeof(IDType))) == NULL) {
fprintf(stderr,"Error: Failed reallocating idsP\n");
ok = 0;
break;
}
}
(*idsP)[nId++] = row->getLongLongAt(row,0);
}
}
if (ok) {
sth->finish(sth);
}
return nId;
}
Vector *HomologyAdaptor_getHomologues(HomologyAdaptor *ha, char *qStr) {
StatementHandle *sth;
ResultRow *row;
Vector *genes;
sth = ha->prepare((BaseAdaptor *)ha, qStr, strlen(qStr));
sth->execute(sth);
genes = Vector_new();
while ((row = sth->fetchRow(sth))) {
Homology *homol = Homology_new();
Homology_setSpecies(homol, row->getStringAt(row,1));
Homology_setStableId(homol, row->getStringAt(row,0));
Homology_setChromosome(homol, row->getStringAt(row,2));
Homology_setChrStart(homol, row->getIntAt(row,3));
Homology_setChrEnd(homol, row->getIntAt(row,4));
Vector_addElement(genes,homol);
}
sth->finish(sth);
return genes;
}
Vector *SyntenyRegionAdaptor_fetchByClusterId(SyntenyRegionAdaptor *sra, IDType clusterId) {
char qStr[256];
StatementHandle *sth;
ResultRow *row;
Vector *out = NULL;
if (!clusterId) {
fprintf(stderr, "Error: fetch_by_cluster_id with no cluster_id!\n");
} else {
sprintf(qStr,
"select synteny_region_id,"
" dnafrag_id,"
" seq_start,"
" seq_end "
" from synteny_region "
" where synteny_cluster_id = " IDFMTSTR, clusterId);
sth = sra->prepare((BaseAdaptor *)sra, qStr, strlen(qStr));
sth->execute(sth);
out = Vector_new();
while ((row = sth->fetchRow(sth))) {
SyntenyRegion *sr = SyntenyRegionAdaptor_newRegionFromArray(sra,
row->getLongLongAt(row,0),
clusterId,
row->getLongLongAt(row,1),
row->getIntAt(row,2),
row->getIntAt(row,3));
Vector_addElement(out,sr);
}
sth->finish(sth);
}
return out;
}
Vector *AttributeAdaptor_doFetchAllByTypeAndTableAndID(AttributeAdaptor *ata, char *type, char *table, IDType objectId, char *code) {
char qStr[1024];
sprintf(qStr, "SELECT at.code, at.name, at.description, t.value "
"FROM %s_attrib t, attrib_type at "
"WHERE at.attrib_type_id = t.attrib_type_id", table);
if (code != NULL){
sprintf(qStr,"%s AND at.code like '%s'", qStr, code);
}
// if(defined($object_id)){
sprintf(qStr,"%s AND t.%s_id = "IDFMTSTR, qStr, type, objectId);
// }
StatementHandle *sth = ata->prepare((BaseAdaptor *)ata,qStr,strlen(qStr));
sth->execute(sth);
Vector *results = AttributeAdaptor_objectsFromStatementHandle(ata, sth);
sth->finish(sth);
return results;
}
IDType DBEntryAdaptor_store(DBEntryAdaptor *dbea, DBEntry *exObj,
IDType ensObject, char *ensType, int ignoreRelease) {
fprintf(stderr,"DBEntryAdaptor_store does not implement ignoreRelease functionality yet\n");
char qStr[512];
StatementHandle *sth;
ResultRow *row;
IDType dbRef;
IDType dbX;
//
// Check for the existance of the external_db, throw if it does not exist
//
sprintf(qStr,
"SELECT external_db_id"
" FROM external_db"
" WHERE db_name = '%s'"
" AND db_release = %s",
DBEntry_getDbName(exObj),
DBEntry_getRelease(exObj));
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
row = sth->fetchRow(sth);
if( row == NULL ) {
sth->finish(sth);
fprintf(stderr,"Error: external_db [%s] release [%s] does not exist\n",
DBEntry_getDbName(exObj), DBEntry_getRelease(exObj));
exit(1);
}
dbRef = row->getLongLongAt(row,0);
sth->finish(sth);
//
// Check for the existance of the external reference, add it if not present
//
sprintf(qStr,
"SELECT xref_id"
" FROM xref"
" WHERE external_db_id = " IDFMTSTR
" AND dbprimary_acc = '%s'"
" AND version = %s",
dbRef,
DBEntry_getPrimaryId(exObj),
DBEntry_getVersion(exObj));
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
row = sth->fetchRow(sth);
if (row != NULL) {
dbX = row->getLongLongAt(row,0);
sth->finish(sth);
} else {
//
// store the new xref
//
// First finish the old sth
sth->finish(sth);
// NIY Handling NULL values
sprintf(qStr,
"INSERT ignore INTO xref"
" SET dbprimary_acc = '%s',"
" display_label = '%s',"
" version = %s,"
" description = '%s',"
" external_db_id = " IDFMTSTR,
DBEntry_getPrimaryId(exObj),
DBEntry_getDisplayId(exObj),
DBEntry_getVersion(exObj),
DBEntry_getDescription(exObj),
dbRef
);
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
dbX = sth->getInsertId(sth);
sth->finish(sth);
//
// store the synonyms for the new xref
//
if (DBEntry_getAllSynonyms(exObj)) {
StatementHandle *checkSth;
StatementHandle *storeSth;
int i;
Vector *synonyms;
sprintf(qStr,
"SELECT xref_id, synonym"
" FROM external_synonym"
" WHERE xref_id = %" IDFMTSTR
" AND synonym = '%%s'");
checkSth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sprintf(qStr,
"INSERT ignore INTO external_synonym"
" SET xref_id = %" IDFMTSTR ", synonym = '%%s'");
storeSth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
synonyms = DBEntry_getAllSynonyms(exObj);
for (i=0;i<Vector_getNumElement(synonyms); i++) {
char *syn = Vector_getElementAt(synonyms,i);
checkSth->execute(checkSth, dbX, syn);
row = checkSth->fetchRow(checkSth);
if (!row) {
storeSth->execute(storeSth, dbX, syn);
}
}
checkSth->finish(checkSth);
storeSth->finish(storeSth);
}
}
//
// check if the object mapping was already stored
//
sprintf(qStr,
"SELECT xref_id"
" FROM object_xref"
" WHERE xref_id = " IDFMTSTR
" AND ensembl_object_type = '%s'"
" AND ensembl_id = " IDFMTSTR,
dbX, ensType, ensObject);
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
row = sth->fetchRow(sth);
// NOTE row will be invalid after this call but will still
// indicate whether something was found
sth->finish(sth);
if (!row) {
IDType Xidt;
//
// Store the reference to the internal ensembl object
//
sprintf(qStr,
"INSERT ignore INTO object_xref"
" SET xref_id = " IDFMTSTR ","
" ensembl_object_type = '%s',"
" ensembl_id = " IDFMTSTR,
dbX, ensType, ensObject);
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
DBEntry_setDbID(exObj, dbX);
DBEntry_setAdaptor(exObj, (BaseAdaptor *)dbea);
Xidt = sth->getInsertId(sth);
//
// If this is an IdentityXref need to store in that table too
//
if (DBEntry_getIdentityXref(exObj)) {
IdentityXref *idx = DBEntry_getIdentityXref(exObj);
sprintf(qStr,
"INSERT ignore INTO identity_xref"
" SET object_xref_id = " IDFMTSTR ","
" query_identity = %f,"
" target_identity = %f",
Xidt,
IdentityXref_getQueryIdentity(idx),
IdentityXref_getTargetIdentity(idx));
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
sth->finish(sth);
}
}
return dbX;
}
// This method was a mess - tidied and rearranged
IDType IntronSupportingEvidenceAdaptor_store(IntronSupportingEvidenceAdaptor *isea, IntronSupportingEvidence *sf) {
if (sf == NULL) {
fprintf(stderr,"sf is NULL in IntronSupportingEvidenceAdaptor_store\n");
exit(1);
}
Class_assertType(CLASS_INTRONSUPPORTINGEVIDENCE, sf->objectType);
DBAdaptor *db = isea->dba;
AnalysisAdaptor *analysisAdaptor = DBAdaptor_getAnalysisAdaptor(db);
if (IntronSupportingEvidence_isStored(sf, db)) {
fprintf(stderr,"ISE already stored\n");
return IntronSupportingEvidence_getDbID(sf);
}
Analysis *analysis = IntronSupportingEvidence_getAnalysis(sf);
if (!Analysis_isStored(analysis, db)) {
AnalysisAdaptor_store(analysisAdaptor, analysis);
}
IDType analysisId = Analysis_getDbID(analysis);
// Think the above is equivalent to this horror
//my $analysis_id = $analysis->is_stored($db) ? $analysis->dbID() : $db->get_AnalysisAdaptor()->store($analysis);
/* No transfer (see GeneAdaptor for why)
my $seq_region_id;
($sf, $seq_region_id) = $self->_pre_store($sf);
*/
IDType seqRegionId = BaseFeatureAdaptor_preStore((BaseFeatureAdaptor *)isea, (SeqFeature*)sf);
char qStr[1024];
sprintf(qStr, "INSERT IGNORE INTO intron_supporting_evidence "
"(analysis_id, seq_region_id, seq_region_start, seq_region_end, seq_region_strand, hit_name, score, score_type, is_splice_canonical) "
"VALUES ("IDFMTSTR","IDFMTSTR",%ld,%ld,%d,'%s',%f,'%s',%d)",
analysisId,
seqRegionId,
IntronSupportingEvidence_getSeqRegionStart((SeqFeature*)sf),
IntronSupportingEvidence_getSeqRegionEnd((SeqFeature*)sf),
IntronSupportingEvidence_getSeqRegionStrand((SeqFeature*)sf),
IntronSupportingEvidence_getHitName(sf),
IntronSupportingEvidence_getScore(sf),
IntronSupportingEvidence_getScoreType(sf),
IntronSupportingEvidence_getIsSpliceCanonical(sf));
StatementHandle *sth = isea->prepare((BaseAdaptor *)isea,qStr,strlen(qStr));
sth->execute(sth);
IDType sfId = sth->getInsertId(sth);
sth->finish(sth);
if (!sfId) {
sprintf(qStr,"SELECT intron_supporting_evidence_id "
"FROM intron_supporting_evidence "
"WHERE analysis_id = "IDFMTSTR
" AND seq_region_id = "IDFMTSTR
" AND seq_region_start = %ld"
" AND seq_region_end = %ld"
" AND seq_region_strand = %d"
" AND hit_name = '%s'",
analysisId,
seqRegionId,
IntronSupportingEvidence_getSeqRegionStart((SeqFeature*)sf),
IntronSupportingEvidence_getSeqRegionEnd((SeqFeature*)sf),
IntronSupportingEvidence_getSeqRegionStrand((SeqFeature*)sf),
IntronSupportingEvidence_getHitName(sf));
sth = isea->prepare((BaseAdaptor *)isea,qStr,strlen(qStr));
sth->execute(sth);
if (sth->numRows(sth) > 0) {
ResultRow *row = sth->fetchRow(sth);
sfId = row->getLongLongAt(row, 0);
}
sth->finish(sth);
}
IntronSupportingEvidence_setAdaptor((SeqFeature*)sf, (BaseAdaptor*)isea);
IntronSupportingEvidence_setDbID(sf, sfId);
return IntronSupportingEvidence_getDbID(sf);
}
// For ordered, the default should be 0 (if you just need to fill out the args)
// Note ONLY stable_id can be char, all other pk's must be IDType (see code)
Vector *BaseAdaptor_listDbIDs(BaseAdaptor *ba, char *table, char *pk, int ordered) {
int ok = 1;
char colName[1024];
if (pk == NULL) {
sprintf(colName, "%s_id", table);
} else {
strcpy(colName, pk);
}
char qStr[1024];
sprintf(qStr,"SELECT `%s` FROM `%s`", colName, table );
if ( BaseAdaptor_isMultiSpecies(BaseAdaptor *ba)
// For now just the multi species because I don't have adaptors in the Class hierarchy
// && $self->isa('Bio::EnsEMBL::DBSQL::BaseFeatureAdaptor')
// && !$self->isa('Bio::EnsEMBL::DBSQL::UnmappedObjectAdaptor')
) {
char tmpStr[1024];
sprintf(tmpStr, "JOIN seq_region USING (seq_region_id) "
"JOIN coord_system cs USING (coord_system_id) "
"WHERE cs.species_id = "IDFMTSTR, BaseAdaptor_getSpeciesId(ba));
sprintf(qStr, "%s %s", qStr, tmpStr);
}
if (ordered) {
sprintf(qStr, "%s ORDER BY seq_region_id, seq_region_start", qStr);
}
StatementHandle *sth = ba->prepare(ba,qStr,strlen(qStr));
sth->execute(sth);
Vector *out = Vector_new();
if (strcmp(pk, "stable_id")) {
ResultRow *row;
while ((row = sth->fetchRow(sth))) {
char *stableId = row->getStringCopyAt(row, 0);
Vector_addElement(out, stableId);
}
} else {
IDType *idP;
ResultRow *row;
while ((row = sth->fetchRow(sth))) {
IDType id = row->getLongLongAt(row, 0);
if ((idP = calloc(1,sizeof(IDType))) == NULL) {
fprintf(stderr, "Failed allocating space for a id\n");
ok = 0;
} else {
*idP = id;
Vector_addElement(out, idP);
}
}
}
if (!ok) {
Vector_free(out);
out = NULL;
}
return out;
}
Vector *GenomicAlignAdaptor_fetchAllByDNAFragGenomeDBDirect( GenomicAlignAdaptor *gaa,
DNAFrag *dnaFrag, GenomeDB *targetGenome, int *startP, int *endP, IDType methodLinkId) {
IDType dnaFragId;
GenomeDB *genomeDB;
char *qStr = NULL;
char tmpStr[512];
Vector *results;
StatementHandle *sth;
int ok = 0;
if (!dnaFrag) {
fprintf(stderr, "Error: Input dnafrag must not be NULL\n");
ok = 0;
}
if (ok) {
// formatting the dnafrag
dnaFragId = DNAFrag_getDbID(dnaFrag);
genomeDB = DNAFrag_getGenomeDB(dnaFrag);
StrUtil_copyString(&qStr,
"SELECT gab.consensus_dnafrag_id,"
" gab.consensus_start,"
" gab.consensus_end,"
" gab.query_dnafrag_id,"
" gab.query_start,"
" gab.query_end,"
" gab.query_strand,"
" gab.method_link_id,"
" gab.score,"
" gab.perc_id,"
" gab.cigar_line"
" FROM genomic_align_block gab ",0);
if (targetGenome) {
qStr = StrUtil_appendString(qStr,", dnafrag d");
}
sprintf(tmpStr," WHERE gab.method_link_id = " IDFMTSTR, methodLinkId);
qStr = StrUtil_appendString(qStr,tmpStr);
results = Vector_new();
if (!targetGenome ||
GenomeDB_hasQuery(genomeDB, targetGenome, methodLinkId)) {
Vector *qres;
sprintf(tmpStr," AND gab.consensus_dnafrag_id = " IDFMTSTR, dnaFragId);
qStr = StrUtil_appendString(qStr, tmpStr);
if (startP && endP) {
int lowerBound = *startP - gaa->maxAlignmentLength;
sprintf(tmpStr,
" AND gab.consensus_start <= %d"
" AND gab.consensus_start >= %d"
" AND gab.consensus_end >= %d", *endP, lowerBound, *startP ) ;
qStr = StrUtil_appendString(qStr, tmpStr);
}
if (targetGenome) {
sprintf(tmpStr,
" AND gab.query_dnafrag_id = d.dnafrag_id"
" AND d.genome_db_id = " IDFMTSTR, GenomeDB_getDbID(targetGenome));
qStr = StrUtil_appendString(qStr, tmpStr);
}
sth = gaa->prepare((BaseAdaptor *)gaa, qStr, strlen(qStr));
sth->execute(sth);
qres = GenomicAlignAdaptor_objectsFromStatementHandle(gaa, sth, 0);
Vector_append(results,qres);
Vector_free(qres);
sth->finish(sth);
}
if (!targetGenome ||
GenomeDB_hasConsensus(genomeDB, targetGenome, methodLinkId)) {
Vector *cres;
sprintf(tmpStr," AND gab.query_dnafrag_id = " IDFMTSTR, dnaFragId);
qStr = StrUtil_appendString(qStr, tmpStr);
if (startP && endP) {
int lowerBound = *startP - gaa->maxAlignmentLength;
sprintf(tmpStr,
" AND gab.query_start <= %d"
" AND gab.query_start >= %d"
" AND gab.query_end >= %d", *endP, lowerBound, *startP ) ;
qStr = StrUtil_appendString(qStr, tmpStr);
}
if (targetGenome) {
sprintf(tmpStr,
" AND gab.consensus_dnafrag_id = d.dnafrag_id"
" AND d.genome_db_id = " IDFMTSTR, GenomeDB_getDbID(targetGenome));
qStr = StrUtil_appendString(qStr, tmpStr);
}
sth = gaa->prepare((BaseAdaptor *)gaa, qStr, strlen(qStr));
sth->execute(sth);
cres = GenomicAlignAdaptor_objectsFromStatementHandle(gaa, sth, 1);
Vector_append(results,cres);
Vector_free(cres);
sth->finish(sth);
}
}
if (qStr)
free(qStr);
return results;
}
void GenomicAlignAdaptor_store(GenomicAlignAdaptor *gaa, Vector *genomicAligns) {
int ok = 1;
char *qStr = NULL;
StatementHandle *sth;
char commaStr[2] = {'\0','\0'};
int i;
char *tmpStr = NULL;
if ((tmpStr = (char *)calloc(65556,sizeof(char))) == NULL) {
fprintf(stderr,"Failed allocating tmpStr\n");
ok = 0;
}
if (ok) {
StrUtil_copyString(&qStr, "INSERT INTO genomic_align_block"
" (consensus_dnafrag_id, consensus_start, consensus_end,"
" query_dnafrag_id, query_start, query_end, query_strand, method_link_id,"
" score, perc_id, cigar_line) VALUES ",0);
for (i=0; i<Vector_getNumElement(genomicAligns); i++) {
GenomicAlign *ga = Vector_getElementAt(genomicAligns,i);
DNAFrag *consDNAFrag = GenomicAlign_getConsensusDNAFrag(ga);
DNAFrag *queryDNAFrag = GenomicAlign_getQueryDNAFrag(ga);
// check that everything has dbIDs
if (!DNAFrag_getDbID(consDNAFrag) || !DNAFrag_getDbID(queryDNAFrag)) {
fprintf(stderr, "Error: dna_fragment in GenomicAlign is not in DB\n");
ok = 0;
break;
}
}
}
GenomicAlign *ga = NULL;
DNAFrag *consDNAFrag = NULL;
DNAFrag *queryDNAFrag = NULL;
IDType methodLinkId = 0;
if (ok) {
// all clear for storing
for (i=0; i<Vector_getNumElement(genomicAligns); i++) {
ga = Vector_getElementAt(genomicAligns,i);
consDNAFrag = GenomicAlign_getConsensusDNAFrag(ga);
queryDNAFrag = GenomicAlign_getQueryDNAFrag(ga);
methodLinkId = GenomicAlignAdaptor_methodLinkIdByAlignmentType(gaa, GenomicAlign_getAlignmentType(ga));
if (!methodLinkId) {
fprintf(stderr, "Error: There is no method_link with this type [%s] in the DB.\n",
GenomicAlign_getAlignmentType(ga));
ok = 0;
break;
}
}
if (ok) {
sprintf(tmpStr," %s(" IDFMTSTR ", %d, %d, " IDFMTSTR ", %d, %d, %d, " IDFMTSTR ", %f, %f, '%s')",
commaStr,
DNAFrag_getDbID(consDNAFrag),
GenomicAlign_getConsensusStart(ga),
GenomicAlign_getConsensusEnd(ga),
DNAFrag_getDbID(queryDNAFrag),
GenomicAlign_getQueryStart(ga),
GenomicAlign_getQueryEnd(ga),
GenomicAlign_getQueryStrand(ga),
methodLinkId,
GenomicAlign_getScore(ga),
GenomicAlign_getPercentId(ga),
GenomicAlign_getCigarString(ga));
qStr = StrUtil_appendString(qStr, tmpStr);
commaStr[0] = ',';
}
sth = gaa->prepare((BaseAdaptor *)gaa, qStr, strlen(qStr));
sth->execute(sth);
sth->finish(sth);
}
if (qStr)
free(qStr);
if (tmpStr)
free(tmpStr);
}
Vector *DBEntryAdaptor_fetchByObjectType(DBEntryAdaptor *dbea, IDType ensObj, char *ensType) {
Vector *out;
char qStr[1024];
StatementHandle *sth;
ResultRow *row;
IDHash *seen;
if (!ensObj) {
fprintf(stderr,"Error: Can't fetchByObjectType without an object\n");
exit(1);
}
if (!ensType) {
fprintf(stderr,"Error: Can't fetchByObjectType without a type\n");
exit(1);
}
// Not sure if idt identities are right way round
sprintf(qStr,
"SELECT xref.xref_id, xref.dbprimary_acc, xref.display_label, xref.version,"
" xref.description,"
" exDB.db_name, exDB.db_release, exDB.status,"
" oxr.object_xref_id,"
" es.synonym,"
" idt.xref_identity, idt.ensembl_identity"
" FROM (external_db exDB, object_xref oxr, xref xref)"
" LEFT JOIN external_synonym es on es.xref_id = xref.xref_id"
" LEFT JOIN identity_xref idt on idt.object_xref_id = oxr.object_xref_id"
" WHERE xref.xref_id = oxr.xref_id"
" AND xref.external_db_id = exDB.external_db_id"
" AND oxr.ensembl_id = " IDFMTSTR
" AND oxr.ensembl_object_type = '%s'",
ensObj,
ensType);
sth = dbea->prepare((BaseAdaptor *)dbea,qStr,strlen(qStr));
sth->execute(sth);
seen = IDHash_new(IDHASH_SMALL);
out = Vector_new();
while ((row = sth->fetchRow(sth))) {
DBEntry *exDB;
IDType refID = row->getLongLongAt(row,0);
// using an outer join on the synonyms as well as on identity_xref, we
// now have to filter out the duplicates (see v.1.18 for
// original). Since there is at most one identity_xref row per xref,
// this is easy enough; all the 'extra' bits are synonyms
if (!IDHash_contains(seen,refID)) {
exDB = DBEntry_new();
DBEntry_setAdaptor(exDB,(BaseAdaptor *)dbea);
DBEntry_setDbID(exDB, refID);
DBEntry_setPrimaryId(exDB, row->getStringAt(row,1));
DBEntry_setDisplayId(exDB, row->getStringAt(row,2));
DBEntry_setVersion(exDB, row->getStringAt(row,3));
DBEntry_setDbName(exDB, row->getStringAt(row,5));
DBEntry_setRelease(exDB, row->getStringAt(row,6));
if (row->col(row,10)) {
IdentityXref *idx = IdentityXref_new();
DBEntry_setIdentityXref(exDB,idx);
IdentityXref_setQueryIdentity(idx, row->getDoubleAt(row,10));
IdentityXref_setTargetIdentity(idx, row->getDoubleAt(row,11));
}
if (row->col(row,4)) DBEntry_setDescription(exDB, row->getStringAt(row,4));
if (row->col(row,7)) DBEntry_setStatus(exDB, row->getStringAt(row,7));
Vector_addElement(out, exDB);
IDHash_add(seen, refID, exDB);
}
exDB = IDHash_getValue(seen, refID);
if (row->col(row,9)) {
DBEntry_addSynonym(exDB,row->getStringAt(row,9));
}
}
IDHash_free(seen, NULL);
sth->finish(sth);
return out;
}
Vector *PredictionTranscriptAdaptor_fetchAllBySlice(PredictionTranscriptAdaptor *pta, Slice *slice, char *logicName, int loadExons) {
//my $transcripts = $self->SUPER::fetch_all_by_Slice($slice,$logic_name);
Vector *transcripts = BaseFeatureAdaptor_fetchAllBySlice((BaseFeatureAdaptor *)pta, slice, logicName);
// if there are 0 or 1 transcripts still do lazy-loading
if ( ! loadExons || Vector_getNumElement(transcripts) < 2 ) {
return transcripts;
}
// preload all of the exons now, instead of lazy loading later
// faster than 1 query per transcript
// get extent of region spanned by transcripts
long minStart = 2000000000;
long maxEnd = -2000000000;
int i;
for (i=0; i<Vector_getNumElement(transcripts); i++) {
PredictionTranscript *t = Vector_getElementAt(transcripts, i);
if (PredictionTranscript_getSeqRegionStart((SeqFeature*)t) < minStart) {
minStart = PredictionTranscript_getSeqRegionStart((SeqFeature*)t);
}
if (PredictionTranscript_getSeqRegionEnd((SeqFeature*)t) > maxEnd) {
maxEnd = PredictionTranscript_getSeqRegionEnd((SeqFeature*)t);
}
}
Slice *extSlice;
if (minStart >= Slice_getStart(slice) && maxEnd <= Slice_getEnd(slice)) {
extSlice = slice;
} else {
SliceAdaptor *sa = DBAdaptor_getSliceAdaptor(pta->dba);
extSlice = SliceAdaptor_fetchByRegion(sa, Slice_getCoordSystemName(slice), Slice_getSeqRegionName(slice),
minStart, maxEnd, Slice_getStrand(slice), CoordSystem_getVersion(Slice_getCoordSystem(slice)), 0);
}
// associate exon identifiers with transcripts
IDHash *trHash = IDHash_new(IDHASH_MEDIUM);
for (i=0; i<Vector_getNumElement(transcripts); i++) {
PredictionTranscript *t = Vector_getElementAt(transcripts, i);
if ( ! IDHash_contains(trHash, PredictionTranscript_getDbID(t))) {
IDHash_add(trHash, PredictionTranscript_getDbID(t), t);
}
}
IDType *uniqueIds = IDHash_getKeys(trHash);
char tmpStr[1024];
char *qStr = NULL;
if ((qStr = (char *)calloc(655500,sizeof(char))) == NULL) {
fprintf(stderr,"Failed allocating qStr\n");
return transcripts;
}
int lenNum;
int endPoint = sprintf(qStr, "SELECT prediction_transcript_id, prediction_exon_id, exon_rank FROM prediction_exon WHERE prediction_transcript_id IN (");
for (i=0; i<IDHash_getNumValues(trHash); i++) {
if (i!=0) {
qStr[endPoint++] = ',';
qStr[endPoint++] = ' ';
}
lenNum = sprintf(tmpStr,IDFMTSTR,uniqueIds[i]);
memcpy(&(qStr[endPoint]), tmpStr, lenNum);
endPoint+=lenNum;
}
qStr[endPoint++] = ')';
qStr[endPoint] = '\0';
free(uniqueIds);
StatementHandle *sth = pta->prepare((BaseAdaptor *)pta,qStr,strlen(qStr));
sth->execute(sth);
IDHash *exTrHash = IDHash_new(IDHASH_MEDIUM);
ResultRow *row;
while ((row = sth->fetchRow(sth))) {
IDType trId = row->getLongLongAt(row,0);
IDType exId = row->getLongLongAt(row,1);
int rank = row->getIntAt(row,2);
if (! IDHash_contains(exTrHash, exId)) {
Vector *vec = Vector_new();
Vector_setFreeFunc(vec, PredictionTranscriptRankPair_free);
IDHash_add(exTrHash, exId, vec);
}
Vector *exVec = IDHash_getValue(exTrHash, exId);
PredictionTranscriptRankPair *trp = PredictionTranscriptRankPair_new(IDHash_getValue(trHash, trId), rank);
Vector_addElement(exVec, trp);
}
IDHash_free(trHash, NULL);
sth->finish(sth);
PredictionExonAdaptor *pea = DBAdaptor_getPredictionExonAdaptor(pta->dba);
Vector *exons = PredictionExonAdaptor_fetchAllBySlice(pea, extSlice);
// move exons onto transcript slice, and add them to transcripts
for (i=0; i<Vector_getNumElement(exons); i++) {
PredictionExon *ex = Vector_getElementAt(exons, i);
// Perl didn't have this line - it was in GeneAdaptor version so I think I'm going to keep it
if (!IDHash_contains(exTrHash, PredictionExon_getDbID(ex))) continue;
PredictionExon *newEx;
if (slice != extSlice) {
newEx = (PredictionExon*)PredictionExon_transfer((SeqFeature*)ex, slice);
if (newEx == NULL) {
fprintf(stderr, "Unexpected. Exon could not be transferred onto PredictionTranscript slice.\n");
exit(1);
}
} else {
newEx = ex;
}
Vector *exVec = IDHash_getValue(exTrHash, PredictionExon_getDbID(newEx));
int j;
for (j=0; j<Vector_getNumElement(exVec); j++) {
PredictionTranscriptRankPair *trp = Vector_getElementAt(exVec, j);
PredictionTranscript_addExon(trp->transcript, newEx, &trp->rank);
}
}
IDHash_free(exTrHash, Vector_free);
free(qStr);
return transcripts;
}
