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 *IntronSupportingEvidenceAdaptor_objectsFromStatementHandle(IntronSupportingEvidenceAdaptor *isea,
StatementHandle *sth,
AssemblyMapper *assMapper,
Slice *destSlice) {
SliceAdaptor *sa = DBAdaptor_getSliceAdaptor(isea->dba);
AnalysisAdaptor *aa = DBAdaptor_getAnalysisAdaptor(isea->dba);
Vector *features = Vector_new();
IDHash *sliceHash = IDHash_new(IDHASH_SMALL);
/* Unneccesary
my %analysis_hash;
my %sr_name_hash;
my %sr_cs_hash;
*/
/* Unused
my $asm_cs;
my $cmp_cs;
my $asm_cs_vers;
my $asm_cs_name;
my $cmp_cs_vers;
my $cmp_cs_name;
if($mapper) {
$asm_cs = $mapper->assembled_CoordSystem();
$cmp_cs = $mapper->component_CoordSystem();
$asm_cs_name = $asm_cs->name();
$asm_cs_vers = $asm_cs->version();
$cmp_cs_name = $cmp_cs->name();
$cmp_cs_vers = $cmp_cs->version();
}
*/
long destSliceStart;
long destSliceEnd;
int destSliceStrand;
long destSliceLength;
//CoordSystem *destSliceCs;
char * destSliceSrName;
IDType destSliceSrId = 0;
//AssemblyMapperAdaptor *asma;
if (destSlice) {
destSliceStart = Slice_getStart(destSlice);
destSliceEnd = Slice_getEnd(destSlice);
destSliceStrand = Slice_getStrand(destSlice);
destSliceLength = Slice_getLength(destSlice);
//??destSliceCs = Slice_getCoordSystem(destSlice);
destSliceSrName = Slice_getSeqRegionName(destSlice);
destSliceSrId = Slice_getSeqRegionId(destSlice);
//??asma = DBAdaptor_getAssemblyMapperAdaptor(ea->dba);
}
ResultRow *row;
while ((row = sth->fetchRow(sth))) {
IDType id = row->getLongLongAt(row,0);
IDType analysisId = row->getLongLongAt(row,1);
IDType seqRegionId = row->getLongLongAt(row,2);
long seqRegionStart = row->getLongAt(row,3);
long seqRegionEnd = row->getLongAt(row,4);
int seqRegionStrand = row->getIntAt(row,5);
char *hitName = row->getStringAt(row,6);
double score = row->getDoubleAt(row,7);
char *scoreType = row->getStringAt(row,8);
int spliceCanonical = row->getIntAt(row,9);
// get the analysis object
Analysis *analysis = AnalysisAdaptor_fetchByDbID(aa, analysisId);
/*
// need to get the internal_seq_region, if present
$seq_region_id = $self->get_seq_region_id_internal($seq_region_id);
#get the slice object
my $slice = $slice_hash{"ID:".$seq_region_id};
if(!$slice) {
$slice = $sa->fetch_by_seq_region_id($seq_region_id);
$slice_hash{"ID:".$seq_region_id} = $slice;
$sr_name_hash{$seq_region_id} = $slice->seq_region_name();
$sr_cs_hash{$seq_region_id} = $slice->coord_system();
}
my $sr_name = $sr_name_hash{$seq_region_id};
my $sr_cs = $sr_cs_hash{$seq_region_id};
*/
if (! IDHash_contains(sliceHash, seqRegionId)) {
IDHash_add(sliceHash, seqRegionId, SliceAdaptor_fetchBySeqRegionId(sa, seqRegionId, POS_UNDEF, POS_UNDEF, STRAND_UNDEF));
}
Slice *slice = IDHash_getValue(sliceHash, seqRegionId);
Slice *iseSlice = slice;
char *srName = Slice_getSeqRegionName(slice);
CoordSystem *srCs = Slice_getCoordSystem(slice);
//
// remap the feature coordinates to another coord system
// if a mapper was provided
//
if (assMapper != NULL) {
MapperRangeSet *mrs;
// Slightly suspicious about need for this if statement so left in perl statements for now
if (destSlice != NULL &&
assMapper->objectType == CLASS_CHAINEDASSEMBLYMAPPER) {
mrs = ChainedAssemblyMapper_map(assMapper, srName, seqRegionStart, seqRegionEnd, seqRegionStrand, srCs, 1, destSlice);
} else {
mrs = AssemblyMapper_fastMap(assMapper, srName, seqRegionStart, seqRegionEnd, seqRegionStrand, srCs, NULL);
}
// skip features that map to gaps or coord system boundaries
//next FEATURE if (!defined($seq_region_id));
if (MapperRangeSet_getNumRange(mrs) == 0) {
continue;
}
MapperRange *range = MapperRangeSet_getRangeAt(mrs, 0);
if (range->rangeType == MAPPERRANGE_GAP) {
fprintf(stderr,"Got a mapper gap in gene obj_from_sth - not sure if this is allowed\n");
exit(1);
} else {
MapperCoordinate *mc = (MapperCoordinate *)range;
seqRegionId = mc->id;
seqRegionStart = mc->start;
seqRegionEnd = mc->end;
seqRegionStrand = mc->strand;
}
MapperRangeSet_free(mrs);
/* Was - but identical if and else so why test???
#get a slice in the coord system we just mapped to
if($asm_cs == $sr_cs || ($cmp_cs != $sr_cs && $asm_cs->equals($sr_cs))) {
$slice = $slice_hash{"ID:".$seq_region_id} ||=
$sa->fetch_by_seq_region_id($seq_region_id);
} else {
$slice = $slice_hash{"ID:".$seq_region_id} ||=
$sa->fetch_by_seq_region_id($seq_region_id);
}
*/
// Instead...
if (! IDHash_contains(sliceHash, seqRegionId)) {
IDHash_add(sliceHash, seqRegionId, SliceAdaptor_fetchBySeqRegionId(sa, seqRegionId, POS_UNDEF, POS_UNDEF, STRAND_UNDEF));
}
iseSlice = IDHash_getValue(sliceHash, seqRegionId);
}
//
// If a destination slice was provided convert the coords
// If the dest_slice starts at 1 and is foward strand, nothing needs doing
//
if (destSlice != NULL) {
if (destSliceStart != 1 || destSliceStrand != 1) {
if (destSliceStrand == 1) {
seqRegionStart = seqRegionStart - destSliceStart + 1;
seqRegionEnd = seqRegionEnd - destSliceStart + 1;
} else {
long tmpSeqRegionStart = seqRegionStart;
seqRegionStart = destSliceEnd - seqRegionEnd + 1;
seqRegionEnd = destSliceEnd - tmpSeqRegionStart + 1;
seqRegionStrand = -seqRegionStrand;
}
}
// throw away features off the end of the requested slice
if (seqRegionEnd < 1 || seqRegionStart > destSliceLength || (destSliceSrId != seqRegionId)) {
continue;
}
iseSlice = destSlice;
}
IntronSupportingEvidence *ise = IntronSupportingEvidence_new();
IntronSupportingEvidence_setStart (ise, seqRegionStart);
IntronSupportingEvidence_setEnd (ise, seqRegionEnd);
IntronSupportingEvidence_setStrand (ise, seqRegionStrand);
IntronSupportingEvidence_setSlice (ise, iseSlice);
IntronSupportingEvidence_setAnalysis (ise, analysis);
IntronSupportingEvidence_setAdaptor (ise, (BaseAdaptor *)isea);
IntronSupportingEvidence_setDbID (ise, id);
IntronSupportingEvidence_setHitName (ise, hitName);
IntronSupportingEvidence_setScore (ise, score);
IntronSupportingEvidence_setScoreType (ise, scoreType);
IntronSupportingEvidence_setIsSpliceCanonical(ise, spliceCanonical);
Vector_addElement(features, ise);
}
return features;
}
Vector *GenomicAlignAdaptor_objectsFromStatementHandle(GenomicAlignAdaptor *gaa, StatementHandle *sth,
int reverse) {
Vector *results = Vector_new();
ResultRow *row;
DNAFragAdaptor *dfa;
IDType consensusDNAFragId;
IDType queryDNAFragId;
int consensusStart;
int consensusEnd;
int queryStart;
int queryEnd;
int queryStrand;
IDType methodLinkId;
double score;
double percId;
char *cigarString;
dfa = ComparaDBAdaptor_getDNAFragAdaptor(gaa->dba);
while ((row = sth->fetchRow(sth))) {
GenomicAlign *genomicAlign;
char *alignmentType;
if (reverse) {
queryDNAFragId = row->getLongLongAt(row,0);
queryStart = row->getIntAt(row,1);
queryEnd = row->getIntAt(row,2);
consensusDNAFragId = row->getLongLongAt(row,3);
consensusStart = row->getIntAt(row,4);
consensusEnd = row->getIntAt(row,5);
} else {
consensusDNAFragId = row->getLongLongAt(row,0);
consensusStart = row->getIntAt(row,1);
consensusEnd = row->getIntAt(row,2);
queryDNAFragId = row->getLongLongAt(row,3);
queryStart = row->getIntAt(row,4);
queryEnd = row->getIntAt(row,5);
}
queryStrand = row->getIntAt(row,6);
methodLinkId = row->getLongLongAt(row,7);
score = row->getDoubleAt(row,8);
percId = row->getDoubleAt(row,9);
cigarString = row->getStringAt(row,10);
alignmentType = GenomicAlignAdaptor_alignmentTypeByMethodLinkId(gaa, methodLinkId);
if (reverse) {
StrUtil_strReplChrs(cigarString,"DI","ID");
// alignment of the opposite strand
if (queryStrand == -1) {
cigarString = CigarStrUtil_reverse(cigarString, strlen(cigarString));
}
}
genomicAlign = GenomicAlign_new();
GenomicAlign_setAdaptor(genomicAlign, (BaseAdaptor *)gaa);
GenomicAlign_setConsensusDNAFrag(genomicAlign, DNAFragAdaptor_fetchByDbID(dfa,consensusDNAFragId));
GenomicAlign_setConsensusStart(genomicAlign, consensusStart);
GenomicAlign_setConsensusEnd(genomicAlign, consensusEnd);
GenomicAlign_setQueryDNAFrag(genomicAlign, DNAFragAdaptor_fetchByDbID(dfa,queryDNAFragId));
GenomicAlign_setQueryStart(genomicAlign, queryStart);
GenomicAlign_setQueryEnd(genomicAlign, queryEnd);
GenomicAlign_setQueryStrand(genomicAlign, queryStrand);
GenomicAlign_setAlignmentType(genomicAlign, alignmentType);
GenomicAlign_setScore(genomicAlign, score);
GenomicAlign_setPercentId(genomicAlign, percId);
GenomicAlign_setCigarString(genomicAlign, cigarString);
Vector_addElement(results, genomicAlign);
}
return results;
}
