Vector *HomologyAdaptor_fetchHomologuesOfGene(HomologyAdaptor *ha, char *sp, char *gene) {
char qStr[1024];
char *species;
int nRelationship;
IDType *relationshipIds;
Vector *genes;
int i;
species = StrUtil_copyString(&species,sp,0);
species = StrUtil_strReplChr(species,'_',' ');
sprintf(qStr,
"select grm.gene_relationship_id "
" from gene_relationship_member grm, "
" genome_db gd "
" where gd.genome_db_id = grm.genome_db_id "
" and gd.name = '%s' "
" and grm.member_stable_id = '%s' "
" group by grm.gene_relationship_id", species, gene);
nRelationship = HomologyAdaptor_getRelationships(ha, qStr,&relationshipIds);
genes = Vector_new();
for (i=0;i<nRelationship;i++) {
Vector *homols;
sprintf(qStr,
"select grm.member_stable_id,"
" gd.name,"
" grm.chromosome,"
" grm.chrom_start,"
" grm.chrom_end"
" from gene_relationship_member grm,"
" genome_db gd"
" where grm.gene_relationship_id = " IDFMTSTR
" and grm.genome_db_id = gd.genome_db_id "
" and NOT (grm.member_stable_id = '%s')", relationshipIds[i], gene);
homols = HomologyAdaptor_getHomologues(ha, qStr);
Vector_append(genes,homols);
Vector_free(homols);
}
free(relationshipIds);
free(species);
return genes;
}
static bool read_piref(Streader* sr, int32_t index, void* userdata)
{
rassert(sr != NULL);
rassert(userdata != NULL);
Order_list* ol = userdata;
// Read the Pattern instance reference
Pat_inst_ref* p = PAT_INST_REF_AUTO;
if (!Streader_read_piref(sr, p))
return false;
// Check if the Pattern instance is already used
Index_mapping* key = INDEX_MAPPING_AUTO;
key->p = *p;
key->ol_index = index;
if (AAtree_contains(ol->index_map, key))
{
Streader_set_error(
sr,
"Duplicate occurrence of pattern instance"
" [%" PRId16 ", %" PRId16 "]",
p->pat, p->inst);
return false;
}
// Add the reference to our containers
if (!Vector_append(ol->pat_insts, p))
{
Streader_set_memory_error(
sr, "Could not allocate memory for order list");
return false;
}
Index_mapping* im = new_Index_mapping(key);
if (im == NULL || !AAtree_ins(ol->index_map, im))
{
memory_free(im);
Streader_set_memory_error(
sr, "Could not allocate memory for order list");
return false;
}
return true;
}
Vector *HomologyAdaptor_fetchHomologuesOfGeneInSpecies(HomologyAdaptor *ha,
char *sp, char *gene, char *hSp) {
char qStr[1024];
Vector *genes;
IDType *relationshipIds;
int nRelationship;
int i;
char *hSpecies;
char *species;
species = StrUtil_copyString(&species,sp,0);
species = StrUtil_strReplChr(species,'_',' ');
hSpecies = StrUtil_copyString(&hSpecies,hSp,0);
hSpecies = StrUtil_strReplChr(hSpecies,'_',' ');
sprintf(qStr,
"select grm.gene_relationship_id "
" from gene_relationship_member grm, "
" genome_db gd "
" where gd.genome_db_id = grm.genome_db_id "
" and gd.name = '%s' "
" and grm.member_stable_id = '%s' "
" group by grm.gene_relationship_id", species, gene);
nRelationship = HomologyAdaptor_getRelationships(ha,qStr,&relationshipIds);
genes = Vector_new();
for (i=0;i<nRelationship;i++) {
Vector *homols = HomologyAdaptor_fetchHomologuesBySpeciesRelationshipId(ha,hSpecies,relationshipIds[i]);
Vector_append(genes, homols);
Vector_free(homols);
}
free(relationshipIds);
free(species);
free(hSpecies);
return genes;
}
// Test your vector here
int main() {
Vector* temp = Vector_create(); //create a vector
Vector_append( temp, "a");//1
Vector_append( temp, "b");//2
Vector_append( temp, "c");//3
Vector_append( temp, "d");//4
Vector_append( temp, NULL);//5
Vector_append( temp, "f");//6
Vector_append( temp, "g");//7
Vector_append( temp, NULL);//8
Vector_append( temp, "");//9
Vector_append( temp, "h");//10
Vector_append( temp, "i");//11
Vector_append( temp, NULL);//12
Vector_resize( temp, 20 );
if( Vector_size(temp) != 20 || strcmp( Vector_get( temp, 10 ), "i" ) || Vector_get( temp, 15 ) != NULL )
perror( "something wrong.\n");
//done for append, resize, get, size
Vector_set( temp, 19, "caibi" );
Vector_insert( temp, 20, "niubi" );
Vector_insert( temp, 30, "wori" );
if( Vector_size(temp) != 31 || strcmp( Vector_get( temp, 19 ), "caibi" ) || strcmp( Vector_get( temp, 20 ), "niubi" ) || Vector_get( temp, 15 ) != NULL || strcmp( Vector_get( temp, 30 ), "wori" ) )
perror( "something wrong.\n");
Vector_delete( temp, 11 );
Vector_delete( temp, 27 );
Vector_delete( temp, 1 );
Vector_delete( temp, 18 );
if( Vector_size(temp) != 27 || strcmp( Vector_get( temp, 4 ), "f" ) || strcmp( Vector_get( temp, 26 ), "wori") || Vector_get( temp, 18 ) !=NULL || strcmp( Vector_get( temp, 17 ), "caibi") )
perror( "something wrong.\n");
Vector_destroy( temp );
return 0;
}
int
Stack_push(Stack s, void *value)
{
return Vector_append(&s->vector, 1, value);
}
Vector *GenomicAlignAdaptor_fetchAllByDNAFragGenomeDB(GenomicAlignAdaptor *gaa,
DNAFrag *dnaFrag, GenomeDB *targetGenome, int *startP, int *endP,
char *alignmentType) {
Vector *result = NULL;
GenomeDB *genomeCons;
IDType methodLinkId;
GenomeDB *genomeQuery;
Vector *mergedAligns;
int ok = 1;
if (!dnaFrag) {
fprintf(stderr, "Error: dnaFrag argument must be non NULL\n");
ok = 0;
}
if (ok) {
methodLinkId = GenomicAlignAdaptor_methodLinkIdByAlignmentType(gaa, alignmentType);
genomeCons = DNAFrag_getGenomeDB(dnaFrag);
genomeQuery = targetGenome;
// direct or indirect ??
if (GenomeDB_hasConsensus(genomeCons, genomeQuery, methodLinkId) ||
GenomeDB_hasQuery(genomeCons, genomeQuery, methodLinkId)) {
result = GenomicAlignAdaptor_fetchAllByDNAFragGenomeDBDirect(gaa,
dnaFrag, targetGenome, startP, endP, methodLinkId);
} else {
// indirect checks
Vector *linkedCons = GenomeDB_linkedGenomesByMethodLinkId(genomeCons, methodLinkId);
Vector *linkedQuery = GenomeDB_linkedGenomesByMethodLinkId(genomeQuery, methodLinkId);
// there are not many genomes, square effort is cheap
Vector *linked = Vector_new();
Vector *set1 = Vector_new();
mergedAligns = Vector_new();
int i;
for (i=0; i<Vector_getNumElement(linkedCons); i++) {
int j;
GenomeDB *g1 = Vector_getElementAt(linkedCons, i);
for (j=0; j<Vector_getNumElement(linkedQuery); j++) {
GenomeDB *g2 = Vector_getElementAt(linkedQuery, i);
if (g1 == g2) {
Vector_addElement(linked, g1);
}
}
}
Vector_free(linkedCons);
Vector_free(linkedQuery);
// collect GenomicAligns from all linked genomes
for (i=0; i<Vector_getNumElement(linked); i++) {
GenomeDB *g = Vector_getElementAt(linked, i);
Vector *gres = GenomicAlignAdaptor_fetchAllByDNAFragGenomeDBDirect(gaa,
dnaFrag, g, startP, endP, methodLinkId);
Vector_append(set1, gres);
Vector_free(gres);
}
// go from each dnafrag in the result set to target_genome
// there is room for improvement here: create start end
// my %frags = map { $_->query_dnafrag->dbID => $_->query_dnafrag } @$set1;
for (i=0; i<Vector_getNumElement(set1); i++) {
GenomicAlign *alignA = Vector_getElementAt(set1,i);
DNAFrag *frag = GenomicAlign_getQueryDNAFrag(alignA);
int qStart = GenomicAlign_getQueryStart(alignA);
int qEnd = GenomicAlign_getQueryEnd(alignA);
Vector *dres = GenomicAlignAdaptor_fetchAllByDNAFragGenomeDBDirect(gaa,
frag, genomeQuery, &qStart, &qEnd, methodLinkId);
int j;
for (j=0; j<Vector_getNumElement(dres); j++) {
GenomicAlign *alignB = Vector_getElementAt(dres,j);
GenomicAlignAdaptor_addDerivedAlignments(gaa, mergedAligns, alignA, alignB);
}
Vector_free(dres);
}
// NIY freeing
result = mergedAligns;
}
}
return result;
}
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;
}
// Note I didn't implement the stable id fetching uggliness here. I'll probably make a separate method for that
// if necessary
Vector *BaseAdaptor_uncachedFetchAllByDbIDList(BaseAdaptor *ba, Vector *idList, Slice *slice) {
if ( idList == NULL) {
fprintf(stderr, "id_list list reference argument is required - bye!");
return NULL;
}
char constraintPref[1024];
if (!Vector_getNumElement(idList)) {
return Vector_new();
}
NameTableType *tables = ba->getTables();
char **t = (*tables)[0];
sprintf(constraintPref, "%s.%s_id ", t[SYN], t[NAME] );
// Ensure that we do not exceed MySQL's max_allowed_packet (defaults to
// 1 MB) splitting large queries into smaller queries of at most 256 KB.
// Assuming a (generous) average dbID string
// length of 16, this means 16384 dbIDs in each query.
int maxSize = 16384;
// Uniquify the list
IDHash *idListHash = IDHash_new(IDHASH_MEDIUM);
int i;
for (i=0; i<Vector_getNumElement(idList); i++) {
IDType id = *(IDType *)(Vector_getElementAt(idList, i));
if (!IDHash_contains(idListHash, id)) {
IDHash_add(idListHash, id, &trueVal);
}
}
IDType *uniqueIds = IDHash_getKeys(idListHash);
int nUniqueId = IDHash_getNumValues(idListHash);
IDHash_free(idListHash, NULL);
Vector *out = Vector_new();
int lenNum;
for (i=0; i<nUniqueId; i+=maxSize) {
char *constraint = NULL;
if ((constraint = (char *)calloc(655500,sizeof(char))) == NULL) {
fprintf(stderr,"Failed allocating constraint\n");
return out;
}
strcpy(constraint, constraintPref);
// Special case for one remaining Id
if (i == nUniqueId-1) {
sprintf(constraint, "%s = "IDFMTSTR, constraint, uniqueIds[i]);
} else {
char tmpStr[1024];
int endPoint = sprintf(constraint, "%s IN (", constraint);
int j;
for (j=0; j<maxSize && j+i<nUniqueId; j++) {
if (j!=0) {
constraint[endPoint++] = ',';
constraint[endPoint++] = ' ';
}
lenNum = sprintf(tmpStr, IDFMTSTR, uniqueIds[i+j]);
memcpy(&(constraint[endPoint]), tmpStr, lenNum);
endPoint+=lenNum;
}
constraint[endPoint++] = ')';
constraint[endPoint] = '\0';
}
Vector *resChunk = BaseAdaptor_genericFetch(ba, constraint, NULL, slice);
Vector_append(out, resChunk);
Vector_free(resChunk);
free(constraint);
}
free(uniqueIds);
return out;
}
