static void fakePslFromGenePred(char *db, char *fileTbl, char *pslOut, char *cdsOut)
/* check a genePred */
{
struct genePredReader *gpr;
struct genePred *gp;
FILE *pslFh = mustOpen(pslOut, "w");
FILE *cdsFh = mustOpen(cdsOut, "w");
struct hash *chromHash = getChromHash(db);
if (fileExists(fileTbl))
{
gpr = genePredReaderFile(fileTbl, NULL);
}
else
{
struct sqlConnection *conn = hAllocConn(db);
gpr = genePredReaderQuery(conn, fileTbl, NULL);
hFreeConn(&conn);
}
while ((gp = genePredReaderNext(gpr)) != NULL)
{
cnvGenePred(chromHash, gp, pslFh, cdsFh);
}
genePredReaderFree(&gpr);
carefulClose(&pslFh);
carefulClose(&cdsFh);
}
int palOutPredsInBeds(struct sqlConnection *conn, struct cart *cart,
struct bed *beds, char *table )
/* output the alignments whose names and coords match a bed*/
{
struct genePred *list = NULL;
for(; beds; beds = beds->next)
{
char where[10 * 1024];
sqlSafefFrag(where, sizeof where,
"name = '%s' and chrom='%s' and txEnd > %d and txStart <= %d",
beds->name, beds->chrom, beds->chromStart, beds->chromEnd);
struct genePredReader *reader = genePredReaderQuery( conn, table, where);
struct genePred *pred;
while ((pred = genePredReaderNext(reader)) != NULL)
slAddHead(&list, pred);
genePredReaderFree(&reader);
}
int outCount = 0;
if (list != NULL)
{
slReverse(&list);
outCount = palOutPredList( conn, cart, list);
genePredFreeList(&list);
}
return outCount;
}
struct genePred *getPredsFromBeds(char *file, char *table, char *db)
{
struct sqlConnection *conn = hAllocConn(db);
struct lineFile *lf = lineFileOpen(file, TRUE);
char *words[5000];
int wordsRead;
struct genePred *list = NULL;
while( (wordsRead = lineFileChopNext(lf, words, sizeof(words)/sizeof(char *)) ))
{
if (wordsRead != 4)
errAbort("file '%s' must be bed4. Line %d has %d fields",
file, lf->lineIx, wordsRead);
char where[10 * 1024];
sqlSafefFrag(where, sizeof where,
"name = '%s' and chrom='%s' and txStart = %d and txEnd = %d",
words[3], words[0], sqlUnsigned(words[1]), sqlUnsigned(words[2]));
//printf("table %s where %s\n",table,where);
struct genePredReader *reader = genePredReaderQuery( conn, table, where);
struct genePred *pred;
while ((pred = genePredReaderNext(reader)) != NULL)
slAddHead(&list, pred);
genePredReaderFree(&reader);
}
hFreeConn(&conn);
if (list != NULL)
slReverse(&list);
return list;
}
static void checkGenePred(char *fileTbl)
/* check a genePred file or table */
{
struct sqlConnection *conn = NULL;
struct genePredReader *gpr;
struct genePred *gp;
int iRec = 0;
if (fileExists(fileTbl))
{
gpr = genePredReaderFile(fileTbl, NULL);
}
else if (gDb != NULL)
{
conn = hAllocConn(gDb);
gpr = genePredReaderQuery(conn, fileTbl, NULL);
}
else
{
errAbort("file %s doesn't exist, must specify -db=db if this is a table", fileTbl);
}
while ((gp = genePredReaderNext(gpr)) != NULL)
{
checkAGenePred(fileTbl, ++iRec, gp);
genePredFree(&gp);
}
genePredReaderFree(&gpr);
hFreeConn(&conn);
}
struct genePred *genePredReaderLoadQuery(struct sqlConnection* conn,
char* table, char* where)
/* Function that encapsulates doing a query and loading the results */
{
struct genePredReader *gpr = genePredReaderQuery(conn, table, where);
struct genePred *gpList = genePredReaderAll(gpr);
genePredReaderFree(&gpr);
return gpList;
}
static void getGeneAnns(struct sqlConnection *conn, struct hash *refSeqVerInfoTbl, char *outFile)
/* get request genePred annotations from database */
{
struct genePredReader *gpr = genePredReaderQuery(conn, "refGene", NULL);
FILE *fh = mustOpen(outFile, "w");
struct genePred *gp;
while ((gp = genePredReaderNext(gpr)) != NULL)
{
processGenePred(fh, refSeqVerInfoTbl, gp);
genePredFree(&gp);
}
carefulClose(&fh);
genePredReaderFree(&gpr);
}
/* load one or more genePreds from the database */
struct genePred *getPredsForName(char *name, char *geneTable, char *db)
{
struct sqlConnection *conn = hAllocConn(db);
struct genePred *list = NULL;
char splitTable[HDB_MAX_TABLE_STRING];
struct genePred *gene;
boolean hasBin;
struct genePredReader *reader;
boolean found = hFindSplitTable(db, NULL, geneTable,
splitTable, &hasBin);
if (!found)
errAbort("can't find table %s\n", geneTable);
char extra[2048];
if (onlyChrom != NULL)
safef(extra, sizeof extra, "name='%s' and chrom='%s'", name, onlyChrom);
else
safef(extra, sizeof extra, "name='%s'", name);
reader = genePredReaderQuery( conn, splitTable, extra);
while ((gene = genePredReaderNext(reader)) != NULL)
{
verbose(2, "got gene %s\n",gene->name);
slAddHead(&list, gene);
}
if (list == NULL)
errAbort("no genePred for gene %s in %s\n",name, geneTable);
slReverse(&list);
genePredReaderFree(&reader);
hFreeConn(&conn);
return list;
}
/* query the list of gene names from the frames table */
static struct genePred *queryPreds(char *dbName, char *geneTable)
{
struct sqlConnection *conn = hAllocConn(dbName);
struct genePred *list = NULL;
char buf[2048];
char *extra = NULL;
struct genePredReader *reader;
if (onlyChrom != NULL)
{
safef(buf, sizeof buf, "chrom='%s'", onlyChrom);
extra = buf;
}
reader = genePredReaderQuery( conn, geneTable, extra);
list = genePredReaderAll(reader);
hFreeConn(&conn);
return list;
}
static void printCcdsHgGeneUrl(struct sqlConnection *conn, char *ccdsId, char* kgId)
/* output a URL to hgGene for a ccds */
{
char where[128];
struct genePredReader *gpr;
struct genePred *ccdsGene = NULL, *kgGene = NULL;
/* get ccds genePred to get location */
sqlSafefFrag(where, sizeof(where), "chrom = '%s' and name = '%s'", seqName, ccdsId);
gpr = genePredReaderQuery(conn, "ccdsGene", where);
ccdsGene = genePredReaderAll(gpr);
genePredReaderFree(&gpr);
if (ccdsGene == NULL)
errAbort("%s not found in ccdsGene table for chrom %s", ccdsId, seqName);
else if (ccdsGene->next != NULL)
errAbort("multiple %s rows found in ccdsGene table for chrom %s", ccdsId, seqName);
/* get KG genePred, as need exact location for link */
sqlSafefFrag(where, sizeof(where), "name = '%s' and strand = '%s'", kgId,
ccdsGene->strand);
gpr = genePredReaderRangeQuery(conn, "knownGene", seqName,
ccdsGene->txStart, ccdsGene->txEnd, where);
kgGene = genePredReaderAll(gpr);
genePredReaderFree(&gpr);
if (kgGene == NULL)
errAbort("%s not found in knownGene table for chrom %s", kgId, seqName);
else if (kgGene->next != NULL)
errAbort("multiple %s rows found in knownGene table for chrom %s", kgId, seqName);
printf("../cgi-bin/hgGene?%s&%s=%s&%s=%s&%s=%s&%s=%d&%s=%d",
cartSidUrlString(cart),
"db", database,
"hgg_gene", kgId,
"hgg_chrom", seqName,
"hgg_start", kgGene->txStart,
"hgg_end", kgGene->txEnd);
genePredFree(&ccdsGene);
genePredFree(&kgGene);
}
void kgGetCds(char *db, char *spDb, char *geneTable, FILE *outf)
/* get CDS info */
{
struct sqlConnection *conn = NULL;
struct genePred *gp;
int cdsCnt;
struct genePredReader *gpr;
int iExon, exonStart, exonEnd;
if (db != NULL)
conn = sqlConnect(db);
gpr = genePredReaderQuery(conn, geneTable, NULL);
while ((gp = genePredReaderNext(gpr)) != NULL)
{
cdsCnt = 0;
for (iExon = 0; iExon < gp->exonCount; iExon++)
{
if (genePredCdsExon(gp, iExon, &exonStart, &exonEnd))
{
sprintf(cdsBloc[cdsCnt], "%d-%d;", exonStart, exonEnd);
cdsCnt++;
}
}
if (cdsCnt > 0)
{
processAlign(db, spDb, gp->name, cdsCnt, outf);
}
else
{
fprintf(stderr, "%s does not have cds.\n", gp->name);
}
}
sqlDisconnect(&conn);
}
struct genePred *getOverlappingGeneDb(struct genePred **list, char *table, char *chrom, int cStart, int cEnd, char *name, int *retOverlap, char *db)
{
/* read all genes from a table find the gene with the biggest overlap.
Cache the list of genes to so we only read it once */
struct genePred *el = NULL, *bestMatch = NULL, *gp = NULL;
int overlap = 0 , bestOverlap = 0, i;
int *eFrames;
if (list == NULL)
return NULL;
if (*list == NULL)
{
struct genePred *gpList = NULL;
struct sqlConnection *conn = sqlConnect(db);
struct genePredReader *gpr = NULL;
if (!hTableExistsDb(db,table))
table = altTable;
if (!hTableExistsDb(db,table))
{
verbose(2,"no table %s in %s\n",table, db);
return NULL;
}
gpr = genePredReaderQuery(conn, table, NULL);
verbose(1,"Loading Predictions from %s in %s\n",table, db);
gpList = genePredReaderAll(gpr);
if (gpList != NULL)
{
hashAdd(geneListHash, db, gpList);
*list = gpList;
}
sqlDisconnect(&conn);
}
for (el = *list; el != NULL; el = el->next)
{
if (chrom != NULL && el->chrom != NULL)
{
overlap = 0;
if ( sameString(chrom, el->chrom))
{
for (i = 0 ; i<(el->exonCount); i++)
{
overlap += positiveRangeIntersection(cStart,cEnd, el->exonStarts[i], el->exonEnds[i]) ;
}
if (overlap > 20 && sameString(name, el->name))
{
bestMatch = el;
bestOverlap = overlap;
*retOverlap = bestOverlap;
}
if (overlap > bestOverlap)
{
bestMatch = el;
bestOverlap = overlap;
*retOverlap = bestOverlap;
}
}
}
}
if (bestMatch != NULL)
{
/* Allocate genePred and fill in values. */
AllocVar(gp);
gp->name = cloneString(bestMatch->name);
gp->chrom = cloneString(bestMatch->chrom);
gp->strand[1] = bestMatch->strand[1];
gp->strand[0] = bestMatch->strand[0];
gp->txStart = bestMatch->txStart;
gp->txEnd = bestMatch->txEnd;
gp->cdsStart = bestMatch->cdsStart;
gp->cdsEnd = bestMatch->cdsEnd;
gp->exonCount = bestMatch->exonCount;
AllocArray(gp->exonStarts, bestMatch->exonCount);
AllocArray(gp->exonEnds, bestMatch->exonCount);
for (i=0; i<bestMatch->exonCount; ++i)
{
gp->exonStarts[i] = bestMatch->exonStarts[i] ;
gp->exonEnds[i] = bestMatch->exonEnds[i] ;
}
gp->optFields = bestMatch->optFields;
gp->id = bestMatch->id;
if (bestMatch->optFields & genePredName2Fld)
gp->name2 = cloneString(bestMatch->name2);
else
gp->name2 = NULL;
if (bestMatch->optFields & genePredCdsStatFld)
{
gp->cdsStartStat = bestMatch->cdsStartStat;
gp->cdsEndStat = bestMatch->cdsEndStat;
}
if (bestMatch->optFields & genePredExonFramesFld)
{
gp->exonFrames = AllocArray(eFrames, bestMatch->exonCount);
for (i = 0; i < bestMatch->exonCount; i++)
gp->exonFrames[i] = bestMatch->exonFrames[i];
}
eFrames = gp->exonFrames;
}
return gp;
}
