static boolean ctdExists(struct section *section,
struct sqlConnection *conn, char *geneId)
/* Return TRUE if CTD database exists and it has an entry with the gene symbol */
{
char query[1024];
char *geneSymbol;
if (isRgdGene(conn))
{
if (sqlTableExists(conn, "rgdGene2Xref") == FALSE) return FALSE;
}
else
{
if (sqlTableExists(conn, "kgXref") == FALSE) return FALSE;
}
if (sqlTableExists(conn, "hgFixed.ctdSorted") == TRUE)
{
if (isRgdGene(conn))
{
sqlSafef(query, sizeof(query), "select ChemicalId from rgdGene2Xref x, hgFixed.ctdSorted c"
" where x.info=c.GeneSymbol and infoType = 'Name' and rgdGeneId='%s' limit 1", geneId);
geneSymbol = sqlQuickString(conn, query);
}
else
{
sqlSafef(query, sizeof(query), "select ChemicalId from kgXref x, hgFixed.ctdSorted c"
" where x.geneSymbol=c.GeneSymbol and kgId='%s' limit 1", geneId);
geneSymbol = sqlQuickString(conn, query);
}
if (geneSymbol != NULL) return(TRUE);
}
return(FALSE);
}
static int reactomeCount(struct pathwayLink *pl, struct sqlConnection *conn,
char *geneId)
/* Count up number of hits. */
{
int ret = 0;
char query[256];
char *spID, *chp;
char condStr[256];
char *origSpID;
/* check the existence of kgXref table first */
if (!isRgdGene(conn))
{
if (!sqlTableExists(conn, "kgXref")) return(0);
}
else
{
if (!sqlTableExists(conn, "rgdGene2Xref")) return(0);
}
if (isRgdGene(conn))
{
sqlSafefFrag(condStr, sizeof(condStr), "name='%s'", geneId);
spID = sqlGetField(database, "rgdGene2ToUniProt", "value", condStr);
}
else
{
sqlSafefFrag(condStr, sizeof(condStr), "kgID='%s'", geneId);
spID = sqlGetField(database, "kgXref", "spID", condStr);
}
if (spID != NULL)
{
origSpID = cloneString(spID);
/* convert splice variant UniProt ID to its main root ID */
chp = strstr(spID, "-");
if (chp != NULL) *chp = '\0';
if (!isRgdGene(conn))
{
sqlSafef(query, sizeof(query),
"select count(*) from %s.spReactomeEvent, %s.spVariant, %s.kgXref where kgID='%s' and kgXref.spID=variant and variant = '%s' and spReactomeEvent.spID=parent",
PROTEOME_DB_NAME, PROTEOME_DB_NAME, database, geneId, origSpID);
}
else
{
sqlSafef(query, sizeof(query),
"select count(*) from %s.spReactomeEvent, %s.spVariant, %s.rgdGene2ToUniProt where name='%s' and value=variant and variant = '%s' and spReactomeEvent.spID=parent",
PROTEOME_DB_NAME, PROTEOME_DB_NAME, database, geneId, origSpID);
}
ret = sqlQuickNum(conn, query);
}
return ret;
}
static int keggCount(struct pathwayLink *pl, struct sqlConnection *conn,
char *geneId)
/* Count up number of hits. */
{
char query[256];
if (!isRgdGene(conn))
{
sqlSafef(query, sizeof(query),
"select count(*) from keggPathway k, kgXref x where k.kgID=x.kgId and x.kgId='%s'", geneId);
}
else
{
sqlSafef(query, sizeof(query),
"select count(*) from rgdGene2KeggPathway k, rgdGene2 x where k.rgdId=x.name and x.name='%s'", geneId);
}
return sqlQuickNum(conn, query);
}
char *getSwissProtAcc(struct sqlConnection *conn, struct sqlConnection *spConn,
char *geneId)
/* Look up SwissProt id. Return NULL if not found. FreeMem this when done.
* spConn is existing SwissProt database conn. May be NULL. */
{
char *proteinSql = genomeSetting("proteinSql");
char query[256];
char *someAcc, *primaryAcc = NULL;
if (isRgdGene(conn))
{
return(getRgdGeneUniProtAcc(curGeneId, conn));
}
sqlSafef(query, sizeof(query), proteinSql, geneId);
someAcc = sqlQuickString(conn, query);
if (someAcc == NULL || someAcc[0] == 0)
return NULL;
primaryAcc = spFindAcc(spConn, someAcc);
freeMem(someAcc);
return primaryAcc;
}
static void keggLink(struct pathwayLink *pl, struct sqlConnection *conn,
char *geneId)
/* Print out kegg database link. */
{
char query[512], **row;
struct sqlResult *sr;
if (isRgdGene(conn))
{
sqlSafef(query, sizeof(query),
"select distinct k.locusID, k.mapID, keggMapDesc.description"
" from rgdGene2KeggPathway k, keggMapDesc, rgdGene2 x"
" where k.rgdId=x.name "
" and x.name='%s'"
" and k.mapID = keggMapDesc.mapID"
, geneId);
}
else
{
sqlSafef(query, sizeof(query),
"select k.locusID, k.mapID, keggMapDesc.description"
" from keggPathway k, keggMapDesc, kgXref x"
" where k.kgID=x.kgId "
" and x.kgID='%s'"
" and k.mapID = keggMapDesc.mapID"
, geneId);
}
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
hPrintf("<A HREF=\"http://www.genome.ad.jp/kegg-bin/show_pathway?%s+%s\" TARGET=_blank>",
row[1], row[0]);
hPrintf("%s</A> - %s<BR>", row[1], row[2]);
}
sqlFreeResult(&sr);
}
void cartMain(struct cart *theCart)
/* We got the persistent/CGI variable cart. Now
* set up the globals and make a web page. */
{
hgBotDelay();
cart = theCart;
getDbAndGenome(cart, &database, &genome, oldVars);
initGenbankTableNames(database);
getGenomeSettings();
if (cartVarExists(cart, hggDoKgMethod))
doKgMethod();
else if (cartVarExists(cart, hggDoTxInfoDescription))
doTxInfoDescription();
else
{
struct sqlConnection *conn = NULL;
char *geneName = cartUsualString(cart, hggGene, NULL);
if (isEmpty(geneName))
{
// Silly googlebots.
hUserAbort("Error: the hgg_gene parameter is missing from the cart and the CGI params.");
}
/* if kgProtMap2 table exists, this means we are doing KG III */
if (hTableExists(database, "kgProtMap2")) kgVersion = KG_III;
conn = hAllocConn(database);
curGeneId = findGeneId(conn, geneName);
getGenePosition(conn);
curGenePred = getCurGenePred(conn);
curGeneName = getGeneName(curGeneId, conn);
spConn = hAllocConn(UNIPROT_DB_NAME);
swissProtAcc = getSwissProtAcc(conn, spConn, curGeneId);
if (isRgdGene(conn)) swissProtAcc=getRgdGeneUniProtAcc(curGeneId, conn);
/* Check command variables, and do the ones that
* don't want to put up the hot link bar etc. */
if (cartVarExists(cart, hggDoGetMrnaSeq))
doGetMrnaSeq(conn, curGeneId, curGeneName);
else if (cartVarExists(cart, hggDoWikiTrack))
doWikiTrack(conn);
else if (cartVarExists(cart, hggDoGetProteinSeq))
doGetProteinSeq(conn, curGeneId, curGeneName);
else if (cartVarExists(cart, hggDoRnaFoldDisplay))
doRnaFoldDisplay(conn, curGeneId, curGeneName);
else if (cartVarExists(cart, hggDoOtherProteinSeq))
doOtherProteinSeq(conn, curGeneName);
else if (cartVarExists(cart, hggDoOtherProteinAli))
doOtherProteinAli(conn, curGeneId, curGeneName);
else
{
/* Default case - start fancy web page. */
measureTiming = isNotEmpty(cartOptionalString(cart, "measureTiming"));
struct trackDb *tdb = hTrackDbForTrack(database, genomeSetting("knownGene"));
isGencode = trackDbSettingOn(tdb, "isGencode");
isGencode2 = trackDbSettingOn(tdb, "isGencode2");
cartWebStart(cart, database, "%s Gene %s (%s) Description and Page Index",
genome, curGeneName, isGencode2 ? curGeneId : curAlignId);
webMain(conn, tdb);
cartWebEnd();
}
hFreeConn(&spConn);
hFreeConn(&conn);
}
cartRemovePrefix(cart, hggDoPrefix);
}
char *linkGetUrl(struct link *link, struct sqlConnection *conn,
char *geneId)
/* Return URL string if possible or NULL if not. FreeMem this when done. */
{
char query[512];
struct sqlResult *sr;
char **row;
char *url = NULL;
/* Some special case code here for things that need to
* do more than check a table. */
if (sameString(link->name, "family"))
{
if (!hgNearOk(database))
return NULL;
}
if (sameString(link->name, "protBrowser"))
{
if (!hgPbOk(database))
return NULL;
/* special processing for PB, since we need the protein ID, instead everything key off from gene ID */
/* use UniProt accession instead of displayID, because display ID sometimes changes */
if (swissProtAcc == NULL || swissProtAcc[0] == 0)
return NULL;
if (isRgdGene(conn))
{
safef(query, sizeof(query), "../cgi-bin/pbGlobal?proteinID=%s", swissProtAcc);
}
else
{
safef(query, sizeof(query), "../cgi-bin/pbTracks?db=%s&proteinID=%s", database, swissProtAcc);
}
return(cloneString(query));
}
if (sameString(link->name, "tbSchema"))
{
char *geneTable = genomeSetting("knownGene");
struct trackDb *tdb = hTrackDbForTrack(sqlGetDatabase(conn), geneTable);
struct dyString *dy = NULL;
if (tdb == NULL)
return NULL;
dy = newDyString(256);
dyStringPrintf(dy, link->url, tdb->grp, geneTable, geneTable);
trackDbFree(&tdb);
addLinkExtras(link, dy);
return dyStringCannibalize(&dy);
}
geneId = cloneAndCut(geneId, link->preCutAt);
safef(query, sizeof(query), link->idSql, geneId);
sr = sqlGetResult(conn, query);
row = sqlNextRow(sr);
if (row != NULL && row[0][0] != 0) /* If not null or empty */
{
struct dyString *dy = newDyString(0);
char *name = cloneAndCut(row[0], link->postCutAt);
dyStringPrintf(dy, link->url, name, row[1], row[2], row[3]);
addLinkExtras(link, dy);
url = dyStringCannibalize(&dy);
freez(&name);
}
sqlFreeResult(&sr);
freeMem(geneId);
return url;
}
static void synonymPrint(struct section *section,
struct sqlConnection *conn, char *id)
/* Print out SwissProt comments - looking up typeId/commentVal. */
{
char *protAcc = getSwissProtAcc(conn, spConn, id);
char *spDisplayId;
char *refSeqAcc = "";
char *mrnaAcc = "";
char *oldDisplayId;
char condStr[255];
char *kgProteinID;
char *parAcc; /* parent accession of a variant splice protein */
char *chp;
if (isRgdGene(conn))
{
rgdGene2SynonymPrint(section,conn, id);
return;
}
if (sqlTablesExist(conn, "kgAlias"))
printAlias(id, conn);
if (sameWord(genome, "Zebrafish"))
{
char *xrefTable = "ensXRefZfish";
char *geneIdCol = "ensGeneId";
/* get Gene Symbol and RefSeq accession from Zebrafish-specific */
/* cross-reference table */
printGeneSymbol(id, xrefTable, geneIdCol, conn);
refSeqAcc = getRefSeqAcc(id, xrefTable, geneIdCol, conn);
hPrintf("<B>ENSEMBL ID:</B> %s", id);
}
else
{
char query[256];
char *toRefTable = genomeOptionalSetting("knownToRef");
if (toRefTable != NULL && sqlTableExists(conn, toRefTable))
{
safef(query, sizeof(query), "select value from %s where name='%s'", toRefTable,
id);
refSeqAcc = emptyForNull(sqlQuickString(conn, query));
}
if (sqlTableExists(conn, "kgXref"))
{
safef(query, sizeof(query), "select mRNA from kgXref where kgID='%s'", id);
mrnaAcc = emptyForNull(sqlQuickString(conn, query));
}
if (sameWord(genome, "C. elegans"))
hPrintf("<B>WormBase ID:</B> %s<BR>", id);
else
hPrintf("<B>UCSC ID:</B> %s<BR>", id);
}
if (refSeqAcc[0] != 0)
{
hPrintf("<B>RefSeq Accession: </B> <A HREF=\"");
printOurRefseqUrl(stdout, refSeqAcc);
hPrintf("\">%s</A><BR>\n", refSeqAcc);
}
else if (mrnaAcc[0] != 0)
{
safef(condStr, sizeof(condStr), "acc = '%s'", mrnaAcc);
if (sqlGetField(database, "gbCdnaInfo", "acc", condStr) != NULL)
{
hPrintf("<B>Representative RNA: </B> <A HREF=\"");
printOurMrnaUrl(stdout, mrnaAcc);
hPrintf("\">%s</A><BR>\n", mrnaAcc);
}
else
/* do not show URL link if it is not found in gbCdnaInfo */
{
hPrintf("<B>Representative RNA: %s </B>", mrnaAcc);
}
}
if (protAcc != NULL)
{
kgProteinID = cloneString("");
if (hTableExists(sqlGetDatabase(conn), "knownGene")
&& (isNotEmpty(cartOptionalString(cart, hggChrom)) &&
differentWord(cartOptionalString(cart, hggChrom),"none")))
{
safef(condStr, sizeof(condStr), "name = '%s' and chrom = '%s' and txStart=%s and txEnd=%s",
id, cartOptionalString(cart, hggChrom),
cartOptionalString(cart, hggStart),
cartOptionalString(cart, hggEnd));
kgProteinID = sqlGetField(database, "knownGene", "proteinID", condStr);
}
hPrintf("<B>Protein: ");
if (strstr(kgProteinID, "-") != NULL)
{
parAcc = cloneString(kgProteinID);
chp = strstr(parAcc, "-");
*chp = '\0';
/* show variant splice protein and the UniProt link here */
hPrintf("<A HREF=\"http://www.uniprot.org/uniprot%s\" "
"TARGET=_blank>%s</A></B>, splice isoform of ",
kgProteinID, kgProteinID);
hPrintf("<A HREF=\"http://www.uniprot.org/uniprot/%s\" "
"TARGET=_blank>%s</A></B>\n",
parAcc, parAcc);
}
else
{
hPrintf("<A HREF=\"http://www.uniprot.org/uniprot/%s\" "
"TARGET=_blank>%s</A></B>\n",
protAcc, protAcc);
}
/* show SWISS-PROT display ID if it is different than the accession ID */
/* but, if display name is like: Q03399 | Q03399_HUMAN, then don't show display name */
spDisplayId = spAnyAccToId(spConn, protAcc);
if (spDisplayId == NULL)
{
errAbort("<br>%s seems to no longer be a valid protein ID in our latest UniProtKB DB.", protAcc);
}
if (strstr(spDisplayId, protAcc) == NULL)
{
hPrintf(" (aka %s", spDisplayId);
/* show once if the new and old displayId are the same */
oldDisplayId = oldSpDisplayId(spDisplayId);
if (oldDisplayId != NULL)
{
if (!sameWord(spDisplayId, oldDisplayId)
&& !sameWord(protAcc, oldDisplayId))
{
hPrintf(" or %s", oldDisplayId);
}
}
hPrintf(")<BR>\n");
}
}
printCcds(id, conn);
}
static void ctdPrint(struct section *section,
struct sqlConnection *conn, char *geneId)
/* Print out CTD section. */
{
char query[256];
struct sqlResult *sr;
char **row;
char *chemId, *chemName;
int chemCnt;
int first = 1;
boolean showCompleteCtdList;
struct dyString *currentCgiUrl;
showCompleteCtdList = FALSE;
if (cgiOptionalString("showAllCtdRef") != NULL)
{
if (sameWord(cgiOptionalString("showAllCtdRef"), "Y") ||
sameWord(cgiOptionalString("showAllCtdRef"), "y") )
{
showCompleteCtdList = TRUE;
}
cartRemove(cart, "showAllCtdRef");
}
currentCgiUrl = cgiUrlString();
/* List chemicals related to this gene */
if (isRgdGene(conn))
{
sqlSafef(query, sizeof(query),
"select ChemicalId, ChemicalName from rgdGene2Xref x, hgFixed.ctdSorted c where x.info=c.GeneSymbol and rgdGeneId='%s' and infoType='Name'",
geneId);
}
else
{
sqlSafef(query, sizeof(query),
"select ChemicalId, ChemicalName from kgXref x, hgFixed.ctdSorted c where x.geneSymbol=c.GeneSymbol and kgId='%s'",
geneId);
}
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
chemCnt = 0;
while (row != NULL)
{
chemId = cloneString(row[0]);
chemName = cloneString(row[1]);
if (first)
{
printf("<B>The following chemicals interact with this gene</B>\n");
printf("<UL>");
first = 0;
}
printf("<LI><A HREF=\"http://ctd.mdibl.org/detail.go?type=chem&acc=%s\" target=_blank>",
chemId);
printf("%s</A>\n", chemId);
printf("%s\n", chemName);
printf("</LI>");
chemCnt++;
row = sqlNextRow(sr);
/* Initially, just show no more than 10 items */
if ((!showCompleteCtdList) && (chemCnt >= 10) && (row != NULL))
{
chemCnt++;
break;
}
}
sqlFreeResult(&sr);
if (!first) printf("</UL>");
/* Offer user the option to see the complete list */
if ((!showCompleteCtdList) && chemCnt > 10)
{
printf("<B> more ... </B>");
printf("<A HREF=\"%s?showAllCtdRef=Y&%s#ctd\">click here to view the complete list</A> "
,"hgGene", currentCgiUrl->string);
}
return;
}
static void reactomeLink(struct pathwayLink *pl, struct sqlConnection *conn,
char *geneId)
{
char condStr[255];
char *spID, *chp;
struct sqlConnection *conn2;
char query2[256];
struct sqlResult *sr2;
char **row2;
char *eventDesc;
char *eventID;
/* check the existence of kgXref table first */
if (isRgdGene(conn))
{
if (!sqlTableExists(conn, "rgdGene2Xref")) return;
}
else
{
if (!sqlTableExists(conn, "kgXref")) return;
}
if (isRgdGene(conn))
{
sqlSafefFrag(condStr, sizeof(condStr), "name='%s'", geneId);
spID = sqlGetField(database, "rgdGene2ToUniProt", "value", condStr);
}
else
{
sqlSafefFrag(condStr, sizeof(condStr), "kgID='%s'", geneId);
spID = sqlGetField(database, "kgXref", "spID", condStr);
}
if (spID != NULL)
{
/* convert splice variant UniProt ID to its main root ID */
chp = strstr(spID, "-");
if (chp != NULL) *chp = '\0';
hPrintf(
"<BR>Protein %s (<A href=\"http://www.reactome.org/cgi-bin/link?SOURCE=UniProt&ID=%s\" TARGET=_blank>Reactome details)</A> participates in the following event(s):<BR><BR>"
, spID, spID);
conn2= hAllocConn(database);
sqlSafef(query2,sizeof(query2),
"select eventID, eventDesc from proteome.spReactomeEvent where spID='%s'", spID);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
eventID = row2[0];
eventDesc = row2[1];
hPrintf(
"<A href=\"http://www.reactome.org/cgi-bin/eventbrowser?DB=gk_current&ID=%s\" TARGET=_blank>%s</A> %s<BR>\n",
eventID, eventID, eventDesc);
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn2);
}
}
