static char *findGeneId(struct sqlConnection *conn, char *name)
/* Given some sort of gene name, see if it is in our primary gene table, and if not
* look it up in alias table if we have one. */
{
/* Just check if it's in the main gene table, and if so return input name. */
char *mainTable = genomeSetting("knownGene");
char query[256];
sqlSafef(query, sizeof(query), "select count(*) from %s where name = '%s'", mainTable, name);
if (sqlQuickNum(conn, query) > 0)
return name;
else
{
/* check OMIM gene symbol table first */
if (sqlTableExists(conn, "omimGeneSymbol"))
{
sqlSafef(query, sizeof(query), "select geneSymbol from omimGeneSymbol where geneSymbol= '%s'", name);
char *symbol = sqlQuickString(conn, query);
if (symbol != NULL)
{
sqlSafef(query, sizeof(query), "select kgId from kgXref where geneSymbol = '%s'", symbol);
char *kgId = sqlQuickString(conn, query);
if (kgId != NULL)
{
/* The canonical gene is preferred */
sqlSafef(query, sizeof(query),
"select c.transcript from knownCanonical c,knownIsoforms i where i.transcript = '%s' and i.clusterId=c.clusterId", kgId);
char *canonicalKgId = sqlQuickString(conn, query);
if (canonicalKgId != NULL)
{
return canonicalKgId;
}
else
return(kgId);
}
}
}
}
char *alias = genomeOptionalSetting("kgAlias");
if (alias != NULL && sqlTableExists(conn, alias))
{
sqlSafef(query, sizeof(query), "select kgID from %s where alias = '%s'", alias, name);
char *id = sqlQuickString(conn, query);
if (id == NULL)
hUserAbort("Couldn't find %s in %s.%s or %s.%s", name, database, mainTable, database, alias);
return id;
}
else
hUserAbort("Couldn't find %s in %s.%s", name, database, mainTable);
return NULL;
}
void botTerminateMessage(char *ip, int millis)
/* Print out message saying why you are terminated. */
{
time_t now = time(NULL);
hUserAbort("There is an exceedingly high volume of traffic coming from your "
"site (IP address %s) as of %s (California time). It looks like "
"a web robot is launching queries quickly, and not even waiting for "
"the results of one query to finish before launching another query. "
"/* We cannot service requests from your IP address under */ these "
"conditions. (code %d)", ip, asctime(localtime(&now)), millis);
}
static void getGenePosition(struct sqlConnection *conn)
/* Get gene position from database. */
{
char *table = genomeSetting("knownGene");
char query[256];
struct sqlResult *sr;
char **row;
sqlSafef(query, sizeof(query),
"select chrom,txStart,txEnd from %s where name = '%s'"
, table, curGeneId);
sr = sqlGetResult(conn, query);
row = sqlNextRow(sr);
if (row != NULL)
{
curGeneChrom = cloneString(row[0]);
curGeneStart = atoi(row[1]);
curGeneEnd = atoi(row[2]);
}
else
hUserAbort("Couldn't find %s in %s.%s", curGeneId, database, table);
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);
}
void doMiddle(struct cart *theCart)
/* Print the body of an html file. */
{
char cond_str[255];
struct sqlConnection *conn;
char *proteinAC;
char *chp, *chp1, *chp9;
char *debugTmp = NULL;
char *chromStr, *cdsStartStr, *cdsEndStr, posStr[255];
char *supportedGenomeDatabase;
char *answer;
char *queryID;
/* Initialize layout and database. */
cart = theCart;
/* Uncomment this to see parameters for debugging. */
/* Be careful though, it breaks if custom track
* is more than 4k */
/*
{ struct dyString *state = cgiUrlString();
hPrintf("State: %s\n", state->string);
}
*/
queryID = cartOptionalString(cart, "proteinID");
if (sameString(queryID, ""))
{
hUserAbort("Please go back and enter a gene symbol or a Swiss-Prot/TrEMBL protein ID.\n");
}
if (cgiVarExists("db"))
{
/* if db is known, get key variables set */
proteinInSupportedGenome = TRUE;
database = cgiOptionalString("db");
organism = hDbOrganism(database);
protDbName = hPdbFromGdb(database);
proteinID = strdup(queryID);
}
else
{
protCntInSwissByGene = searchProteinsInSwissProtByGene(queryID);
/* no CGI 'db' variable means it did not come in from GB but from pbGateway */
/* search existing GB databases to see if this protein can be found */
protCntInSupportedGenomeDb =
searchProteinsInSupportedGenomes(queryID, &supportedGenomeDatabase);
if ((protCntInSupportedGenomeDb > 1) || protCntInSwissByGene >= 1)
{
/* more than 1 proteins match the query ID, present selection web page */
proteinInSupportedGenome = 1;
presentProteinSelections(queryID, protCntInSwissByGene, protCntInSupportedGenomeDb);
return;
}
else
{
if (protCntInSupportedGenomeDb == 1)
{
/* one and only one protein found in a genome DB that support KG and PB */
proteinInSupportedGenome = TRUE;
database = strdup(supportedGenomeDatabase);
organism = hDbOrganism(database);
protDbName = hPdbFromGdb(database);
proteinID=strdup(queryID);
}
else
{
/* not found in genome DBs that support KG/PB */
/* now search PROTEOME_DB_NAMES to see if this protein is there. */
answer = uniProtFindPrimAcc(queryID);
if (answer == NULL)
{
hUserAbort("'%s' does not seem to be a valid UniProtKB protein ID or a gene "
"symbol.<br><br>Click <A HREF=\"../cgi-bin/pbGateway\">here</A> "
"to start another query.", queryID);
}
proteinInSupportedGenome = FALSE;
database = strdup(GLOBAL_PB_DB);
organism = strdup("");
protDbName = strdup(PROTEOME_DB_NAME);
proteinID = strdup(answer);
}
}
if (proteinInSupportedGenome)
{
spConn = sqlConnect(database);
sqlSafefFrag(cond_str, sizeof(cond_str), "alias='%s'", queryID);
proteinID = sqlGetField(database, "kgSpAlias", "spID", cond_str);
sqlSafefFrag(cond_str, sizeof(cond_str), "spID='%s'", proteinID);
answer = sqlGetField(database, "kgXref", "spDisplayID", cond_str);
sqlSafefFrag(cond_str, sizeof(cond_str), "proteinID='%s'", answer);
chromStr = sqlGetField(database, "knownGene", "chrom", cond_str);
if (chromStr)
{
cdsStartStr = sqlGetField(database, "knownGene", "cdsStart", cond_str);
cdsEndStr = sqlGetField( database, "knownGene", "cdsEnd", cond_str);
safef(posStr, sizeof(posStr), "%s:%s-%s", chromStr, cdsStartStr, cdsEndStr);
positionStr = strdup(posStr);
cartSetString(cart, "position", positionStr);
cartSetString(cart, "organism", organism);
}
}
}
/* print out key variables for debugging */
/* printf("<br>before enter main section: <br>proteinInSupportedGenome=%d<br>proteinID=%s <br>database=%s <br>organism=%s <br>protDbName=%s\n",
proteinInSupportedGenome, proteinID, database, organism, protDbName);fflush(stdout);
*/
if (hTableExists(database, "kgProtMap2"))
{
kgVersion = KG_III;
strcpy(kgProtMapTableName, "kgProtMap2");
}
debugTmp = cartUsualString(cart, "hgDebug", "off");
if(sameString(debugTmp, "on"))
hgDebug = TRUE;
else
hgDebug = FALSE;
conn = hAllocConn(database);
hgsid = cartOptionalString(cart, "hgsid");
if (hgsid != NULL)
{
safef(hgsidStr, sizeof(hgsidStr), "&hgsid=%s", hgsid);
}
else
{
strcpy(hgsidStr, "");
}
/* check proteinID to see if it is a valid SWISS-PROT/TrEMBL accession or display ID */
/* then assign the accession number to global variable proteinID */
sqlSafefFrag(cond_str, sizeof(cond_str), "accession='%s'", proteinID);
proteinAC = sqlGetField(protDbName, "spXref3", "accession", cond_str);
if (proteinAC == NULL)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "displayID='%s'", proteinID);
proteinAC = sqlGetField(protDbName, "spXref3", "accession", cond_str);
if (proteinAC == NULL)
{
hUserAbort("'%s' does not seem to be a valid Swiss-Prot/TrEMBL protein ID or gene symbol.<br><br>Click <A HREF=\"../cgi-bin/pbGateway\">here</A> to start another query."
, proteinID);
}
else
{
protDisplayID = proteinID;
proteinID = proteinAC;
}
}
else
{
sqlSafefFrag(cond_str, sizeof(cond_str), "accession='%s'", proteinID);
protDisplayID = sqlGetField(protDbName, "spXref3", "displayID", cond_str);
}
if (proteinInSupportedGenome)
{
if (kgVersion == KG_III)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "spId='%s'", proteinID);
mrnaID = sqlGetField(database, "kgXref", "kgId", cond_str);
}
else
{
sqlSafefFrag(cond_str, sizeof(cond_str), "proteinID='%s'", protDisplayID);
mrnaID = sqlGetField(database, "knownGene", "name", cond_str);
}
}
else
{
mrnaID = NULL;
positionStr = NULL;
}
sqlSafefFrag(cond_str, sizeof(cond_str), "accession='%s'", proteinID);
description = sqlGetField(protDbName, "spXref3", "description", cond_str);
if (positionStr != NULL)
{
chp = strstr(positionStr, ":");
*chp = '\0';
prevGBChrom = cloneString(positionStr);
chp1 = chp + 1;
chp9 = strstr(chp1, "-");
*chp9 = '\0';
prevGBStartPos = atoi(chp1);
chp1 = chp9 + 1;
prevGBEndPos = atoi(chp1);
}
else
{
prevGBChrom = NULL;
prevGBStartPos = -1;
prevGBEndPos = -1;
}
/* Do main display. */
if (cgiVarExists("pbt.psOutput"))
handlePostscript();
else
{
doTrackForm(NULL, NULL);
}
}
void makeActiveImagePB(char *psOutput, char *psOutput2)
/* Make image and image map. */
{
char *mapName = "map";
int pixWidth, pixHeight;
char *answer;
char cond_str[255];
struct sqlConnection *conn;
struct sqlConnection *connCentral;
char query[256];
struct sqlResult *sr;
char **row;
int iypos;
char *blatGbDb;
char *sciName, *commonName;
char *spDisplayId;
char *oldDisplayId;
conn = sqlConnect(UNIPROT_DB_NAME);
hPrintf("<br><font size=4>Protein ");
hPrintf("<A HREF=\"http://www.uniprot.org/uniprot/%s\" TARGET=_blank><B>%s</B></A>\n",
proteinID, proteinID);
spDisplayId = spAccToId(conn, spFindAcc(conn, proteinID));
if (strstr(spDisplayId, spFindAcc(conn, proteinID)) == 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))
{
hPrintf(" or %s", oldSpDisplayId(spDisplayId));
}
}
hPrintf(")\n");
}
hPrintf(" %s\n", description);
hPrintf("</font><br>");
hPrintf("Organism: ");
/* get scientific and Genbank common name of this organism */
sciName = NULL;
commonName = NULL;
sqlSafefFrag(cond_str, sizeof(cond_str),"accession='%s'", proteinID);
answer = sqlGetField(PROTEOME_DB_NAME, "spXref3", "division", cond_str);
if (answer != NULL)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "id=%s and nameType='scientific name'", answer);
sciName = sqlGetField(PROTEOME_DB_NAME, "taxonNames", "name", cond_str);
sqlSafefFrag(cond_str, sizeof(cond_str), "id=%s and nameType='genbank common name'", answer);
commonName = sqlGetField(PROTEOME_DB_NAME, "taxonNames", "name", cond_str);
}
if (sciName != NULL)
{
hPrintf("%s", sciName);
}
if (commonName != NULL)
{
hPrintf(" (%s)", commonName);
}
hPrintf("<br>");
protSeq = getAA(proteinID);
if (protSeq == NULL)
{
hUserAbort("%s is not a current valid entry in UniProtKB\n", proteinID);
}
protSeqLen = strlen(protSeq);
fflush(stdout);
iypos = 15;
doTracks(proteinID, mrnaID, protSeq, &iypos, psOutput);
if (!hTableExists(database, "pbStamp")) goto histDone;
pbScale = 3;
pixWidth = 765;
insideWidth = pixWidth-gfxBorder;
pixHeight = 350;
if (psOutput2)
{
vg2 = vgOpenPostScript(pixWidth, pixHeight, psOutput2);
}
else
{
trashDirFile(&gifTn2, "pbt", "pbt", ".png");
vg2 = vgOpenPng(pixWidth, pixHeight, gifTn2.forCgi, FALSE);
}
g_vg = vg2;
pbRed = vgFindColorIx(vg2, 0xf9, 0x51, 0x59);
pbBlue = vgFindColorIx(g_vg, 0x00, 0x00, 0xd0);
normalColor = pbBlue;
abnormalColor = pbRed;
bkgColor = vgFindColorIx(vg2, 255, 254, 232);
vgBox(vg2, 0, 0, insideWidth, pixHeight, bkgColor);
/* Start up client side map. */
mapName=cloneString("pbStamps");
hPrintf("\n<MAP Name=%s>\n", mapName);
vgSetClip(vg2, 0, gfxBorder, insideWidth, pixHeight - 2*gfxBorder);
iypos = 15;
/* Draw stamps. */
doStamps(proteinID, mrnaID, protSeq, vg2, &iypos);
/* Finish map. */
hPrintf("</MAP>\n");
/* Save out picture and tell html file about it. */
vgClose(&vg2);
hPrintf("<P>");
hPrintf("\n<IMG SRC=\"%s\" BORDER=1 WIDTH=%d HEIGHT=%d USEMAP=#%s><BR>",
gifTn2.forCgi, pixWidth, pixHeight, mapName);
if (proteinInSupportedGenome)
{
hPrintf("\n<A HREF=\"../goldenPath/help/pbTracksHelpFiles/pbTracksHelp.shtml#histograms\" TARGET=_blank>");
}
else
{
hPrintf("\n<A HREF=\"../goldenPath/help/pbTracksHelpFiles/pbTracksHelp.shtml#histograms\" TARGET=_blank>");
}
hPrintf("Explanation of Protein Property Histograms</A><BR>");
hPrintf("<P>");
histDone:
hPrintf("<P>");
fflush(stdout);
/* See if a UCSC Genome Browser exist for this organism. If so, display BLAT link. */
connCentral = hConnectCentral();
sqlSafef(query, sizeof(query),
"select defaultDb.name from dbDb, defaultDb where dbDb.scientificName='%s' and dbDb.name=defaultDb.name",
sciName);
sr = sqlGetResult(connCentral, query);
row = sqlNextRow(sr);
if (row != NULL)
{
blatGbDb = strdup(row[0]);
}
else
{
blatGbDb = NULL;
}
sqlFreeResult(&sr);
hDisconnectCentral(&connCentral);
if (proteinInSupportedGenome || (blatGbDb != NULL))
{
hPrintf("\n<B>UCSC Links:</B><BR>\n ");
hPrintf("<UL>\n");
/* Show GB links only if the protein belongs to a supported genome */
if (proteinInSupportedGenome)
{
doGenomeBrowserLink(proteinID, mrnaID, hgsidStr);
doGeneDetailsLink(proteinID, mrnaID, hgsidStr);
}
/* Show Gene Sorter link only if it is valid for this genome */
if (hgNearOk(database))
{
doGeneSorterLink(protDisplayID, mrnaID, hgsidStr);
}
/* Show BLAT link if we have UCSC Genome Browser for it */
if (blatGbDb != NULL)
{
doBlatLink(blatGbDb, sciName, commonName, protSeq);
}
hPrintf("</UL><P>");
}
/* This section shows various types of domains */
conn = sqlConnect(UNIPROT_DB_NAME);
domainsPrint(conn, proteinID);
hPrintf("<P>");
/* Do Pathway section only if the protein belongs to a supported genome */
if (proteinInSupportedGenome);
{
doPathwayLinks(proteinID, mrnaID);
}
printFASTA(proteinID, protSeq);
}
