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);
}
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2;
char query2[256];
struct sqlResult *sr2;
char **row2;
char cond_str[256];
char *kgID;
char *proteinID;
char *seq;
char *acc;
char protDbName[100];
char spDbName[100];
char *dbName;
char *ro_dbName;
FILE *o1, *o2;
struct dnaSeq *kgSeq;
if (argc != 4) usage();
o1 = fopen("j.dat", "w");
o2 = fopen("jj.dat", "w");
dbName = argv[1];
ro_dbName = argv[3];
sprintf(protDbName, "proteins%s", argv[2]);
sprintf(spDbName, "sp%s", argv[2]);
conn= hAllocConn(ro_dbName);
conn2= hAllocConn(ro_dbName);
sqlSafef(query2, sizeof query2, "select name from %s.knownGene;", dbName);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
kgID = row2[0];
sqlSafefFrag(cond_str, sizeof cond_str, "name = '%s';", kgID);
seq = sqlGetField(dbName, "knownGenePep", "seq", cond_str);
if (seq != NULL)
{
fprintf(o1, "%s\t%s\n", kgID, seq);fflush(o1);
}
else
{
sqlSafefFrag(cond_str, sizeof cond_str, "name = '%s';", kgID);
proteinID=sqlGetField(dbName, "knownGene", "proteinID", cond_str);
if (proteinID != NULL)
{
sqlSafefFrag(cond_str, sizeof cond_str, "val = '%s';", proteinID);
acc = sqlGetField(spDbName, "displayId", "acc", cond_str);
if (acc == NULL)
{
fprintf(stderr, "NO acc.displayId.%s: %s from name.knownGene.%s: %s\n", spDbName, proteinID, dbName, kgID);
fflush(stderr);
}
else
{
sqlSafefFrag(cond_str, sizeof cond_str, "acc = '%s';", acc);
seq = sqlGetField(spDbName, "protein", "val", cond_str);
if (seq == NULL)
{
fprintf(stderr, "NO protein seq for %s\n", kgID);
fprintf(stderr, "proteinID.knownGene.%s: %s, acc.displayID.%s: %s\n", dbName, proteinID, spDbName, acc);
fflush(stderr);
}
else
{
fprintf(o1, "%s\t%s\n", kgID, seq);
}
}
} else {
fprintf(stderr, "kgID: %s not in knownGenePep or knownGene\n", kgID);
}
}
sqlSafefFrag(cond_str, sizeof cond_str, "name = '%s';", kgID);
seq = sqlGetField(dbName, "knownGeneMrna", "seq", cond_str);
if (seq != NULL)
{
fprintf(o2, "%s\t%s\n", kgID, seq);fflush(o1);
}
else
{
kgSeq = hGenBankGetMrna(dbName, kgID, NULL);
if (kgSeq != NULL)
{
fprintf(o2, "%s\t%s\n", kgID, kgSeq->dna);fflush(o1);
}
else
{
fprintf(stderr, "NO mRNA seq for %s\n", kgID);fflush(stderr);
}
}
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn);
hFreeConn(&conn2);
fclose(o1);
fclose(o2);
mustSystem("cat j.dat |sort|uniq > knownGenePep.tab");
mustSystem("cat jj.dat|sort|uniq > knownGeneMrna.tab");
mustSystem("rm j.dat");
mustSystem("rm jj.dat");
return(0);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn;
FILE *inf;
FILE *o1;
char cond_str[256];
char *database;
char *proteinFileName;
char *outputFileName;
char *answer;
char *alias;
char *id;
char *chp0, *chp1, *chp2, *chp;
char *kgID;
char line[2000];
if (argc != 4) usage();
database = cloneString(argv[1]);
proteinFileName = cloneString(argv[2]);
outputFileName = cloneString(argv[3]);
conn = hAllocConn(database);
o1 = mustOpen(outputFileName, "w");
if ((inf = mustOpen(proteinFileName, "r")) == NULL)
{
fprintf(stderr, "Can't open file %s.\n", proteinFileName);
exit(8);
}
while (fgets(line, 1000, inf) != NULL)
{
chp = strstr(line, "ID ");
if (chp != line)
{
fprintf(stderr, "expected ID line, but got: %s\n", line);
exit(1);
}
chp = chp + strlen("ID ");
id = chp;
chp = strstr(id, " ");
*chp = '\0';
id = strdup(id);
sqlSafefFrag(cond_str, sizeof cond_str, "proteinID = '%s'", id);
answer = sqlGetField(database, "knownGene", "name", cond_str);
kgID = NULL;
if (answer != NULL)
{
kgID = strdup(answer);
}
if (fgets(line, 1000, inf) == NULL)
{
break;
}
do
{
/* "//" signal end of a record */
if ((line[0] == '/') && (line[1] == '/')) break;
// work on GN (Gene Name) line only
chp = strstr(line, "GN ");
if (chp != NULL)
{
chp = line + strlen(line) -2;
if (*chp == '.')
{
*chp = '\0';
}
else
{
chp++;
*chp = '\0';
}
chp0 = line + 5;
while (chp0 != NULL)
{
while (*chp0 == ' ') chp0++;
chp1 = strstr(chp0, " OR ");
chp2 = strstr(chp0, " AND ");
chp = NULL;
if (chp1 != NULL)
{
if (chp2 != NULL)
{
if (chp1 < chp2)
{
chp = chp1;
}
else
{
chp = chp2;
}
}
else
{
chp = chp1;
}
}
if (chp2!= NULL)
{
if (chp1 != NULL)
{
if (chp1 < chp2)
{
chp = chp1;
}
else
{
chp = chp2;
}
}
else
{
chp = chp2;
}
}
if (chp == NULL)
{
alias = strdup(chp0);
chp0 = NULL;
}
else
{
*chp = '\0';
alias = strdup(chp0);
chp0 = chp+4;
}
if (kgID != NULL)
{
// clean up "(XXXX" or "XXXX)"
if (*alias == '(') alias++;
chp = strstr(alias, ")");
if (chp != NULL) *chp = '\0';
fprintf(o1, "%s\t%s\n", kgID, alias);
}
}
}
} while (fgets(line, 1000, inf) != NULL);
}
fclose(o1);
hFreeConn(&conn);
return(0);
}
