char *newSpDisplayId(char *oldSpDisplayId)
/* Convert from old Swiss-Prot display ID to new display ID */
{
static struct sqlConnection *conn=NULL;
char condStr[255];
char *newSpDisplayId;
if (conn==NULL)
{
conn = sqlConnect(PROTEOME_DB_NAME);
if (conn == NULL) return NULL;
}
safef(condStr, sizeof(condStr), "oldDisplayId='%s'", oldSpDisplayId);
newSpDisplayId = sqlGetField(PROTEOME_DB_NAME, "spOldNew", "newDisplayId", condStr);
return(newSpDisplayId);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2;
char query2[256];
struct sqlResult *sr2;
char **row2;
char *chp0, *chp;
char *kgID;
FILE *o1, *o2;
char cond_str[256];
char *database;
char *proteinDB;
boolean doingAlias, bothDone;
char *answer;
char *symbol, *alias, *aliases;
if (argc != 3) usage();
database = cloneString(argv[1]);
proteinDB = cloneString(argv[2]);
conn = hAllocConn(database);
conn2= hAllocConn(database);
o1 = fopen("j.dat", "w");
o2 = fopen("jj.dat", "w");
doingAlias = TRUE;
bothDone = FALSE;
while (!bothDone)
{
if (doingAlias)
{
sqlSafef(query2, sizeof query2, "select symbol, aliases from %s.hgnc;", proteinDB);
}
else
{
sqlSafef(query2, sizeof query2, "select symbol, prvSymbols from %s.hgnc;", proteinDB);
}
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
symbol = row2[0];
aliases = row2[1];
if ( (symbol != NULL) && (strlen(symbol) != 0) )
{
sqlSafefFrag(cond_str, sizeof cond_str, "geneSymbol = '%s'", symbol);
answer = sqlGetField(database, "kgXref", "kgID", cond_str);
if (answer != NULL)
{
kgID = strdup(answer);
fprintf(o2, "%s\t%s\n", kgID, symbol);
}
if ( (aliases != NULL) && (strlen(aliases) != 0) && (answer != NULL) )
{
kgID = strdup(answer);
chp0 = aliases;
while (chp0 != NULL)
{
while (*chp0 == ' ') chp0++;
chp = strstr(chp0, ",");
if (chp == NULL)
{
alias = strdup(chp0);
/* get rid of quote character in some aliases */
if (*alias == '"')
{
*(alias + strlen(alias) - 1) = '\0';
alias++;
printf("%s\n", alias);fflush(stdout);
}
chp0 = NULL;
}
else
{
*chp = '\0';
/* get rid of quote character in some aliases */
if (*chp0 == '"')
{
*(chp0 + strlen(chp0) - 1) = '\0';
chp0++;
printf("%s\n", chp0);fflush(stdout);
}
alias = strdup(chp0);
chp0 = chp+1;
}
if (kgID != NULL)
{
fprintf(o1, "%s\t%s\t%s\n", kgID, symbol, alias);
fprintf(o2, "%s\t%s\n", kgID, alias);
}
}
}
}
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
if (doingAlias)
{
doingAlias = FALSE;
}
else
{
bothDone = TRUE;
}
}
fclose(o1);
fclose(o2);
/* geneAlias.tab has 3 columns, the 2nd is HUGO.symbol
and 3rd contains aliases and withdraws */
mustSystem("cat j.dat|sort|uniq >geneAlias.tab");
/* kgAliasM.tab has 2 columns, all entries from HUGO.symbol, HUGO.aliass,
and HUGO.withdraws are listed in the 2nd column. */
mustSystem("cat jj.dat|sort|uniq >kgAliasM.tab");
mustSystem("rm j.dat");
mustSystem("rm jj.dat");
return(0);
}
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);
}
}
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);
}
int getSuperfamilies(char *proteinID)
/* preserved here for previous older genomes.
Newer genomes should be using getSuperfamilies2(). 6/16/04 Fan*/
{
struct sqlConnection *conn, *conn2;
char query[MAXNAMELEN];
struct sqlResult *sr;
char **row;
char cond_str[255];
char *genomeID, *seqID, *modelID, *eValue, *sfID, *sfDesc;
char *region;
int done;
char *ensPep;
char *transcriptName;
char *chp, *chp2;
int ii = 0;
int int_start, int_end;
if (!hTableExists(database, "sfAssign")) return(0);
conn = hAllocConn(database);
conn2 = hAllocConn(database);
if (hTableExists(database, "ensemblXref3"))
{
/* use ensemblXref3 for Ensembl data release after ensembl34d */
sqlSafefFrag(cond_str, sizeof(cond_str), "tremblAcc='%s'", proteinID);
ensPep = sqlGetField(database, "ensemblXref3", "protein", cond_str);
if (ensPep == NULL)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "swissAcc='%s'", proteinID);
ensPep = sqlGetField(database, "ensemblXref3", "protein", cond_str);
if (ensPep == NULL) return(0);
}
}
else
{
if (! (hTableExists(database, "ensemblXref") || hTableExists(database, "ensTranscript") ) )
return(0);
/* two steps query needed because the recent Ensembl gene_xref 11/2003 table does not have
valid translation_name */
sqlSafefFrag(cond_str, sizeof(cond_str), "external_name='%s'", protDisplayID);
transcriptName = sqlGetField(database, "ensGeneXref", "transcript_name", cond_str);
if (transcriptName == NULL)
{
return(0);
}
else
{
sqlSafefFrag(cond_str, sizeof(cond_str), "transcript_name='%s';", transcriptName);
ensPep = sqlGetField(database, "ensTranscript", "translation_name", cond_str);
if (ensPep == NULL)
{
hFreeConn(&conn);
return(0);
}
}
}
ensPepName = ensPep;
sqlSafef(query, sizeof(query), "select * from %s.sfAssign where seqID='%s' and evalue <= 0.02;", database, ensPep);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
if (row == NULL) return(0);
while (row != NULL)
{
genomeID = row[0];
seqID = row[1];
modelID = row[2];
region = row[3];
eValue = row[4];
sfID = row[5];
/* sfDesc = row[6]; */
/* !!! the recent Suprefamily sfAssign table does not have valid sf description */
sqlSafefFrag(cond_str, sizeof(cond_str), "id=%s;", sfID);
sfDesc = sqlGetField(database, "sfDes", "description", cond_str);
/* !!! refine logic here later to be defensive against illegal syntax */
chp = region;
done = 0;
while (!done)
{
chp2 = strstr(chp, "-");
*chp2 = '\0';
chp2++;
sscanf(chp, "%d", &int_start);
chp = chp2;
chp2 = strstr(chp, ",");
if (chp2 != NULL)
{
*chp2 = '\0';
}
else
{
done = 1;
}
chp2++;
sscanf(chp, "%d", &int_end);
sfId[ii] = atoi(sfID);
sfStart[ii] = int_start;
sfEnd[ii] = int_end;
strncpy(superfam_name[ii], sfDesc, MAXNAMELEN-1);
ii++;
chp = chp2;
}
row = sqlNextRow(sr);
}
sqlFreeResult(&sr);
hFreeConn(&conn);
hFreeConn(&conn2);
return(ii);
}
void doAnomalies(char *aa, int len, int *yOffp)
/* draw the AA Anomalies track */
{
char res;
int index;
char cond_str[255];
char *answer;
int xx, yy;
int i, j;
char *chp;
int aaResCnt[20];
double aaResFreqDouble[20];
int abnormal;
int ia = -1;
double pctLow[20], pctHi[20];
/* count frequency for each residue for current protein */
chp = aa;
for (j=0; j<20; j++)
{
aaResCnt[j] = 0;
/* get cutoff threshold value pairs */
sqlSafefFrag(cond_str, sizeof(cond_str), "AA='%c'", aaAlphabet[j]);
answer = sqlGetField(database, "pbAnomLimit", "pctLow", cond_str);
pctLow[j] = (double)(atof(answer));
answer = sqlGetField(database, "pbAnomLimit", "pctHi", cond_str);
pctHi[j] = (double)(atof(answer));
}
for (i=0; i<len; i++)
{
for (j=0; j<20; j++)
{
if (*chp == aaChar[j])
{
aaResCnt[j] ++;
break;
}
}
chp++;
}
for (j=0; j<20; j++)
{
aaResFreqDouble[j] = ((double)aaResCnt[j])/((double)len);
}
currentYoffset = *yOffp;
for (index=0; index < len; index++)
{
res = aa[index];
ia = -1;
for (j=0; j<20; j++)
{
if (res == aaChar[j])
{
ia = j;
break;
}
}
/* skip non-standard AA alphabets */
if (ia == -1) continue;
calxy(index, *yOffp, &xx, &yy);
abnormalColor = pbRed;
abnormal = chkAnomaly(aaResFreqDouble[ia], pctLow[ia], pctHi[ia]);
if (abnormal > 0)
{
vgBox(g_vg, xx, yy-5, 1*pbScale, 5, abnormalColor);
}
else
{
if (abnormal < 0)
{
vgBox(g_vg, xx, yy, 1*pbScale, 5, abnormalColor);
}
}
vgBox(g_vg, xx, yy, 1*pbScale, 1, MG_BLACK);
}
calxy0(0, *yOffp, &xx, &yy);
vgBox(g_vg, 0, yy-10, xx, 20, bkgColor);
trackTitle = cloneString("AA Anomalies");
vgTextRight(g_vg, xx-25, yy-4, 10, 10, MG_BLACK, g_font, trackTitle);
trackTitleLen = strlen(trackTitle);
mapBoxTrackTitle(xx-25-trackTitleLen*6, yy-6, trackTitleLen*6+12, 14, trackTitle, "pepAnom");
/* update y offset */
*yOffp = *yOffp + 15;
}
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);
}
static void gadPrint(struct section *section,
struct sqlConnection *conn, char *geneId)
/* Print out GAD section. */
{
int refPrinted = 0;
boolean showCompleteGadList;
char condStr[256];
char query[256];
struct sqlResult *sr;
char **row;
struct dyString *currentCgiUrl;
char *upperDisease;
char *url =
cloneString("http://geneticassociationdb.nih.gov/cgi-bin/tableview.cgi?table=allview&cond=gene=");
char *itemName;
if (url != NULL && url[0] != 0)
{
safef(condStr, sizeof(condStr),
"k.kgId='%s' and k.geneSymbol = g.geneSymbol", geneId);
itemName = sqlGetField(database, "kgXref k, gadAll g", "k.geneSymbol", condStr);
showCompleteGadList = FALSE;
if (cgiOptionalString("showAllRef") != NULL)
{
if (sameWord(cgiOptionalString("showAllRef"), "Y") ||
sameWord(cgiOptionalString("showAllRef"), "y") )
{
showCompleteGadList = TRUE;
}
}
currentCgiUrl = cgiUrlString();
printf("<B>Genetic Association Database: ");
printf("<A HREF=\"%s'%s'\" target=_blank>", url, itemName);
printf("%s</B></A>\n", itemName);
printf("<BR><B>CDC HuGE Published Literature: ");
printf("<A HREF=\"%s%s%s\" target=_blank>",
"http://hugenavigator.net/HuGENavigator/searchSummary.do?firstQuery=",
itemName,
"&publitSearchType=now&whichContinue=firststart&check=n&dbType=publit&Mysubmit=go");
printf("%s</B></A>\n", itemName);
/* List diseases associated with the gene */
safef(query, sizeof(query),
"select distinct broadPhen from gadAll where geneSymbol='%s' and association = 'Y' order by broadPhen",
itemName);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
if (row != NULL)
{
upperDisease = replaceChars(row[0], "'", "''");
touppers(upperDisease);
printf("<BR><B>Positive Disease Associations: </B>");
printf("<A HREF=\"%s%s%s%s%s\" target=_blank>",
"http://geneticassociationdb.nih.gov/cgi-bin/tableview.cgi?table=allview&cond=upper(DISEASE)%20like%20'%25",
cgiEncode(upperDisease), "%25'%20AND%20upper(GENE)%20%20like%20'%25", itemName, "%25'");
printf("%s</B></A>\n", row[0]);
row = sqlNextRow(sr);
}
while (row != NULL)
{
upperDisease = replaceChars(row[0], "'", "''");
touppers(upperDisease);
printf(", <A HREF=\"%s%s%s%s%s\" target=_blank>",
"http://geneticassociationdb.nih.gov/cgi-bin/tableview.cgi?table=allview&cond=upper(DISEASE)%20like%20'%25",
cgiEncode(upperDisease), "%25'%20AND%20upper(GENE)%20%20like%20'%25", itemName, "%25'");
printf("%s</B></A>\n", row[0]);
row = sqlNextRow(sr);
}
sqlFreeResult(&sr);
refPrinted = 0;
safef(query, sizeof(query),
"select broadPhen,reference,title,journal, pubMed, conclusion from gadAll where geneSymbol='%s' and association = 'Y' order by broadPhen",
itemName);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
if (row != NULL) printf("<BR><B>Related Studies: </B><OL>");
while (row != NULL)
{
printf("<LI><B>%s </B>", row[0]);
printf("<br>%s, %s, %s.\n", row[1], row[2], row[3]);
if (!sameWord(row[4], ""))
{
printf(" [PubMed ");
printf("<A HREF=\"%s%s%s'\" target=_blank>",
"http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstract&list_uids=",
row[4],"&query_hl=1&itool=genome.ucsc.edu");
printf("%s</B></A>]\n", row[4]);
}
printf("<br><i>%s</i>\n", row[5]);
printf("</LI>\n");
refPrinted++;
if ((!showCompleteGadList) && (refPrinted >= 3)) break;
row = sqlNextRow(sr);
}
sqlFreeResult(&sr);
printf("</OL>");
if ((!showCompleteGadList) && (row != NULL))
{
printf("<B>   more ... </B>");
printf(
"<A HREF=\"%s?showAllRef=Y&%s#gad\">click here to view the complete list</A> ",
"hgGene", currentCgiUrl->string);
}
}
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn2, *conn3, *conn4;
char query2[256], query3[256];
struct sqlResult *sr2, *sr3;
char **row2, **row3;
char *accession;
char *extDB;
char *extAC;
char condStr[255];
char *id, *subId, *avStr, *pos;
char *baseAAStr, *subsAAStr;
char baseAA, subsAA;
char *genomeDb;
char *aaSeq;
char ch;
int aaPos, aaLen;
int nTotal = 0;
int nOK = 0;
int nBase = 0;
int nErr = 0;
int nSubs = 0;
boolean gotAMatch = FALSE;
FILE *outf;
if (argc != 3) usage();
genomeDb = argv[1];
outf = fopen(argv[2], "w");
conn2= hAllocConn();
conn3= hAllocConn();
conn4= hAllocConn();
/* loop thru all recordd in the omimAvPos table */
sqlSafef(query2, sizeof query2, "select * from %s.omimAvPos", genomeDb);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
id = row2[0];
subId = row2[1];
avStr = row2[2];
pos = row2[3];
baseAAStr = row2[4];
subsAAStr = row2[5];
baseAA = *baseAAStr;
subsAA = *subsAAStr;
aaPos = atoi(pos);
/* find corresponding protein for each OMIM record */
sqlSafef(query3, sizeof query3,
"select distinct accession, extDB, extAC from %s.spXref2 where extAC='%s' and extDB='MIM';",
PROTEOME_DB_NAME, id);
sr3 = sqlMustGetResult(conn3, query3);
row3 = sqlNextRow(sr3);
while (row3 != NULL)
{
accession = row3[0];
extDB = row3[1];
extAC = row3[2];
nTotal++;
gotAMatch = FALSE;
/* get protein sequence */
sqlSafefFrag(condStr, sizeof condStr, "acc='%s'", accession);
aaSeq = sqlGetField(UNIPROT_DB_NAME, "protein", "val", condStr);
aaLen = strlen(aaSeq);
/* check AA (both base and substitition) of the AV entry against
AA in the protein sequence */
if (aaPos <= aaLen)
{
ch = *(aaSeq+aaPos-1);
if (ch == baseAA)
{
gotAMatch = TRUE;
nOK++;
nBase++;
}
else
{
if (ch == subsAA)
{
gotAMatch = TRUE;
nOK++;
nSubs++;
}
}
if (gotAMatch)
{
fprintf(outf, "%s\t%s\t%s\t%s\n", id, subId, accession, pos);
}
else
{
nErr++;
}
}
else
{
nErr++;
}
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn2);
hFreeConn(&conn3);
fclose(outf);
fprintf(stderr, "nTotal\t= %6d\n", nTotal);
fprintf(stderr, "nOk\t= %6d\n", nOK);
fprintf(stderr, "nBase\t= %6d\n", nBase);
fprintf(stderr, "nSub\t= %6d\n", nSubs);
fprintf(stderr, "nErr\t= %6d\n", nErr);
return(0);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2, *conn3, *conn5;
char query2[256], query3[256], query5[256];
struct sqlResult *sr2, *sr3, *sr5;
char **row2, **row3, **row5;
char cond_str[512];
char *chp;
FILE *o1, *o2;
boolean hasKGmRNA;
char *proteinDisplayID;
char *gbAC;
char *locusID; /* LocusLink ID */
char *refAC; /* Refseq accession.version */
char *giNCBI2; /* NCBI gi for the protein record associated with the CDS */
char *revStatus; /* review status */
char *proteinAC2; /* protein accession.version */
char *taxID2; /* tax id */
char *protDbName;
char *refSeq;
char *name, *chrom, *strand, *txStart, *txEnd, *cdsStart, *cdsEnd,
*exonCount, *exonStarts, *exonEnds;
char *gseq, *hseq, *swissprot;
int alignmentID=0;
if (argc != 4) usage();
dbName = argv[1];
protDbName = argv[2];
genomeReadOnly = argv[3];
sprintf(tempDbName, "%sTemp", dbName);
hSetDb(genomeReadOnly);
conn = hAllocConn();
conn2= hAllocConn();
conn3= hAllocConn();
conn5= hAllocConn();
o1 = fopen("dnaGene.tab", "w");
o2 = fopen("j.dat", "w");
// scan all RefSeq entries
sqlSafef(query2, sizeof query2, "select * from %s.locus2Ref0;", tempDbName);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
locusID = row2[0];
refAC = row2[1];
giNCBI2 = row2[2];
revStatus = row2[3];
proteinAC2 = row2[4];
taxID2 = row2[5];
refSeq = strdup(refAC);
chp = strstr(refAC, ".");
if (chp != NULL) *chp = '\0';
proteinDisplayID = NULL;
/* check if the locusID of this RefSeq points to a KG mRNA */
hasKGmRNA = checkMrna(locusID);
/* check if this RefSeq has 'g' type sequence(s) referenced */
sqlSafefFrag(cond_str, sizeof cond_str, "locusID=%s and seqType='g';", locusID);
gseq = sqlGetField(tempDbName, "locus2Acc0", "gbac", cond_str);
/* process only 'g' type record which does not have corresponding KG entry */
if ((!hasKGmRNA) && (gseq != NULL))
{
sqlSafefFrag(cond_str, sizeof cond_str, "name='%s'", refAC);
hseq = sqlGetField(genomeReadOnly, "refGene", "name", cond_str);
if (hseq != NULL)
{
sqlSafefFrag(cond_str, sizeof cond_str, "refseq='%s';", refAC);
swissprot = sqlGetField(protDbName, "hugo", "swissprot", cond_str);
if (swissprot != NULL)
{
if (strlen(swissprot) >0)
{
// HUGO has an entry with swissprot ID, get display ID
sqlSafefFrag(cond_str, sizeof cond_str, "accession='%s';", swissprot);
proteinDisplayID = sqlGetField(protDbName,
"spXref2", "displayID", cond_str);
if (proteinDisplayID == NULL)
{
fprintf(stderr, "%s: a HUGO.swissprot, ", swissprot);
fprintf(stderr, "but not a SP Primary AC.\n");
fflush(stdout);
}
}
else
{
//printf("HGNC has a non-NULL but empty swissprot field ");
//printf("for %s\n", refAC);fflush(stdout);
}
}
// not finding it in HUGO does not mean not a valid one for sure
if (proteinDisplayID == NULL)
{
// get gbAC and check if spXref2 actually has it
sqlSafef(query3, sizeof query3, "select gbAC from %s.locus2Acc0 where locusID=%s;",
tempDbName, locusID);
sr3 = sqlMustGetResult(conn3, query3);
row3 = sqlNextRow(sr3);
while (row3 != NULL)
{
gbAC = row3[0];
chp = strstr(gbAC, ".");
if (chp != NULL) *chp = '\0';
sqlSafefFrag(cond_str, sizeof cond_str, "extAC='%s'", gbAC);
proteinDisplayID = sqlGetField(protDbName,
"spXref2", "displayID", cond_str);
if (proteinDisplayID == NULL)
{
//printf("%s %s is in refGene, but has no SWISS-PROT.\n",
// locusID, refAC);
//fflush(stdout);
}
else
{
//printf("%s %s got 2nd chance.\n", refAC, gbAC);fflush(stdout);
break;
}
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
}
if (proteinDisplayID != NULL)
{
// generate KG entry
sqlSafef(query5, sizeof query5, "select * from %s.refGene where name='%s';", genomeReadOnly, refAC);
sr5 = sqlMustGetResult(conn5, query5);
row5 = sqlNextRow(sr5);
while (row5 != NULL)
{
name = row5[0];
chrom = row5[1];
strand = row5[2];
txStart = row5[3];
txEnd = row5[4];
cdsStart = row5[5];
cdsEnd = row5[6];
exonCount = row5[7];
exonStarts = row5[8];
exonEnds = row5[9];
fprintf(o1, "%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\tdna%d\n",
name,
chrom,
strand,
txStart,
txEnd,
cdsStart,
cdsEnd,
exonCount,
exonStarts,
exonEnds,
proteinDisplayID,
alignmentID);
alignmentID++;
fprintf(o2, "%s\t%c\t%s\n", name, 'g', proteinAC2);
row5 = sqlNextRow(sr5);
}
sqlFreeResult(&sr5);
}
}
}
row2 = sqlNextRow(sr2);
}
fclose(o1);
fclose(o2);
sqlFreeResult(&sr2);
hFreeConn(&conn);
hFreeConn(&conn2);
hFreeConn(&conn5);
system("sort j.dat|uniq >dnaLink.tab");
system("rm j.dat");
return(0);
}
void doStamps(char *proteinID, char *mrnaID, char *aa, struct vGfx *vg, int *yOffp)
/* draw proteome browser stamps */
{
int i,j,l;
char cond_str[200];
char *valStr;
char valStr2[50];
char *answer;
double pI=0.0;
double exonCount;
char *chp;
int len;
int cCnt;
int xPosition;
int yPosition;
int stampWidth, stampHeight;
int aaResCnt[30];
double aaResFreqDouble[30];
int aaResFound;
int totalResCnt;
double molWeight=0.0;
double hydroSum;
struct pbStamp *stampDataPtr;
for (j=0; j<23; j++)
{
aaResCnt[j] = 0;
}
l=len = strlen(aa);
chp = aa;
for (i=0; i<l; i++)
{
aaResFound = 0;
for (j=0; j<23; j++)
{
if (*chp == aaAlphabet[j])
{
aaResFound = 1;
aaResCnt[j] ++;
}
}
chp++;
}
totalResCnt = 0;
for (i=0; i<23; i++)
{
totalResCnt = totalResCnt + aaResCnt[i];
}
for (i=0; i<20; i++)
{
aaResFreqDouble[i] = ((double)aaResCnt[i])/((double)totalResCnt);
}
AllocVar(stampPictPtr);
stampWidth = 75*(1+pbScale/3);
stampHeight = 60*(1+pbScale/3);
xPosition = 15;
yPosition = *yOffp + 135;
if (pbScale >= 6) yPosition = yPosition + 20;
boundaryColor = vgFindColorIx(g_vg, 170, 170, 170);
/* draw pI stamp */
safef(cond_str, sizeof(cond_str), "accession='%s'", proteinID);
answer = sqlGetField(database, "pepPi", "count(*)", cond_str);
/* either 0 or multiple rows are not valid */
if (strcmp(answer, "1") == 0)
{
answer = sqlGetField(database, "pepPi", "pI", cond_str);
pI = (double)atof(answer);
stampDataPtr = getStampData("pepPi");
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScaleNonInt(stampDataPtr, stampPictPtr, 2);
safef(valStr2, sizeof(valStr2), "%.1f", pI);
markStamp(stampDataPtr, stampPictPtr, pI, valStr2, tx, ty);
pbStampFree(&stampDataPtr);
}
else
{
stampDataPtr = getStampData("pepPi");
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScale(stampDataPtr, stampPictPtr, 2);
safef(valStr2, sizeof(valStr2), "N/A");
markStamp0(stampDataPtr, stampPictPtr, pI, valStr2, tx, ty);
pbStampFree(&stampDataPtr);
}
/* skip Mol Wt, if it is GSID */
if (!hIsGsidServer())
{
/* draw Mol Wt stamp */
safef(cond_str, sizeof(cond_str), "accession='%s'", proteinID);
answer = sqlGetField(database, "pepMwAa", "MolWeight", cond_str);
if (answer != NULL)
{
safef(valStr2, sizeof(valStr2), "%s Da", answer);
molWeight = (double)atof(answer);
stampDataPtr = getStampData("pepMolWt");
xPosition = xPosition + stampWidth + stampWidth/8;
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScaleMW(stampDataPtr, stampPictPtr, 50000);
markStamp(stampDataPtr, stampPictPtr, molWeight, valStr2, tx, ty);
pbStampFree(&stampDataPtr);
}
else
{
safef(valStr2, sizeof(valStr2), "N/A");
stampDataPtr = getStampData("pepMolWt");
xPosition = xPosition + stampWidth + stampWidth/8;
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScaleMW(stampDataPtr, stampPictPtr, 50000);
markStamp0(stampDataPtr, stampPictPtr, molWeight, valStr2, tx, ty);
pbStampFree(&stampDataPtr);
}
}
if (!proteinInSupportedGenome)
{
if (!hIsGsidServer())
xPosition = xPosition + stampWidth + stampWidth/8;
goto skip_exon;
}
/* draw exon count stamp */
if (kgVersion == KG_III)
{
safef(cond_str, sizeof(cond_str), "qName='%s'", mrnaID);
}
else
{
safef(cond_str, sizeof(cond_str), "qName='%s'", proteinID);
}
answer = sqlGetField(database, kgProtMapTableName, "blockCount", cond_str);
if (answer != NULL)
{
valStr = cloneString(answer);
exonCount = (double)atoi(answer);
stampDataPtr = getStampData("exonCnt");
xPosition = xPosition + stampWidth + stampWidth/8;
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScale(stampDataPtr, stampPictPtr, 5);
markStamp(stampDataPtr, stampPictPtr, exonCount, valStr, tx, ty);
pbStampFree(&stampDataPtr);
}
skip_exon:
if (!hIsGsidServer())
{
/* draw AA residual anomolies stamp */
if (answer != NULL)
{
stampDataPtr = getStampData("pepRes");
xPosition = xPosition + stampWidth + stampWidth/8;
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition,
3*stampWidth/2, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
for (i=0; i<20; i++)
{
markResStamp(aaAlphabet[i], stampDataPtr, stampPictPtr, i, aaResFreqDouble[i],
tx, ty, avg, stddev);
}
pbStampFree(&stampDataPtr);
}
xPosition = 15;
yPosition = yPosition + 170;
}
/* skip swInterPro if it is GSID */
if (!hIsGsidServer())
{
/* draw family size stamp */
safef(cond_str, sizeof(cond_str), "accession='%s'", proteinID);
answer = sqlGetField(protDbName, "swInterPro", "count(*)", cond_str);
if (answer != NULL)
{
valStr = cloneString(answer);
stampDataPtr = getStampData("intPCnt");
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScale(stampDataPtr, stampPictPtr, 1);
markStamp(stampDataPtr, stampPictPtr, (double)(atoi(answer)), valStr, tx, ty);
pbStampFree(&stampDataPtr);
}
else
{
valStr = cloneString("N/A");
stampDataPtr = getStampData("intPCnt");
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScale(stampDataPtr, stampPictPtr, 1);
markStamp0(stampDataPtr, stampPictPtr, (double)(atoi(answer)), valStr, tx, ty);
pbStampFree(&stampDataPtr);
}
}
/* draw hydrophobicity stamp */
chp = protSeq;
hydroSum = 0;
for (i=0; i<protSeqLen; i++)
{
hydroSum = hydroSum + aa_hydro[(int)(*chp)];
chp++;
}
stampDataPtr = getStampData("hydro");
xPosition = xPosition + stampWidth + stampWidth/8;
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScaleHydro(stampDataPtr, stampPictPtr, 1.0);
safef(valStr2, sizeof(valStr2), "%.1f", hydroSum/(double)len);
markStamp(stampDataPtr, stampPictPtr, hydroSum/(double)len, valStr2, tx, ty);
pbStampFree(&stampDataPtr);
/* draw Cystein Count stamp */
chp = protSeq;
cCnt = 0;
for (i=0; i<len; i++)
{
if (*chp == 'C') cCnt ++;
chp++;
}
stampDataPtr = getStampData("cCnt");
xPosition = xPosition + stampWidth + stampWidth/8;
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScale(stampDataPtr, stampPictPtr, 10);
safef(valStr2, sizeof(valStr2), "%d", cCnt);
markStamp(stampDataPtr, stampPictPtr, (double)cCnt, valStr2, tx, ty);
pbStampFree(&stampDataPtr);
/* if it is GSID, draw AA residual anomolies here */
if (hIsGsidServer())
{
xPosition = 15;
yPosition = yPosition + 170;
/* draw AA residual anomolies stamp */
if (answer != NULL)
{
stampDataPtr = getStampData("pepRes");
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition,
3*stampWidth/2, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
for (i=0; i<20; i++)
{
markResStamp(aaAlphabet[i], stampDataPtr, stampPictPtr, i, aaResFreqDouble[i],
tx, ty, avg, stddev);
}
pbStampFree(&stampDataPtr);
}
}
/* draw AA residual anomolies stddev stamp */
if (answer != NULL)
{
exonCount = (double)atof(answer);
stampDataPtr = getStampData("pepRes");
if (hIsGsidServer())
{
xPosition = xPosition + stampWidth*1.62 + stampWidth/8;
}
else
{
xPosition = xPosition + stampWidth + stampWidth/8;
}
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, 3*stampWidth/2, stampHeight);
stampDataPtr->ymin = -4.0;
stampDataPtr->ymax = 4.0;
for (i=0; i<20; i++)
{
markResStdvStamp(stampDataPtr, stampPictPtr, i, aaResFreqDouble[i], tx, ty, avg, stddev);
}
/* draw background after bars drawn so that "... stddev" labels do not get covered by bars */
stampDataPtr = getStampData("pepRes");
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, 3*stampWidth/2, stampHeight);
drawPbStampB(stampDataPtr, stampPictPtr);
pbStampFree(&stampDataPtr);
}
/* The follwing section was used to plot freq distribution for each AA so that we can view than to decide on
whether +/- 2 stddev is applicable and what cutoff thresholds to use. Keep it here for possible
future reuse. */
/*
vertLabel = cloneString("Frequency");
for (i=strlen(vertLabel)-1; i>=0; i--)
{
vertLabel[i+1] = '\0';
vgTextCentered(g_vg, 3, 45+i*10, 10, 10, MG_BLACK, g_font, vertLabel+i);
vgTextCentered(g_vg, 3, 215+i*10, 10, 10, MG_BLACK, g_font, vertLabel+i);
}
xPosition = xPosition + 80;
for (j=0; j<20; j++)
{
safef(tempStr, sizeof(tempStr), "%c", aaAlphabet[j]);
stampDataPtr = getStampData(tempStr);
xPosition = xPosition + stampWidth + stampWidth/8;
setPbStampPict(stampPictPtr, stampDataPtr, xPosition, yPosition, stampWidth, stampHeight);
drawPbStamp(stampDataPtr, stampPictPtr);
drawXScale(stampDataPtr, stampPictPtr, 10);
safef(valStr2, sizeof(valStr2), "%c", aaAlphabet[j]);
markStamp(stampDataPtr, stampPictPtr, 0.0, valStr2, tx, ty);
pbStampFree(&stampDataPtr);
}
*/
}
void markResStdvStamp(struct pbStamp *pbStampPtr, struct pbStampPict *stampPictPtr,
int iTarget, double yValueIn, double tx[], double ty[],
double avg[], double stddev[])
/* mark the AA residual stddev stamp */
{
int ix, iy;
double txmin, tymin, txmax, tymax;
double yValue, yPlotValue;
int len;
int xx, yy;
double pctLow, pctHi;
char cond_str[255];
char *answer;
char aaChar;
len = pbStampPtr->len;
txmin = pbStampPtr->xmin;
txmax = pbStampPtr->xmax;
/* force fit for the stddev stamp plot */
tymin = -4.0;
tymax = 4.0;
ix = stampPictPtr->xOrig;
iy = stampPictPtr->yOrig;
aaChar = aaAlphabet[iTarget];
safef(cond_str, sizeof(cond_str), "AA='%c'", aaChar);
answer = sqlGetField(database, "pbAnomLimit", "pctLow", cond_str);
pctLow = (double)(atof(answer));
answer = sqlGetField(database, "pbAnomLimit", "pctHi", cond_str);
pctHi = (double)(atof(answer));
yScale = (double)(120)/8.0;
calStampXY(stampPictPtr, (txmax-txmin)/2.0, tymax, &xx, &yy);
yValue = (yValueIn - avg[iTarget])/stddev[iTarget];
if (yValue > tymax)
{
yPlotValue = tymax;
}
else
{
if (yValue < tymin)
{
yPlotValue = tymin;
}
else
{
yPlotValue = yValue;
}
}
if (yValueIn > pctHi)
{
vLine(tx[iTarget]+0.4, 0.0, yPlotValue, 3, abnormalColor);
}
else
{
if (yValueIn <= pctLow)
{
vLine(tx[iTarget]+0.4, 0.0+yPlotValue, -yPlotValue, 3, abnormalColor);
}
else
{
/* normal range */
if ((yValueIn - avg[iTarget]) >= 0.0)
{
vLine(tx[iTarget]+0.4, 0.0, yPlotValue, 2, normalColor);
}
else
{
vLine(tx[iTarget]+0.4, 0.0+yPlotValue, -yPlotValue, 2, normalColor);
}
}
}
}
int main(int argc, char *argv[])
{
char *skippedKgId;
char *lastValidKgId;
struct sqlConnection *conn2, *conn3;
struct sqlResult *sr2;
char query2[256];
char **row2;
char *proteinID;
FILE *o3, *o7;
char *name, *chrom, *strand, *txStart, *txEnd, *cdsStart, *cdsEnd,
*exonCount, *exonStarts, *exonEnds;
char *alignID;
char *chp;
int i, j;
int isDuplicate;
char *genomeDBname;
char *proteinDataDate;
char proteinsDB[40];
char spDB[40];
char *acc;
#define MAX_EXON 1000
int exStart[MAX_EXON], exEnd[MAX_EXON];
int exCount;
int aaStart[MAX_EXON], aaEnd[MAX_EXON];
char *sp, *ep;
int aalen;
int cdsS, cdsE;
int eS, eE;
if (argc != 3) usage();
proteinDataDate = argv[1];
genomeDBname = argv[2];
safef(spDB, sizeof(spDB), "sp%s", proteinDataDate);
safef(proteinsDB, sizeof(proteinsDB), "proteins%s", proteinDataDate);
o3 = fopen("j.dat", "w");
o7 = fopen("jj.dat", "w");
conn2= hAllocConn(genomeDBname);
conn3= hAllocConn(genomeDBname);
inf = mustOpen("sorted.lis", "r");
strcpy(oldInfo, "");
skippedKgId = cloneString("");
lastValidKgId = cloneString("");
isDuplicate = 0;
oldMrnaStr = cloneString("");
oldAlignStr = cloneString("");
oldProteinStr = cloneString("");
mrnaStr = cloneString("");
proteinStr = cloneString("");
alignStr = cloneString("");
while (fgets(line_in, 10000, inf) != NULL)
{
strcpy(line, line_in);
chp = strstr(line, "\t"); /* chrom */
chp ++;
chp = strstr(chp, "\t"); /* cds block start position */
chp ++;
chp = strstr(chp, "\t"); /* cds block end position */
*chp = '\0';
chp++;
strcpy(newInfo, line);
if (sameString(oldInfo, newInfo))
{
isDuplicate = 1;
}
else
{
/* remember previous record as old only if it is not a duplicate */
if (!isDuplicate)
{
oldMrnaStr = mrnaStr;
oldProteinStr = proteinStr;
oldAlignStr = alignStr;
}
strcpy(oldInfo, newInfo);
isDuplicate = 0;
}
chp = strstr(chp, "\t"); /* priority score */
chp ++;
chp = strstr(chp, "\t"); /* mRNA transcription length */
chp ++;
chp = strstr(chp, "\t"); /* mRNA date */
chp ++;
mrnaStr = chp;
chp = strstr(chp, "\t"); /* mRNA ID */
*chp = '\0';
chp ++;
mrnaStr = cloneString(mrnaStr);
proteinStr = chp;
chp = strstr(chp, "\t"); /* protein ID */
*chp = '\0';
chp ++;
proteinStr = cloneString(proteinStr);
alignID = chp;
/* get rid of "end-of-line" character at the end of the string */
alignStr = trimSpaces(alignID);
if (isDuplicate)
{
/* only put out records for valid KG entries */
if (!sameString(oldMrnaStr, skippedKgId) || sameString(oldMrnaStr, lastValidKgId))
{
fprintf(o7, "%s\t%s\t%s\t%s\n", oldMrnaStr, oldProteinStr, mrnaStr, proteinStr);
}
}
else
{
safef(query2, sizeof(query2), "select * from %sTemp.knownGene0 where alignID='%s';", genomeDBname, alignID);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
name = row2[0];
chrom = row2[1];
strand = row2[2];
txStart = row2[3];
txEnd = row2[4];
cdsStart = row2[5];
cdsEnd = row2[6];
exonCount = row2[7];
exonStarts = row2[8];
exonEnds = row2[9];
proteinID = row2[10];
alignID = row2[11];
sscanf(exonCount, "%d", &exCount);
sp = cloneString(exonStarts);
ep = cloneString(exonEnds);
sscanf(cdsStart, "%d", &cdsS);
sscanf(cdsEnd, "%d", &cdsE);
aalen = 0;
j=0;
for (i=0; i<exCount; i++)
{
chp = strstr(sp, ",");
*chp = '\0';
sscanf(sp, "%d", &(exStart[i]));
chp++;
sp = chp;
chp = strstr(ep, ",");
*chp = '\0';
sscanf(ep, "%d", &(exEnd[i]));
eS = exStart[i];
eE = exEnd[i];
if (cdsS > eS)
{
eS = cdsS;
}
if (cdsE < eE)
{
eE = cdsE;
}
if (eS > eE)
{
eS = 0;
eE = 0;
}
if (eS != eE)
{
aaStart[j] = aalen;
aaEnd[j] = aaStart[j] + (eE- eS +1)/3 -1;
aalen = aalen + (eE- eS +1)/3;
j++;
}
chp++;
ep = chp;
}
cdsLen = aalen;
safef(cond_str, sizeof(cond_str), "val='%s'", proteinID);
acc = sqlGetField(spDB, "displayId", "acc", cond_str);
safef(cond_str, sizeof(cond_str), "acc='%s'", acc);
aaStr=sqlGetField(spDB, "protein", "val", cond_str);
aaLen = strlen(aaStr);
if ((cdsLen > 50) || ((cdsLen * 100)/aaLen > 50))
{
fprintf(o3,"%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n",
name,
chrom,
strand,
txStart,
txEnd,
cdsStart,
cdsEnd,
exonCount,
exonStarts,
exonEnds,
proteinID,
alignID);
lastValidKgId = cloneString(name);
}
else
{
printf("skipping %s %d \n", name, cdsLen);
skippedKgId = cloneString(name);
}
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
}
}
hFreeConn(&conn2);
hFreeConn(&conn3);
fclose(o3);
fclose(o7);
mustSystem("cat j.dat|sort|uniq >knownGene.tab");
mustSystem("cat jj.dat|sort|uniq >duplicate.tab");
mustSystem("rm j.dat");
mustSystem("rm jj.dat");
return(0);
}
void doSamT02(char *proteinId, char *database)
/* display the UCSC SAM-T02 Protein Structure Analysis and Prediction section */
{
char *itemName = NULL;
char query2[256];
struct sqlResult *sr2;
char **row2;
struct sqlConnection *conn, *conn2 = hAllocConn(database);
char condStr[256];
char *chp;
char *samSubDir;
char *samHttpStr0 = NULL; /* SAM server*/
char *samHttpStr = NULL; /* UCSC GB site */
int homologCount;
char *homologID;
char *SCOPdomain;
char *chain;
char *bestEValStr = NULL;
float eValue, bestEVal;
char goodSCOPdomain[40];
int first = 1;
/* return if this genome does not have SAM protein analysis results */
/* defensive logic to guard against the situation that the binary program is pushed, but the data tables are not */
conn = sqlConnect(database);
if (!(sqlTableExists(conn, "samSubdir") && sqlTableExists(conn, "protHomolog")))
{
return;
}
sqlDisconnect(&conn);
if (!sameWord(database, "sacCer1"))
{
return;
}
itemName = proteinId;
if (sameWord(database, "sacCer1"))
{
samHttpStr0 = strdup("http://www.soe.ucsc.edu/research/compbio/yeast-protein-predictions");
samHttpStr = strdup("../goldenPath/sacCer1/sam");
/* SAM analysis of SGD proteins uses SGD ID, not Swiss-Prot AC */
itemName = getSgdId(proteinId, database);
}
if (itemName == NULL) return;
sqlSafefFrag(condStr, sizeof condStr, "proteinId='%s'", itemName);
samSubDir = sqlGetField(database, "samSubdir", "subdir", condStr);
if (samSubDir == NULL) return;
hPrintf("<B>UCSC ");
hPrintf("<A HREF=\"http://www.soe.ucsc.edu/research/compbio/SAM_T02/sam-t02-faq.html\"");
hPrintf(" TARGET=_blank>SAM-T02</A>\n");
hPrintf(" Protein Structure Analysis and Prediction on %s", proteinId);
if (!sameWord(proteinId, itemName)) hPrintf(" (aka %s)", itemName);
hPrintf("</B><BR>\n");
hPrintf(" <B>Multiple Alignment (sequence logo):</B> \n");
hPrintf("<A HREF=\"%s/%s/%s/%s.t2k.w0.5-logo.pdf\"", samHttpStr, samSubDir, itemName, itemName);
hPrintf(" TARGET=_blank>%s</A> (pdf)<BR>\n", itemName);
hPrintf("<B> Secondary Structure Predictions:</B> \n");
hPrintf("<A HREF=\"%s/%s/%s/%s.t2k.dssp-ehl2-logo.pdf\"", samHttpStr, samSubDir, itemName, itemName);
hPrintf(" TARGET=_blank>%s</A> (pdf)<BR>\n", itemName);
hPrintf("<B> Close Homologs:</B> \n");
conn2= hAllocConn(database);
sqlSafef(query2, sizeof query2,
"select homologID,eValue,SCOPdomain,chain from %s.protHomolog where proteinID='%s' and evalue <= 0.01 order by evalue;",
database, itemName);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
homologCount = 0;
strcpy(goodSCOPdomain, "dummy");
bestEVal = 100;
while (row2 != NULL)
{
homologID = row2[0];
sscanf(row2[1], "%e", &eValue);
if (first)
{
bestEVal = eValue;
bestEValStr = strdup(row2[1]);
}
SCOPdomain = row2[2];
chp = SCOPdomain+strlen(SCOPdomain)-1;
while (*chp != '.') chp--;
*chp = '\0';
chain = row2[3];
if (eValue <= 1.0e-10)
{
strcpy(goodSCOPdomain, SCOPdomain);
}
else
{
if (strcmp(goodSCOPdomain,SCOPdomain) != 0)
{
goto skip;
}
else
{
if (eValue > 0.1) goto skip;
}
}
if (first)
{
first = 0;
}
else
{
printf(", ");
}
hPrintf("\n<A HREF=\"http://www.rcsb.org/pdb/cgi/explore.cgi?job=graphics&pdbId=%s", homologID);
if (strlen(chain) >= 1)
{
hPrintf("\"TARGET=_blank>%s</A>(chain %s)\n", homologID, chain);
}
else
{
hPrintf("\"TARGET=_blank>%s</A>\n", homologID);
}
homologCount++;
skip:
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn2);
if (homologCount == 0)
{
hPrintf("None\n");
}
hPrintf("<BR> <B>More Details:</B> \n");
hPrintf("<A HREF=\"%s/%s/%s/summary.html\"", samHttpStr0, samSubDir, itemName);
hPrintf("\" TARGET=_blank>%s</A><BR>\n", itemName);
if (homologCount > 0)
{
hPrintf(" <B>3D Structure Prediction: </B> \n");
hPrintf("<A HREF=\"%s/%s/%s/%s.t2k.undertaker-align.pdb.gz\"",
samHttpStr, samSubDir, itemName, itemName);
hPrintf("\" TARGET=_blank>%s</A> (PDB format, gzipped)<BR>\n", itemName);
hPrintf(" <B>3D Pictures of the Best Model");
hPrintf(" (E Value: %s):</B><BR>\n", bestEValStr);fflush(stdout);
hPrintf("<TABLE><TR>\n");
hPrintf("<TD> </TD>");
hPrintf("<TD><IMG SRC=\"%s/%s/%s/%s.view1_200.jpg\"></A></TD>\n",
samHttpStr, samSubDir, itemName, itemName);
hPrintf("<TD><IMG SRC=\"%s/%s/%s/%s.view2_200.jpg\"></A></TD>\n",
samHttpStr, samSubDir, itemName, itemName);
hPrintf("<TD><IMG SRC=\"%s/%s/%s/%s.view3_200.jpg\"></A></TD>\n",
samHttpStr, samSubDir, itemName, itemName);
hPrintf("</TR>\n");
hPrintf("<TR>");
hPrintf("<TD> </TD>");
hPrintf("<TD ALIGN=CENTER>Front</TD>");
hPrintf("<TD ALIGN=CENTER>Top</TD>");
hPrintf("<TD ALIGN=CENTER>Side</TD>");
hPrintf("</TR>\n");
hPrintf("<TR>");
hPrintf("<TD> </TD>");
hPrintf("<TD ALIGN=CENTER><A HREF=\"%s/%s/%s/%s.view1_500.jpg\">500x500</A></TD>\n",
samHttpStr, samSubDir, itemName, itemName);
hPrintf("<TD ALIGN=CENTER><A HREF=\"%s/%s/%s/%s.view2_500.jpg\">500x500</A></TD>\n",
samHttpStr, samSubDir, itemName, itemName);
hPrintf("<TD ALIGN=CENTER><A HREF=\"%s/%s/%s/%s.view3_500.jpg\">500x500</A></TD>\n",
samHttpStr, samSubDir, itemName, itemName);
hPrintf("</TR>\n");
hPrintf("</TABLE>\n");
}
else
{
hPrintf(" <B>3D Structure Prediction: </B> \n");
hPrintf("No models presented, because none has E-value <= 0.01.<BR>");
}
hPrintf("<BR>");
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2;
char query2[256];
struct sqlResult *sr2;
char **row2;
char cond_str[255];
char *proteinDatabaseName;
FILE *o1, *o2, *o3;
FILE *fh[23];
char temp_str[1000];;
char *accession;
char *aaSeq;
char *chp;
int i, j, len;
int ihi, ilow;
char *answer;
char *protDisplayId;
int aaResCnt[30];
char aaAlphabet[30];
int aaResFound;
float fvalue1, fvalue2;
float p1, p2;
int icnt, jcnt;
char *taxon;
char *database;
int sortedCnt;
if (argc != 4) usage();
strcpy(aaAlphabet, "WCMHYNFIDQKRTVPGEASLXZB");
proteinDatabaseName = argv[1];
taxon = argv[2];
database = argv[3];
o2 = mustOpen("pbResAvgStd.tab", "w");
for (i=0; i<20; i++)
{
safef(temp_str, sizeof(temp_str), "%c.txt", aaAlphabet[i]);
fh[i] = mustOpen(temp_str, "w");
}
conn = hAllocConn(hDefaultDb());
conn2 = hAllocConn(hDefaultDb());
safef(query2, sizeof(query2), "select proteinID from %s.knownGene;", database);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
icnt = 0;
jcnt = 0;
for (j=0; j<MAXRES; j++)
{
sumJ[j] = 0;
}
while (row2 != NULL)
{
protDisplayId = row2[0];
safef(cond_str, sizeof(cond_str), "val='%s'", protDisplayId);
accession = sqlGetField(proteinDatabaseName, "displayId", "acc", cond_str);
if (accession == NULL)
{
safef(cond_str, sizeof(cond_str), "acc='%s'", protDisplayId);
accession = sqlGetField(proteinDatabaseName, "displayId", "acc", cond_str);
if (accession == NULL)
{
verbose(2, "'%s' not found.\n", protDisplayId);
goto skip;
}
}
safef(cond_str, sizeof(cond_str), "accession='%s'", accession);
answer = sqlGetField("proteins040115", "spXref2", "biodatabaseID", cond_str);
if (answer == NULL)
{
/* this protein might be a variant splice protein, and then it won't be in spXref2 */
goto skip;
}
if (answer[0] != '1')
{
/* printf("%s not in SWISS-PROT\n", protDisplayId);fflush(stdout); */
goto skip;
}
safef(cond_str, sizeof(cond_str), "acc='%s'", accession);
aaSeq = sqlGetField(proteinDatabaseName, "protein", "val", cond_str);
if (aaSeq == NULL)
{
printf("Can't find peptide sequence for %s, exiting ...\n", protDisplayId);
fflush(stdout);
exit(1);
}
len = strlen(aaSeq);
if (len < 100) goto skip;
lenDouble = (double)len;
for (j=0; j<MAXRES; j++)
{
aaResCnt[j] = 0;
}
chp = aaSeq;
for (i=0; i<len; i++)
{
aaResFound = 0;
for (j=0; j<MAXRES; j++)
{
if (*chp == aaAlphabet[j])
{
aaResFound = 1;
aaResCnt[j] ++;
}
}
if (!aaResFound)
{
fprintf(stderr, "%c %d not a valid AA residue.\n", *chp, *chp);
}
chp++;
}
for (j=0; j<MAXRES; j++)
{
freq[icnt][j] = (double)aaResCnt[j]/lenDouble;
sumJ[j] = sumJ[j] + freq[icnt][j];
}
for (j=0; j<20; j++)
{
fprintf(fh[j], "%15.7f\t%s\n", freq[icnt][j], accession);
fflush(fh[j]);
}
icnt++;
if (icnt >= MAXN)
errAbort("Too many proteins - please set MAXN to be more than %d\n", MAXN);
skip:
row2 = sqlNextRow(sr2);
}
recordCnt = icnt;
recordCntDouble = (double)recordCnt;
for (j=0; j<20; j++)
{
carefulClose(&(fh[j]));
}
sqlFreeResult(&sr2);
hFreeConn(&conn);
hFreeConn(&conn2);
for (j=0; j<MAXRES; j++)
{
avg[j] = sumJ[j]/recordCntDouble;
}
for (j=0; j<20; j++)
{
sum = 0.0;
for (i=0; i<recordCnt; i++)
{
sum = sum + (freq[i][j] - avg[j]) * (freq[i][j] - avg[j]);
}
sigma[j] = sqrt(sum/(double)(recordCnt-1));
fprintf(o2, "%c\t%f\t%f\n", aaAlphabet[j], avg[j], sigma[j]);
}
carefulClose(&o2);
o1 = mustOpen("pbAnomLimit.tab", "w");
for (j=0; j<20; j++)
{
safef(temp_str, sizeof(temp_str), "cat %c.txt|sort|uniq > %c.srt", aaAlphabet[j], aaAlphabet[j]);
mustSystem(temp_str);
/* figure out how many unique entries */
safef(temp_str, sizeof(temp_str), "wc %c.srt > %c.tmp", aaAlphabet[j], aaAlphabet[j]);
mustSystem(temp_str);
safef(temp_str, sizeof(temp_str), "%c.tmp", aaAlphabet[j]);
o3 = mustOpen(temp_str, "r");
mustGetLine(o3, temp_str, 1000);
chp = temp_str;
while (*chp == ' ') chp++;
while (*chp != ' ') chp++;
*chp = '\0';
sscanf(temp_str, "%d", &sortedCnt);
safef(temp_str, sizeof(temp_str), "rm %c.tmp", aaAlphabet[j]);
mustSystem(temp_str);
/* cal hi and low cutoff threshold */
ilow = (int)((float)sortedCnt * 0.025);
ihi = (int)((float)sortedCnt * 0.975);
safef(temp_str, sizeof(temp_str), "%c.srt", aaAlphabet[j]);
o2 = mustOpen(temp_str, "r");
i=0;
for (i=0; i<ilow; i++)
{
mustGetLine(o2, temp_str, 1000);
}
sscanf(temp_str, "%f", &fvalue1);
mustGetLine(o2, temp_str, 1000);
sscanf(temp_str, "%f", &fvalue2);
p1 = (fvalue1 + fvalue2)/2.0;
for (i=ilow+1; i<ihi; i++)
{
mustGetLine(o2, temp_str, 1000);
}
sscanf(temp_str, "%f", &fvalue1);
mustGetLine(o2, temp_str, 1000);
sscanf(temp_str, "%f", &fvalue2);
p2 = (fvalue1 + fvalue2)/2.0;
carefulClose(&o2);
fprintf(o1, "%c\t%f\t%f\n", aaAlphabet[j], p1, p2);
fflush(stdout);
for (i=0; i<recordCnt; i++)
{
measure[i] = freq[i][j];
}
safef(temp_str, sizeof(temp_str), "pbAaDist%c.tab", aaAlphabet[j]);
calDist(measure, recordCnt, 51, 0.0, 0.005, temp_str);
}
carefulClose(&o1);
return(0);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2;
char query2[256];
struct sqlResult *sr2;
char **row2;
char cond_str[255];
char *proteinDatabaseName; /* example: sp031112 */
char *protDbName; /* example: proteins031112 */
char emptyStr[1] = {""};
FILE *o2;
char *accession;
char *aaSeq;
char *chp;
int i, j, len;
int cCnt;
char *answer, *answer2;
double hydroSum;
char *protDisplayId;
int aaResCnt[30];
double aaResCntDouble[30];
char aaAlphabet[30];
int aaResFound;
int totalResCnt;
int molWtCnt;
double molWt[100000];
int pIcnt;
double pI[100000];
double aa_hydro[256];
int icnt, jExon, pcnt, ipcnt = 0;
double aaLenDouble[100000];
double avgHydro[100000];
double cCountDouble[100000];
double exonCountDouble[100000];
double interProCountDouble[100000];
char *taxon;
char *database;
char *exonCnt;
int interProCount;
char *kgId;
if (argc != 5) usage();
strcpy(aaAlphabet, "WCMHYNFIDQKRTVPGEASLXZB");
/* Ala: 1.800 Arg: -4.500 Asn: -3.500 Asp: -3.500 Cys: 2.500 Gln: -3.500 */
aa_hydro['A'] = 1.800;
aa_hydro['R'] = -4.500;
aa_hydro['N'] = -3.500;
aa_hydro['D'] = -3.500;
aa_hydro['C'] = 2.500;
aa_hydro['Q'] = -3.500;
/* Glu: -3.500 Gly: -0.400 His: -3.200 Ile: 4.500 Leu: 3.800 Lys: -3.900 */
aa_hydro['E'] = -3.500;
aa_hydro['G'] = -0.400;
aa_hydro['H'] = -3.200;
aa_hydro['I'] = 4.500;
aa_hydro['L'] = 3.800;
aa_hydro['K'] = -3.900;
/* Met: 1.900 Phe: 2.800 Pro: -1.600 Ser: -0.800 Thr: -0.700 Trp: -0.900 */
aa_hydro['M'] = 1.900;
aa_hydro['F'] = 2.800;
aa_hydro['P'] = -1.600;
aa_hydro['S'] = -0.800;
aa_hydro['T'] = -0.700;
aa_hydro['W'] = -0.900;
/* Tyr: -1.300 Val: 4.200 Asx: -3.500 Glx: -3.500 Xaa: -0.490 */
aa_hydro['Y'] = -1.300;
aa_hydro['V'] = 4.200;
proteinDatabaseName = argv[1];
protDbName = argv[2];
taxon = argv[3];
database = argv[4];
o2 = mustOpen("pepResDist.tab", "w");
conn = hAllocConn(database);
conn2 = hAllocConn(database);
for (j=0; j<23; j++)
{
aaResCnt[j] = 0;
}
icnt = jExon = pcnt = 0;
pIcnt = 0;
molWtCnt = 0;
sqlSafef(query2, sizeof(query2), "select acc from %s.accToTaxon where taxon=%s;", proteinDatabaseName, taxon);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
accession = row2[0];
sqlSafefFrag(cond_str, sizeof(cond_str), "acc='%s'", accession);
protDisplayId = sqlGetField(proteinDatabaseName, "displayId", "val", cond_str);
sqlSafefFrag(cond_str, sizeof(cond_str), "proteinID='%s'", protDisplayId);
answer = sqlGetField(database, "knownGene", "name", cond_str);
/* count InterPro domains */
if (answer != NULL)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "accession='%s'", accession);
answer2 = sqlGetField(protDbName, "swInterPro", "count(*)", cond_str);
if (answer2 != NULL)
{
interProCount = interProCount + atoi(answer2);
interProCountDouble[ipcnt] = (double)(atoi(answer2));
ipcnt++;
}
else
{
printf("%s is not in InterPro DB.\n", accession);fflush(stdout);
}
}
/* count exons, using coding exons from kgProtMap2 (KG-III) table */
sqlSafefFrag(cond_str, sizeof(cond_str), "spID='%s'", accession);
kgId = sqlGetField(database, "kgXref", "kgID", cond_str);
sqlSafefFrag(cond_str, sizeof(cond_str), "qName='%s'", kgId);
answer2 = sqlGetField(database, "kgProtMap2", "blockCount", cond_str);
if (answer2 != NULL)
{
exonCnt = strdup(answer2);
if (atoi(exonCnt) == 0)
{
errAbort("%s %s has 0 block count\n", accession, protDisplayId);
}
exonCountDouble[jExon] = (double)(atoi(exonCnt));
jExon++;
}
else
{
exonCnt = emptyStr;
}
/* process Mol Wt */
sqlSafefFrag(cond_str, sizeof(cond_str), "accession='%s'", accession);
answer2 = sqlGetField(database, "pepMwAa", "molWeight", cond_str);
if (answer2 != NULL)
{
molWt[molWtCnt] = (double)(atof(answer2));
molWtCnt++;
}
/* process pI */
sqlSafefFrag(cond_str, sizeof(cond_str), "accession='%s'", accession);
answer2 = sqlGetField(database, "pepPi", "pI", cond_str);
if (answer2 != NULL)
{
pI[pIcnt] = (double)(atof(answer2));
pIcnt++;
}
sqlSafefFrag(cond_str, sizeof(cond_str), "acc='%s'", accession);
aaSeq = sqlGetField(proteinDatabaseName, "protein", "val", cond_str);
if (aaSeq == NULL)
{
errAbort("%s does not have protein sequence data in %s, aborting ...\n", accession,
proteinDatabaseName);
}
len = strlen(aaSeq);
chp = aaSeq;
for (i=0; i<len; i++)
{
aaResFound = 0;
for (j=0; j<23; j++)
{
if (*chp == aaAlphabet[j])
{
aaResFound = 1;
aaResCnt[j] ++;
}
}
if (!aaResFound)
{
warn("%c %d not a valid AA residue in %s:\n%s", *chp, *chp, accession, aaSeq);
}
chp++;
}
/* calculate hydrophobicity */
chp = aaSeq;
cCnt = 0;
hydroSum = 0;
for (i=0; i<len; i++)
{
hydroSum = hydroSum + aa_hydro[(int)(*chp)];
/* count Cysteines */
if ((*chp == 'C') || (*chp == 'c'))
{
cCnt ++;
}
chp++;
}
aaLenDouble[icnt] = len;
cCountDouble[icnt] = (double)cCnt;
avgHydro[icnt] = hydroSum/(double)len;
icnt++;
row2 = sqlNextRow(sr2);
}
totalResCnt = 0;
for (i=0; i<23; i++)
{
totalResCnt = totalResCnt + aaResCnt[i];
}
/* write out residue count distribution */
for (i=0; i<20; i++)
{
aaResCntDouble[i] = ((double)aaResCnt[i])/((double)totalResCnt);
fprintf(o2, "%d\t%f\n", i+1, (float)aaResCntDouble[i]);
}
fprintf(o2, "%d\t%f\n", i+1, 0.0);
carefulClose(&o2);
/* calculate and write out various distributions */
calDist(molWt, molWtCnt, 21, 0.0, 10000.0,"pepMolWtDist.tab");
calDist(pI, pIcnt, 61, 3.0, 0.2, "pepPiDist.tab");
calDist(avgHydro, icnt, 41, -2.0, 0.1, "pepHydroDist.tab");
calDist(cCountDouble, icnt, 51, 0.0, 1.0, "pepCCntDist.tab");
calDist(exonCountDouble, jExon, 31, 0.0, 1.0, "pepExonCntDist.tab");
calDist(interProCountDouble, ipcnt, 16, 0.0, 1.0, "pepIPCntDist.tab");
sqlFreeResult(&sr2);
hFreeConn(&conn);
hFreeConn(&conn2);
return(0);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2, *conn3;
char query[256], query2[256], query3[256];
struct sqlResult *sr, *sr2, *sr3;
char **row, **row2, **row3;
char *chp;
FILE *o1, *o2;
char *locusID; /* LocusLink ID */
char *gbAC; /* GenBank accession.version */
char *locusID2; /* LocusLink ID */
char *refAC; /* Refseq accession.version */
char *dbName;
char cond_str[200];
char *kgID;
char *mapID;
char *desc;
if (argc != 2) usage();
dbName = argv[1];
conn = hAllocConn(dbName);
conn2= hAllocConn(dbName);
conn3= hAllocConn(dbName);
o1 = fopen("j.dat", "w");
o2 = fopen("jj.dat", "w");
sprintf(query2,"select * from %sTemp.locus2Ref0;", dbName);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
locusID2 = row2[0];
refAC = row2[1];
sprintf(query, "select * from %sTemp.locus2Acc0 where locusID=%s and seqType='m';",
dbName, locusID2);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
while (row != NULL)
{
locusID = row[0];
gbAC = row[1];
chp = strstr(gbAC, ".");
if (chp != NULL) *chp = '\0';
chp = strstr(refAC, ".");
if (chp != NULL) *chp = '\0';
sprintf(cond_str, "name='%s'", gbAC);
kgID = sqlGetField(dbName, "knownGene", "name", cond_str);
if (kgID != NULL)
{
sprintf(query3, "select * from %sTemp.keggList where locusID = '%s'", dbName, locusID);
sr3 = sqlGetResult(conn3, query3);
while ((row3 = sqlNextRow(sr3)) != NULL)
{
mapID = row3[1];
desc = row3[2];
fprintf(o1, "%s\t%s\t%s\n", kgID, locusID, mapID);
fprintf(o2, "%s\t%s\n", mapID, desc);
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
}
row = sqlNextRow(sr);
}
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
fclose(o1);
fclose(o2);
hFreeConn(&conn);
hFreeConn(&conn2);
mustSystem("cat j.dat|sort|uniq >keggPathway.tab");
mustSystem("cat jj.dat|sort|uniq >keggMapDesc.tab");
mustSystem("rm j.dat");
mustSystem("rm jj.dat");
return(0);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2, *conn3;
char query2[256];
struct sqlResult *sr2;
char **row2;
char cond_str[256];
char *protDbDate;
char *kgID;
char *protDisplayId;
FILE *o1;
char *kgTempDb;
char spDb[255],proteinsDb[255];
char *ro_DB;
char *refSeqName;
char *hugoID;
char *protAcc; /* protein Accession number from NCBI */
char *answer;
char *emptyStr;
char *parSpID;
int leg; /* marker for debugging */
char *spID, *kgProteinID, *geneSymbol, *refseqID, *desc;
if (argc != 4) usage();
kgTempDb = cloneString(argv[1]);
protDbDate = cloneString(argv[2]);
ro_DB = cloneString(argv[3]);
safef(spDb, sizeof(spDb), "sp%s", protDbDate);
safef(proteinsDb, sizeof(proteinsDb), "proteins%s", protDbDate);
conn = hAllocConn(ro_DB);
conn2= hAllocConn(ro_DB);
conn3= hAllocConn(ro_DB);
o1 = mustOpen("j.dat", "w");
emptyStr = strdup("");
sqlSafef(query2, sizeof query2, "select name, proteinID from %s.knownGene;", kgTempDb);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
kgID = row2[0];
kgProteinID = row2[1];
refseqID = strdup("");
geneSymbol = strdup("");
desc = strdup("");
protAcc = strdup("");
sqlSafefFrag(cond_str, sizeof cond_str, "displayID='%s'", kgProteinID);
spID = sqlGetField(proteinsDb, "spXref3", "accession", cond_str);
/* process variant splice proteins */
if (spID == NULL)
{
sqlSafefFrag(cond_str, sizeof cond_str, "varAcc='%s'", kgProteinID);
spID = kgProteinID;
parSpID = sqlGetField(proteinsDb, "splicProt", "parAcc", cond_str);
if (parSpID != NULL)
{
sqlSafefFrag(cond_str, sizeof cond_str, "accession='%s'", parSpID);
protDisplayId = sqlGetField(proteinsDb, "spXref3", "displayID", cond_str);
}
else
{
fprintf(stderr, "%s not found in kgXref3 nor in varProtein.\n", kgProteinID);
exit(1);
}
}
else
{
protDisplayId = kgProteinID;
}
/* use description for the protein as default, replace it with HUGO desc if available. */
sqlSafefFrag(cond_str, sizeof cond_str, "displayID='%s'", protDisplayId);
desc = sqlGetField(proteinsDb, "spXref3", "description", cond_str);
if (strstr(kgID, "NM_") != NULL)
{
leg = 1;
/* special processing for RefSeq DNA based genes */
sqlSafefFrag(cond_str, sizeof cond_str, "mrnaAcc = '%s'", kgID);
refSeqName = sqlGetField(ro_DB, "refLink", "name", cond_str);
if (refSeqName != NULL)
{
geneSymbol = cloneString(refSeqName);
refseqID = kgID;
sqlSafefFrag(cond_str, sizeof cond_str, "mrnaAcc = '%s'", kgID);
desc = sqlGetField(ro_DB, "refLink", "product", cond_str);
sqlSafefFrag(cond_str, sizeof cond_str, "mrnaAcc='%s'", refseqID);
answer = sqlGetField(ro_DB, "refLink", "protAcc", cond_str);
if (answer != NULL)
{
protAcc = strdup(answer);
}
}
}
else
{
sqlSafefFrag(cond_str, sizeof cond_str, "displayID = '%s'", protDisplayId);
hugoID = sqlGetField(proteinsDb, "spXref3", "hugoSymbol", cond_str);
if (!((hugoID == NULL) || (*hugoID == '\0')) )
{
leg = 21;
geneSymbol = cloneString(hugoID);
sqlSafefFrag(cond_str, sizeof cond_str, "displayID = '%s'", protDisplayId);
desc = sqlGetField(proteinsDb, "spXref3", "hugoDesc", cond_str);
if (desc == NULL)
{
printf("%s/%s don't have hugo desc ...\n", kgProteinID, protDisplayId);
fflush(stdout);
}
}
refseqID = emptyStr;
protAcc = emptyStr;
sqlSafefFrag(cond_str, sizeof cond_str, "mrna = '%s'", kgID);
answer = sqlGetField(ro_DB, "mrnaRefseq", "refseq", cond_str);
if (answer != NULL)
{
refseqID = answer;
}
else
{
/*printf("%s does not have a related RefSeq.\n", kgID);fflush(stdout); */
}
if (strlen(geneSymbol) == 0)
{
leg = 23;
if (strlen(refseqID) != 0)
{
sqlSafefFrag(cond_str, sizeof cond_str, "mrnaAcc = '%s'", refseqID);
answer = sqlGetField(ro_DB, "refLink", "name", cond_str);
if (answer != NULL)
{
leg = 24;
geneSymbol = strdup(answer);
}
}
}
}
/* fix missing fields */
if (strlen(refseqID) == 0)
{
/* printf("%3d %s reseqID is empty.\n", leg, kgID); */
}
if (strlen(geneSymbol) == 0)
{
/* printf("%3d %s geneSymbol is empty.\n", leg, kgID);fflush(stdout);*/
geneSymbol = strdup(kgID);
}
if (strlen(desc) == 0)
{
/* printf("%3d %s desc is empty.\n", leg, kgID);fflush(stdout); */
desc = strdup("N/A");
}
fprintf(o1, "%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n",
kgID, kgID, spID, protDisplayId, geneSymbol, refseqID, protAcc, desc);
row2 = sqlNextRow(sr2);
}
fclose(o1);
hFreeConn(&conn);
hFreeConn(&conn2);
hFreeConn(&conn3);
mustSystem("cat j.dat|sort|uniq >kgXref.tab");
mustSystem("rm j.dat");
return(0);
}
void processAlign(char *kgTempDb, char *spDb, char *alignID, int cdsCnt, FILE *outf)
{
struct sqlConnection *conn2, *conn3, *conn4;
char query2[256], query3[256];
struct sqlResult *sr2, *sr3;
char **row2, **row3;
char *score;
char *chrom;
char *protAcc;
char *mrnaID;
char *ranking;
int protDbId;
char condStr[255];
int i;
char *chp;
char *isCurated;
conn2= hAllocConn(kgTempDb);
conn3= hAllocConn(kgTempDb);
conn4= hAllocConn(kgTempDb);
sqlSafef(query2, sizeof(query2), "select * from %s.kgCandidate where alignID='%s'", kgTempDb, alignID);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
mrnaID = row2[0];
chrom = row2[1];
ranking = row2[11];
/* check if it is a composite mrnaID */
/* if yes, select from entries with both protein and mrna specified */
if (alignID[0] == 'U')
{
chp = strstr(row2[0], "_");
*chp = '\0';
protAcc = row2[0];
chp ++;
mrnaID = chp;
sqlSafef(query3, sizeof(query3),
"select protAcc, score from %s.protMrnaScore where mrnaAcc='%s' and protAcc='%s'",
kgTempDb, mrnaID, protAcc);
}
else
{
sqlSafef(query3, sizeof(query3),
"select protAcc, score from %s.protMrnaScore where mrnaAcc='%s' order by score desc",
kgTempDb, mrnaID);
}
sr3 = sqlMustGetResult(conn3, query3);
row3 = sqlNextRow(sr3);
while(row3 != NULL)
{
protAcc = row3[0];
score = row3[1];
chp = strstr(protAcc, "-");
if (chp == NULL)
{
sqlSafefFrag(condStr, sizeof(condStr), "acc='%s'", protAcc);
isCurated = sqlGetField(spDb, "info", "isCurated", condStr);
if (sameWord(isCurated, "1"))
{
protDbId = 1;
}
else
{
protDbId = 2;
}
}
else
{
protDbId = 4;
}
fprintf(outf, "%s:", chrom);
for (i=0; i<cdsCnt; i++) fprintf(outf, "%s", cdsBloc[i]);
fprintf(outf, "\t%s\t%d\t%8s\t%s\t%s\t%s\n",
ranking, protDbId, score, mrnaID, protAcc, alignID);
/* for composite type, process just one record */
if (alignID[0] == 'U') break;
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn2);
hFreeConn(&conn3);
hFreeConn(&conn4);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn3;
char query[256], query3[256];
struct sqlResult *sr, *sr3;
char **row, **row3;
FILE *o1, *o2;
char *locusID; /* LocusLink ID */
char *kgTempDbName, *roDbName;
char cond_str[200];
char *kgId;
char *mapID;
char *desc;
char *mRNA;
optionInit(&argc, argv, options);
if (argc != 3) usage();
kgTempDbName = argv[1];
roDbName = argv[2];
conn = hAllocConn(roDbName);
conn3= hAllocConn(roDbName);
o1 = fopen("j.dat", "w");
o2 = fopen("jj.dat", "w");
table = optionVal("table", "knownGene");
sqlSafef(query, sizeof(query), "select name from %s.%s", roDbName, table);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
while (row != NULL)
{
kgId = row[0];
sqlSafefFrag(cond_str, sizeof(cond_str), "kgId='%s'", kgId);
mRNA = sqlGetField(roDbName, "kgXref", "mRNA", cond_str);
sqlSafefFrag(cond_str, sizeof(cond_str), "mrna='%s'", mRNA);
locusID = sqlGetField("entrez", "entrezMrna", "geneId", cond_str);
/* look for RefSeq if not found in mRNAs */
if (locusID == NULL)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "refseq='%s'", mRNA);
locusID = sqlGetField("entrez", "entrezRefseq", "geneId", cond_str);
}
if (locusID != NULL)
{
sqlSafef(query3, sizeof(query3), "select * from %s.keggList where locusID = '%s'", kgTempDbName, locusID);
sr3 = sqlGetResult(conn3, query3);
while ((row3 = sqlNextRow(sr3)) != NULL)
{
mapID = row3[1];
desc = row3[2];
fprintf(o1, "%s\t%s\t%s\n", kgId, locusID, mapID);
fprintf(o2, "%s\t%s\n", mapID, desc);
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
}
else
{
/* printf("%s not found in Entrez.\n", kgId);fflush(stdout);*/
if (differentString(table, "knownGene"))
{
sqlSafefFrag(cond_str, sizeof(cond_str), "name='%s'", kgId);
locusID = sqlGetField(roDbName, table, "name2", cond_str);
sqlSafef(query3, sizeof(query3), "select * from %s.keggList where locusID = '%s'", kgTempDbName, kgId);
sr3 = sqlGetResult(conn3, query3);
while ((row3 = sqlNextRow(sr3)) != NULL)
{
mapID = row3[1];
desc = row3[2];
fprintf(o1, "%s\t%s\t%s\n", kgId, locusID, mapID);
fprintf(o2, "%s\t%s\n", mapID, desc);
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
}
}
row = sqlNextRow(sr);
}
fclose(o1);
fclose(o2);
hFreeConn(&conn);
mustSystem("cat j.dat|sort|uniq >keggPathway.tab");
mustSystem("cat jj.dat|sort|uniq >keggMapDesc.tab");
mustSystem("rm j.dat");
mustSystem("rm jj.dat");
return(0);
}
void doTracks(char *proteinID, char *mrnaID, char *aa, int *yOffp, char *psOutput)
/* draw various protein tracks */
{
int l;
char aaOrigOffsetStr[20];
int hasResFreq;
char uniProtDbName[50];
char *protDbDate;
char *chrom;
char strand;
char *kgId, *kgPep, *protPep;
char cond_str[255];
char *answer;
//int i, ll;
//char *chp1, *chp2;
g_font = mgSmallFont();
safef(pbScaleStr, sizeof(pbScaleStr), "%d", pbScale);
if (psOutput != NULL)
{
pbScale = atoi(cartOptionalString(cart, "pbt.pbScaleStr"));
}
if (cgiOptionalString("trackOffset") != NULL)
{
trackOrigOffset = atoi(cgiOptionalString("trackOffset"));
}
if (cgiOptionalString("pbScaleStr") != NULL)
{
pbScale = atoi(cgiOptionalString("pbScaleStr"));
}
if (cgiOptionalString("pbScale") != NULL)
{
scaleButtonPushed = TRUE;
if (strcmp(cgiOptionalString("pbScale"), "1/6") == 0) pbScale = 1;
if (strcmp(cgiOptionalString("pbScale"), "1/2") == 0) pbScale = 3;
if (strcmp(cgiOptionalString("pbScale"), "FULL") == 0) pbScale = 6;
if (strcmp(cgiOptionalString("pbScale"), "DNA") == 0) pbScale =22;
safef(pbScaleStr, sizeof(pbScaleStr), "%d", pbScale);
cgiMakeHiddenVar("pbScaleStr", pbScaleStr);
}
else
{
scaleButtonPushed = FALSE;
}
if (psOutput == NULL)
{
if (cgiVarExists("pbt.left3"))
{
relativeScroll(-0.95);
initialWindow = FALSE;
}
else if (cgiVarExists("pbt.left2"))
{
relativeScroll(-0.475);
initialWindow = FALSE;
}
else if (cgiVarExists("pbt.left1"))
{
relativeScroll(-0.02);
initialWindow = FALSE;
}
else if (cgiVarExists("pbt.right1"))
{
relativeScroll(0.02);
initialWindow = FALSE;
}
else if (cgiVarExists("pbt.right2"))
{
relativeScroll(0.475);
initialWindow = FALSE;
}
else if (cgiVarExists("pbt.right3"))
{
relativeScroll(0.95);
initialWindow = FALSE;
}
}
dnaUtilOpen();
l=strlen(aa);
/* initialize AA properties */
aaPropertyInit(&hasResFreq);
sfCount = getSuperfamilies2(proteinID);
if (sfCount == 0)
{
sfCount = getSuperfamilies(proteinID);
}
if (mrnaID != NULL)
{
if (kgVersion == KG_III)
{
doExonTrack = FALSE;
sqlSafefFrag(cond_str, sizeof(cond_str), "spId='%s'", proteinID);
kgId = sqlGetField(database, "kgXref", "kgId", cond_str);
if (kgId != NULL)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "name='%s'", kgId);
kgPep = sqlGetField(database, "knownGenePep", "seq", cond_str);
//printf("<pre><br>%s", kgPep);fflush(stdout);
if (kgPep != NULL)
{
if (strstr(protDbName, "proteins") != NULL)
{
protDbDate = strstr(protDbName, "proteins") + strlen("proteins");
safef(uniProtDbName, sizeof(uniProtDbName),"sp%s", protDbDate);
sqlSafefFrag(cond_str, sizeof(cond_str), "acc='%s'", proteinID);
protPep = sqlGetField(uniProtDbName, "protein", "val", cond_str);
//printf("<br>%s\n", protPep);fflush(stdout);
if (protPep != NULL)
{
if (sameWord(kgPep, protPep))
{
//printf("<br>MATCH!\n");fflush(stdout);
sqlSafefFrag(cond_str, sizeof(cond_str), "qName='%s'", kgId);
answer = sqlGetField(database, kgProtMapTableName,
"qName", cond_str);
if (answer != NULL)
{
/* NOTE: passing in kgId instead of proteinID because
kgProtMap2's qName uses kgId instead of
protein display ID */
getExonInfo(kgId, &exCount, &chrom, &strand);
assert(exCount > 0);
doExonTrack = TRUE;
}
}
/*
else
{
chp1 = kgPep;
printf("<br>");
chp2 = protPep;
ll = strlen(kgPep);
if (strlen(protPep) < ll) ll= strlen(protPep);
for (i=0; i<ll; i++)
{
if (*chp1 != *chp2)
{
printf("%c", *chp1);
}
else
{
printf(".");
}
chp1++; chp2++;
}
}
//printf("</pre>");fflush(stdout);
*/
}
}
}
}
}
else
{
doExonTrack = TRUE;
getExonInfo(proteinID, &exCount, &chrom, &strand);
assert(exCount > 0);
}
/* do the following only if pbTracks called doTracks() */
if (initialWindow && IAmPbTracks)
{
prevGBOffsetSav = calPrevGB(exCount, chrom, strand, l, yOffp, proteinID, mrnaID);
trackOrigOffset = prevGBOffsetSav;
if (trackOrigOffset > (protSeqLen*pbScale - 600))
trackOrigOffset = protSeqLen*pbScale - 600;
/* prevent negative value */
if (trackOrigOffset < 0) trackOrigOffset = 0;
}
/* if this if for PDF/Postscript, the trackOrigOffset is already calculated previously,
use the saved value */
if (psOutput != NULL)
{
trackOrigOffset = atoi(cartOptionalString(cart, "pbt.trackOffset"));
}
}
/*printf("<br>%d %d<br>%d %d\n", prevGBStartPos, prevGBEndPos,
blockGenomeStartPositive[exCount-1], blockGenomeStartPositive[0]); fflush(stdout);
*/
if (strand == '-')
{
if ((prevGBStartPos <= blockGenomeStartPositive[exCount-1]) && (prevGBEndPos >= blockGenomeStartPositive[0]))
{
showPrevGBPos = FALSE;
}
}
else
{
if ((prevGBStartPos <= blockGenomeStartPositive[0]) && (prevGBEndPos >= blockGenomeStartPositive[exCount-1]))
{
showPrevGBPos = FALSE;
}
}
if ((cgiOptionalString("aaOrigOffset") != NULL) && scaleButtonPushed)
{
trackOrigOffset = atoi(cgiOptionalString("aaOrigOffset"))*pbScale;
}
pixWidth = 160+ protSeqLen*pbScale;
if (pixWidth > MAX_PB_PIXWIDTH)
{
pixWidth = MAX_PB_PIXWIDTH;
}
if ((protSeqLen*pbScale - trackOrigOffset) < MAX_PB_PIXWIDTH)
{
pixWidth = protSeqLen*pbScale - trackOrigOffset + 160;
}
if (pixWidth < 550) pixWidth = 550;
insideWidth = pixWidth-gfxBorder;
if (proteinInSupportedGenome)
{
pixHeight = 250;
}
else
{
pixHeight = 215;
}
if (sfCount > 0) pixHeight = pixHeight + 20;
/* make room for individual residues display */
if (pbScale >=6) pixHeight = pixHeight + 20;
if (pbScale >=18) pixHeight = pixHeight + 30;
if (psOutput)
{
vg = vgOpenPostScript(pixWidth, pixHeight, psOutput);
suppressHtml = TRUE;
hideControls = TRUE;
}
else
{
trashDirFile(&gifTn, "pbt", "pbt", ".png");
vg = vgOpenPng(pixWidth, pixHeight, gifTn.forCgi, FALSE);
}
/* Put up horizontal scroll controls. */
hWrites("Move ");
hButton("pbt.left3", "<<<");
hButton("pbt.left2", " <<");
hButton("pbt.left1", " < ");
hButton("pbt.right1", " > ");
hButton("pbt.right2", ">> ");
hButton("pbt.right3", ">>>");
hPrintf("     ");
/* Put up scaling controls. */
hPrintf("Current scale: ");
if (pbScale == 1) hPrintf("1/6 ");
if (pbScale == 3) hPrintf("1/2 ");
if (pbScale == 6) hPrintf("FULL ");
if (pbScale == 22) hPrintf("DNA ");
hPrintf("    Rescale to ");
hPrintf("<INPUT TYPE=SUBMIT NAME=\"pbScale\" VALUE=\"1/6\">\n");
hPrintf("<INPUT TYPE=SUBMIT NAME=\"pbScale\" VALUE=\"1/2\">\n");
hPrintf("<INPUT TYPE=SUBMIT NAME=\"pbScale\" VALUE=\"FULL\">\n");
if (kgVersion == KG_III)
{
/* for KG III, the protein has to exist in the kgProtMap2 table
(which will turn on doExonTrack flag)
to provide the genomic position data needed for DNA sequence display */
if ((proteinInSupportedGenome) && (doExonTrack))
hPrintf("<INPUT TYPE=SUBMIT NAME=\"pbScale\" VALUE=\"DNA\">\n");
}
else
{
if (proteinInSupportedGenome)
hPrintf("<INPUT TYPE=SUBMIT NAME=\"pbScale\" VALUE=\"DNA\">\n");
}
hPrintf("<FONT SIZE=1><BR><BR></FONT>\n");
g_vg = vg;
pbRed = vgFindColorIx(g_vg, 0xf9, 0x51, 0x59);
pbBlue = vgFindColorIx(g_vg, 0x00, 0x00, 0xd0);
bkgColor = vgFindColorIx(vg, 255, 254, 232);
vgBox(vg, 0, 0, insideWidth, pixHeight, bkgColor);
/* Start up client side map. */
hPrintf("<MAP Name=%s>\n", mapName);
vgSetClip(vg, 0, gfxBorder, insideWidth, pixHeight - 2*gfxBorder);
/* start drawing indivisual tracks */
doAAScale(l, yOffp, 1);
if (pbScale >= 6) doResidues(aa, l, yOffp);
if (pbScale >= 18) doDnaTrack(chrom, strand, exCount, l, yOffp);
if ((mrnaID != NULL) && showPrevGBPos)
{
doPrevGB(exCount, chrom, strand, l, yOffp, proteinID, mrnaID);
}
if (mrnaID != NULL)
{
if (doExonTrack) doExon(exCount, chrom, l, yOffp, proteinID, mrnaID);
}
doCharge(aa, l, yOffp);
doHydrophobicity(aa, l, yOffp);
doCysteines(aa, l, yOffp);
if (sfCount > 0) doSuperfamily(ensPepName, sfCount, yOffp);
if (hasResFreq) doAnomalies(aa, l, yOffp);
doAAScale(l, yOffp, -1);
vgClose(&vg);
/* Finish map and save out picture and tell html file about it. */
hPrintf("</MAP>\n");
/* put tracks image here */
hPrintf(
"\n<IMG SRC=\"%s\" BORDER=1 WIDTH=%d HEIGHT=%d USEMAP=#%s><BR>",
gifTn.forCgi, pixWidth, pixHeight, mapName);
if (proteinInSupportedGenome)
{
hPrintf("<A HREF=\"../goldenPath/help/pbTracksHelpFiles/pbTracksHelp.shtml#tracks\" TARGET=_blank>");
}
else
{
if (hIsGsidServer())
{
hPrintf("<A HREF=\"../goldenPath/help/pbTracksHelpFiles/pbGsid/pbTracksHelp.shtml#tracks\" TARGET=_blank>");
}
else
{
hPrintf("<A HREF=\"../goldenPath/help/pbTracksHelpFiles/pbTracksHelp.shtml#tracks\" TARGET=_blank>");
}
}
hPrintf("Explanation of Protein Tracks</A><br>");
safef(trackOffset, sizeof(trackOffset), "%d", trackOrigOffset);
cgiMakeHiddenVar("trackOffset", trackOffset);
/* remember where the AA base origin is so that it can be passed to next PB page */
aaOrigOffset = trackOrigOffset/pbScale;
safef(aaOrigOffsetStr, sizeof(aaOrigOffsetStr), "%d", aaOrigOffset);
cgiMakeHiddenVar("aaOrigOffset", aaOrigOffsetStr);
/* save the following state variables, to be used by PDF/Postcript processing */
cartSetString(cart,"pbt.pbScaleStr", pbScaleStr);
cartSetString(cart,"pbt.trackOffset", trackOffset);
cartSaveSession(cart);
fflush(stdout);
}
static void showProtH1n1(char *item, char *geneSymbol)
{
char query2[256];
struct sqlResult *sr2;
char **row2;
struct sqlConnection *conn2 = hAllocConn(database);
char *subjId, *dnaSeqId;
char *aaSeqId= NULL;
char *gene=NULL;
char cond_str[256];
char *predFN;
char *homologID;
char *SCOPdomain;
char *chain;
char goodSCOPdomain[40];
int first = 1;
float eValue;
char *chp;
int homologCount;
int gotPDBFile = 0;
sqlSafef(query2, sizeof(query2),
"select subjId, dnaSeqId, aaSeqId, gene from gisaidXref where dnaSeqId='%s'", item);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
if (row2 != NULL)
{
subjId = strdup(row2[0]);
dnaSeqId = strdup(row2[1]);
aaSeqId = strdup(row2[2]);
gene = strdup(row2[3]);
}
else
{
errAbort("%s not found.", item);
}
sqlFreeResult(&sr2);
printf("<H3>Protein Structure Analysis and Prediction</H3>");
printf("<B>Comparison to 1918 Flu Virus:</B> ");
printf("<A HREF=\"%s/%s/%s/1918_%s.mutate", getH1n1StructUrl(), gene, aaSeqId, aaSeqId);
printf("\" TARGET=_blank>%s</A><BR>\n", aaSeqId);
printf("<B>Comparison to A H1N1 gene %s concensus:</B> ", gene);
printf("<A HREF=\"%s/%s/%s/consensus_%s.mutate", getH1n1StructUrl(), gene, aaSeqId, aaSeqId);
printf("\" TARGET=_blank>%s</A><BR>\n", aaSeqId);
printf("<BR><B>3D Structure Prediction of %s concensus sequence (with variation of sequence %s highlighted):", geneSymbol, item);
printf("<BR>PDB file:</B> ");
char pdbUrl[PATH_LEN];
safef(pdbUrl, sizeof(pdbUrl), "%s/%s/decoys/%s.try1-opt3.pdb.gz", getH1n1StructUrl(), item, item);
// Modeller stuff
char modelPdbUrl[PATH_LEN];
if (getH1n1Model(gene, modelPdbUrl))
{
struct tempName imageFile, chimeraScript, chimerax;
mkH1n1StructData(gene, NULL, aaSeqId, &imageFile, &chimeraScript);
mkChimerax(gene, modelPdbUrl, chimeraScript.forCgi, &chimerax);
printf("<A HREF=\"%s\" TARGET=_blank>%s</A>, view with <A HREF=\"%s\">Chimera</A><BR>\n", modelPdbUrl, gene, chimerax.forHtml);
printf("<TABLE>\n");
printf("<TR>\n");
printf("<TD ALIGN=\"center\"><img src=\"%s\"></TD>", imageFile.forHtml);
printf("</TR>\n");
printf("</TABLE>\n");
}
return;
gotPDBFile = 0;
sqlSafefFrag(cond_str, sizeof(cond_str), "proteinID='%s' and evalue <1.0e-5;", item);
printf("<TABLE>\n");
printf("<TR><TD ALIGN=\"center\">Front</TD>\n");
printf("<TD ALIGN=\"center\">Top</TD>\n");
printf("<TD ALIGN=\"center\">Side</TD>\n");
printf("</TR>\n");
printf("<TR>\n");
printf("<TD ALIGN=\"center\"><img src=\"%s/%s/%s.undertaker-align.view1_200.jpg\"></TD>", getH1n1StructUrl(), item, item);
printf("<TD ALIGN=\"center\"><img src=\"%s/%s/%s.undertaker-align.view2_200.jpg\"></TD>", getH1n1StructUrl(), item, item);
printf("<TD ALIGN=\"center\"><img src=\"%s/%s/%s.undertaker-align.view3_200.jpg\"></TD>", getH1n1StructUrl(), item, item);
printf("</TR>\n");
printf("<TR>\n");
printf("<TD ALIGN=\"center\"><A HREF=\"%s/%s/%s.undertaker-align.view1_500.jpg\">500x500</A></TD>",
getH1n1StructUrl(), item, item);
printf("<TD ALIGN=\"center\"><A HREF=\"%s/%s/%s.undertaker-align.view2_500.jpg\">500x500</A></TD>",
getH1n1StructUrl(), item, item);
printf("<TD ALIGN=\"center\"><A HREF=\"%s/%s/%s.undertaker-align.view3_500.jpg\">500x500</A></TD>",
getH1n1StructUrl(), item, item);
printf("</TR>\n");
printf("</TABLE>\n");
printf("<BR><B>Detailed results of SAM-T02:</B> ");
printf("<A HREF=\"%s/%s/summary.html", getH1n1StructUrl(), item);
printf("\" TARGET=_blank>%s</A><BR>\n", item);
/* by pass the following additional processing for now, until two necessary tables are built */
hFreeConn(&conn2);
return;
if (sqlGetField(database, "protHomolog", "proteinID", cond_str) != NULL)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "proteinID='%s'", item);
predFN = sqlGetField(database, "protPredFile", "predFileName", cond_str);
if (predFN != NULL)
{
printf("<A HREF=\"../SARS/%s/", item);
/* printf("%s.t2k.undertaker-align.pdb\">%s</A><BR>\n", item,item); */
printf("%s\">%s</A><BR>\n", predFN,item);
gotPDBFile = 1;
}
}
if (!gotPDBFile)
{
printf("No high confidence level structure prediction available for this sequence.");
printf("<BR>\n");
}
printf("<B>3D Structure of Close Homologs:</B> ");
homologCount = 0;
strcpy(goodSCOPdomain, "dummy");
conn2= hAllocConn(database);
sqlSafef(query2, sizeof(query2),
"select homologID,eValue,SCOPdomain,chain from sc1.protHomolog where proteinID='%s' and evalue <= 0.01;",
item);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
if (row2 != NULL)
{
while (row2 != NULL)
{
homologID = row2[0];
sscanf(row2[1], "%e", &eValue);
SCOPdomain = row2[2];
chp = SCOPdomain+strlen(SCOPdomain)-1;
while (*chp != '.') chp--;
*chp = '\0';
chain = row2[3];
if (eValue <= 1.0e-10)
strcpy(goodSCOPdomain, SCOPdomain);
else
{
if (strcmp(goodSCOPdomain,SCOPdomain) != 0)
goto skip;
else
if (eValue > 0.1) goto skip;
}
if (first)
first = 0;
else
printf(", ");
printf("<A HREF=\"http://www.rcsb.org/pdb/cgi/explore.cgi?job=graphics&pdbId=%s",
homologID);
if (strlen(chain) >= 1)
printf("\"TARGET=_blank>%s(chain %s)</A>", homologID, chain);
else
printf("\"TARGET=_blank>%s</A>", homologID);
homologCount++;
skip:
row2 = sqlNextRow(sr2);
}
}
hFreeConn(&conn2);
sqlFreeResult(&sr2);
if (homologCount == 0)
printf("None<BR>\n");
printf("<BR><B>Details:</B> ");
printf("<A HREF=\"../SARS/%s/summary.html", item);
printf("\" TARGET=_blank>%s</A><BR>\n", item);
htmlHorizontalLine();
}
int getSuperfamilies2(char *proteinID)
/* getSuperfamilies2() superceed getSuperfamilies() starting from hg16,
it gets Superfamily data of a protein
from ensemblXref3, sfAssign, and sfDes from the proteinsXXXXXX database,
and placed them in arrays to be used by doSuperfamily().*/
{
struct sqlConnection *conn, *conn2, *conn3;
char query[MAXNAMELEN], query2[MAXNAMELEN];
struct sqlResult *sr, *sr2;
char **row, **row2;
char cond_str[255];
char *sfID, *seqID, *sfDesc, *region;
int done;
int j;
char *chp, *chp2;
int sfCnt;
int int_start, int_end;
if (!hTableExists(protDbName, "sfAssign")) return(0);
if (!hTableExists(protDbName, "ensemblXref3")) return(0);
conn = hAllocConn(database);
conn2 = hAllocConn(database);
conn3 = hAllocConn(database);
sqlSafef(query2, sizeof(query),
"select distinct sfID, seqID from %s.ensemblXref3 x, %s.sfAssign a where (swissAcc='%s' or tremblAcc='%s') and seqID=x.protein and protein != '' and evalue <= 0.02",
protDbName, protDbName, proteinID, proteinID);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
sfCnt=0;
while (row2 != NULL)
{
sfID = row2[0];
seqID= row2[1];
sqlSafef(query, sizeof(query),
"select region from %s.sfAssign where sfID='%s' and seqID='%s' and evalue <=0.02",
protDbName, sfID, seqID);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
while (row != NULL)
{
region = row[0];
for (j=0; j<sfCnt; j++)
{
if (sfId[j] == atoi(sfID)) goto skip;
}
sqlSafefFrag(cond_str, sizeof(cond_str), "id=%s;", sfID);
sfDesc = sqlGetField(protDbName, "sfDes", "description", cond_str);
/* !!! refine logic here later to be defensive against illegal syntax */
chp = region;
done = 0;
while (!done)
{
chp2 = strstr(chp, "-");
*chp2 = '\0';
chp2++;
sscanf(chp, "%d", &int_start);
chp = chp2;
chp2 = strstr(chp, ",");
if (chp2 != NULL)
{
*chp2 = '\0';
}
else
{
done = 1;
}
chp2++;
sscanf(chp, "%d", &int_end);
sfId[sfCnt] = atoi(sfID);
sfStart[sfCnt] = int_start;
sfEnd[sfCnt] = int_end;
strncpy(superfam_name[sfCnt], sfDesc, MAXNAMELEN-1);
sfCnt++;
chp = chp2;
}
skip:
row = sqlNextRow(sr);
}
sqlFreeResult(&sr);
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn);
hFreeConn(&conn2);
hFreeConn(&conn3);
return(sfCnt);
}
void showSAM_h1n1(char *item)
{
char query2[256];
struct sqlResult *sr2;
char **row2;
struct sqlConnection *conn2 = hAllocConn(database);
char cond_str[256];
char *predFN;
char *homologID;
char *SCOPdomain;
char *chain;
char goodSCOPdomain[40];
int first = 1;
float eValue;
char *chp;
int homologCount;
int gotPDBFile = 0;
printf("<H3>Protein Structure Analysis and Prediction by ");
printf("<A HREF=\"http://www.soe.ucsc.edu/research/compbio/SAM_T02/sam-t02-faq.html\"");
printf(" TARGET=_blank>SAM-T02</A></H3>\n");
printf("<B>Multiple Alignment:</B> ");
printf("<A HREF=\"%s/%s/summary.html#alignment", getH1n1StructUrl(), item);
printf("\" TARGET=_blank>%s</A><BR>\n", item);
printf("<B>Secondary Structure Predictions:</B> ");
printf("<A HREF=\"%s/%s/summary.html#secondary-structure", getH1n1StructUrl(), item);
printf("\" TARGET=_blank>%s</A><BR>\n", item);
printf("<B>3D Structure Prediction (PDB file):</B> ");
char pdbUrl[PATH_LEN];
safef(pdbUrl, sizeof(pdbUrl), "%s/%s/decoys/%s.try1-opt3.pdb.gz", getH1n1StructUrl(), item, item);
struct tempName chimerax;
mkChimerax(item, pdbUrl, NULL, &chimerax);
printf("<A HREF=\"%s\" TARGET=_blank>%s</A>, view with <A HREF=\"%s\">Chimera</A><BR>\n", pdbUrl, item, chimerax.forHtml);
gotPDBFile = 0;
sqlSafefFrag(cond_str, sizeof(cond_str), "proteinID='%s' and evalue <1.0e-5;", item);
printf("<TABLE>\n");
printf("<TR><TD ALIGN=\"center\">Front</TD>\n");
printf("<TD ALIGN=\"center\">Top</TD>\n");
printf("<TD ALIGN=\"center\">Side</TD>\n");
printf("</TR>\n");
printf("<TR>\n");
printf("<TD ALIGN=\"center\"><img src=\"%s/%s/%s.undertaker-align.view1_200.jpg\"></TD>", getH1n1StructUrl(), item, item);
printf("<TD ALIGN=\"center\"><img src=\"%s/%s/%s.undertaker-align.view2_200.jpg\"></TD>", getH1n1StructUrl(), item, item);
printf("<TD ALIGN=\"center\"><img src=\"%s/%s/%s.undertaker-align.view3_200.jpg\"></TD>", getH1n1StructUrl(), item, item);
printf("</TR>\n");
printf("<TR>\n");
printf("<TD ALIGN=\"center\"><A HREF=\"%s/%s/%s.undertaker-align.view1_500.jpg\">500x500</A></TD>",
getH1n1StructUrl(), item, item);
printf("<TD ALIGN=\"center\"><A HREF=\"%s/%s/%s.undertaker-align.view2_500.jpg\">500x500</A></TD>",
getH1n1StructUrl(), item, item);
printf("<TD ALIGN=\"center\"><A HREF=\"%s/%s/%s.undertaker-align.view3_500.jpg\">500x500</A></TD>",
getH1n1StructUrl(), item, item);
printf("</TR>\n");
printf("</TABLE>\n");
printf("<BR><B>Detailed results of SAM-T02:</B> ");
printf("<A HREF=\"%s/%s/summary.html", getH1n1StructUrl(), item);
printf("\" TARGET=_blank>%s</A><BR>\n", item);
/* by pass the following additional processing for now, until two necessary tables are built */
hFreeConn(&conn2);
return;
if (sqlGetField(database, "protHomolog", "proteinID", cond_str) != NULL)
{
sqlSafefFrag(cond_str, sizeof(cond_str), "proteinID='%s'", item);
predFN = sqlGetField(database, "protPredFile", "predFileName", cond_str);
if (predFN != NULL)
{
printf("<A HREF=\"../SARS/%s/", item);
/* printf("%s.t2k.undertaker-align.pdb\">%s</A><BR>\n", item,item); */
printf("%s\">%s</A><BR>\n", predFN,item);
gotPDBFile = 1;
}
}
if (!gotPDBFile)
{
printf("No high confidence level structure prediction available for this sequence.");
printf("<BR>\n");
}
printf("<B>3D Structure of Close Homologs:</B> ");
homologCount = 0;
strcpy(goodSCOPdomain, "dummy");
conn2= hAllocConn(database);
sqlSafef(query2, sizeof(query2),
"select homologID,eValue,SCOPdomain,chain from sc1.protHomolog where proteinID='%s' and evalue <= 0.01;",
item);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
if (row2 != NULL)
{
while (row2 != NULL)
{
homologID = row2[0];
sscanf(row2[1], "%e", &eValue);
SCOPdomain = row2[2];
chp = SCOPdomain+strlen(SCOPdomain)-1;
while (*chp != '.') chp--;
*chp = '\0';
chain = row2[3];
if (eValue <= 1.0e-10)
strcpy(goodSCOPdomain, SCOPdomain);
else
{
if (strcmp(goodSCOPdomain,SCOPdomain) != 0)
goto skip;
else
if (eValue > 0.1) goto skip;
}
if (first)
first = 0;
else
printf(", ");
printf("<A HREF=\"http://www.rcsb.org/pdb/cgi/explore.cgi?job=graphics&pdbId=%s",
homologID);
if (strlen(chain) >= 1)
printf("\"TARGET=_blank>%s(chain %s)</A>", homologID, chain);
else
printf("\"TARGET=_blank>%s</A>", homologID);
homologCount++;
skip:
row2 = sqlNextRow(sr2);
}
}
hFreeConn(&conn2);
sqlFreeResult(&sr2);
if (homologCount == 0)
printf("None<BR>\n");
printf("<BR><B>Details:</B> ");
printf("<A HREF=\"../SARS/%s/summary.html", item);
printf("\" TARGET=_blank>%s</A><BR>\n", item);
htmlHorizontalLine();
}
void processRefSeq(char *database, char *faFile, char *raFile, char *pslFile, char *loc2refFile,
char *pepFile, char *mim2locFile)
/* hgRefSeqMrna - Load refSeq mRNA alignments and other info into
* refSeqGene table. */
{
struct lineFile *lf;
struct hash *raHash, *rsiHash = newHash(0);
struct hash *loc2mimHash = newHash(0);
struct refSeqInfo *rsiList = NULL, *rsi;
char *s, *line, *row[5];
int wordCount, dotMod = 0;
int noLocCount = 0;
int rsiCount = 0;
int noProtCount = 0;
struct psl *psl;
struct sqlConnection *conn = hgStartUpdate(database);
struct hash *productHash = loadNameTable(conn, "productName", 16);
struct hash *geneHash = loadNameTable(conn, "geneName", 16);
char *kgName = "refGene";
FILE *kgTab = hgCreateTabFile(".", kgName);
FILE *productTab = hgCreateTabFile(".", "productName");
FILE *geneTab = hgCreateTabFile(".", "geneName");
FILE *refLinkTab = hgCreateTabFile(".", "refLink");
FILE *refPepTab = hgCreateTabFile(".", "refPep");
FILE *refMrnaTab = hgCreateTabFile(".", "refMrna");
struct exon *exonList = NULL, *exon;
char *answer;
char cond_str[200];
/* Make refLink and other tables table if they don't exist already. */
sqlMaybeMakeTable(conn, "refLink", refLinkTableDef);
sqlUpdate(conn, "NOSQLINJ delete from refLink");
sqlMaybeMakeTable(conn, "refGene", refGeneTableDef);
sqlUpdate(conn, "NOSQLINJ delete from refGene");
sqlMaybeMakeTable(conn, "refPep", refPepTableDef);
sqlUpdate(conn, "NOSQLINJ delete from refPep");
sqlMaybeMakeTable(conn, "refMrna", refMrnaTableDef);
sqlUpdate(conn, "NOSQLINJ delete from refMrna");
/* Scan through locus link to omim ID file and put in hash. */
{
char *row[2];
printf("Scanning %s\n", mim2locFile);
lf = lineFileOpen(mim2locFile, TRUE);
while (lineFileRow(lf, row))
{
hashAdd(loc2mimHash, row[1], intToPt(atoi(row[0])));
}
lineFileClose(&lf);
}
/* Scan through .ra file and make up start of refSeqInfo
* objects in hash and list. */
printf("Scanning %s\n", raFile);
lf = lineFileOpen(raFile, TRUE);
while ((raHash = hashNextRa(lf)) != NULL)
{
if (clDots > 0 && ++dotMod == clDots )
{
dotMod = 0;
dotOut();
}
AllocVar(rsi);
slAddHead(&rsiList, rsi);
if ((s = hashFindVal(raHash, "acc")) == NULL)
errAbort("No acc near line %d of %s", lf->lineIx, lf->fileName);
rsi->mrnaAcc = cloneString(s);
if ((s = hashFindVal(raHash, "siz")) == NULL)
errAbort("No siz near line %d of %s", lf->lineIx, lf->fileName);
rsi->size = atoi(s);
if ((s = hashFindVal(raHash, "gen")) != NULL)
rsi->geneName = cloneString(s);
//!!!else
//!!! warn("No gene name for %s", rsi->mrnaAcc);
if ((s = hashFindVal(raHash, "cds")) != NULL)
parseCds(s, 0, rsi->size, &rsi->cdsStart, &rsi->cdsEnd);
else
rsi->cdsEnd = rsi->size;
if ((s = hashFindVal(raHash, "ngi")) != NULL)
rsi->ngi = atoi(s);
rsi->geneNameId = putInNameTable(geneHash, geneTab, rsi->geneName);
s = hashFindVal(raHash, "pro");
if (s != NULL)
rsi->productName = cloneString(s);
rsi->productNameId = putInNameTable(productHash, productTab, s);
hashAdd(rsiHash, rsi->mrnaAcc, rsi);
freeHashAndVals(&raHash);
}
lineFileClose(&lf);
if (clDots) printf("\n");
/* Scan through loc2ref filling in some gaps in rsi. */
printf("Scanning %s\n", loc2refFile);
lf = lineFileOpen(loc2refFile, TRUE);
while (lineFileNext(lf, &line, NULL))
{
char *mrnaAcc;
if (line[0] == '#')
continue;
wordCount = chopTabs(line, row);
if (wordCount < 5)
errAbort("Expecting at least 5 tab-separated words line %d of %s",
lf->lineIx, lf->fileName);
mrnaAcc = row[1];
mrnaAcc = accWithoutSuffix(mrnaAcc);
if (mrnaAcc[2] != '_')
warn("%s is and odd name %d of %s",
mrnaAcc, lf->lineIx, lf->fileName);
if ((rsi = hashFindVal(rsiHash, mrnaAcc)) != NULL)
{
rsi->locusLinkId = lineFileNeedNum(lf, row, 0);
rsi->omimId = ptToInt(hashFindVal(loc2mimHash, row[0]));
rsi->proteinAcc = cloneString(accWithoutSuffix(row[4]));
}
}
lineFileClose(&lf);
/* Report how many seem to be missing from loc2ref file.
* Write out knownInfo file. */
printf("Writing %s\n", "refLink.tab");
for (rsi = rsiList; rsi != NULL; rsi = rsi->next)
{
++rsiCount;
if (rsi->locusLinkId == 0)
++noLocCount;
if (rsi->proteinAcc == NULL)
++noProtCount;
fprintf(refLinkTab, "%s\t%s\t%s\t%s\t%u\t%u\t%u\t%u\n",
emptyForNull(rsi->geneName),
emptyForNull(rsi->productName),
emptyForNull(rsi->mrnaAcc),
emptyForNull(rsi->proteinAcc),
rsi->geneNameId, rsi->productNameId,
rsi->locusLinkId, rsi->omimId);
}
if (noLocCount)
printf("Missing locusLinkIds for %d of %d\n", noLocCount, rsiCount);
if (noProtCount)
printf("Missing protein accessions for %d of %d\n", noProtCount, rsiCount);
/* Process alignments and write them out as genes. */
lf = pslFileOpen(pslFile);
dotMod = 0;
while ((psl = pslNext(lf)) != NULL)
{
if (hashFindVal(rsiHash, psl->qName) != NULL)
{
if (clDots > 0 && ++dotMod == clDots )
{
dotMod = 0;
dotOut();
}
sqlSafefFrag(cond_str, sizeof cond_str, "extAC='%s'", psl->qName);
answer = sqlGetField(proteinDB, "spXref2", "displayID", cond_str);
if (answer == NULL)
{
fprintf(stderr, "%s NOT FOUND.\n", psl->qName);
fflush(stderr);
}
if (answer != NULL)
{
struct genePred *gp = NULL;
exonList = pslToExonList(psl);
fprintf(kgTab, "%s\t%s\t%c\t%d\t%d\t",
psl->qName, psl->tName, psl->strand[0], psl->tStart, psl->tEnd);
rsi = hashMustFindVal(rsiHash, psl->qName);
gp = genePredFromPsl(psl, rsi->cdsStart, rsi->cdsEnd, genePredStdInsertMergeSize);
if (!gp)
errAbort("Cannot convert psl (%s) to genePred.\n", psl->qName);
fprintf(kgTab, "%d\t%d\t", gp->cdsStart, gp->cdsEnd);
fprintf(kgTab, "%d\t", slCount(exonList));
fflush(kgTab);
for (exon = exonList; exon != NULL; exon = exon->next)
fprintf(kgTab, "%d,", exon->start);
fprintf(kgTab, "\t");
for (exon = exonList; exon != NULL; exon = exon->next)
fprintf(kgTab, "%d,", exon->end);
fprintf(kgTab, "\n");
slFreeList(&exonList);
}
}
else
{
fprintf(stderr, "%s found in psl, but not in .fa or .ra data files.\n", psl->qName);
fflush(stderr);
}
}
if (clDots) printf("\n");
if (!clTest)
{
writeSeqTable(pepFile, refPepTab, FALSE, TRUE);
writeSeqTable(faFile, refMrnaTab, FALSE, FALSE);
}
carefulClose(&kgTab);
carefulClose(&productTab);
carefulClose(&geneTab);
carefulClose(&refLinkTab);
carefulClose(&refPepTab);
carefulClose(&refMrnaTab);
if (!clTest)
{
printf("Loading database with %s\n", kgName);
fflush(stdout);
hgLoadTabFile(conn, ".", kgName, NULL);
printf("Loading database with %s\n", "productName");
fflush(stdout);
hgLoadTabFile(conn, ".", "productName", NULL);
printf("Loading database with %s\n", "geneName");
fflush(stdout);
hgLoadTabFile(conn, ".", "geneName", NULL);
printf("Loading database with %s\n", "refLink");
fflush(stdout);
hgLoadTabFile(conn, ".", "refLink", NULL);
printf("Loading database with %s\n", "refPep");
fflush(stdout);
hgLoadTabFile(conn, ".", "refPep", NULL);
printf("Loading database with %s\n", "refMrna");
fflush(stdout);
hgLoadTabFile(conn, ".", "refMrna", NULL);
}
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2, *conn3;
char query2[256], query3[256];
struct sqlResult *sr2, *sr3;
char **row2, **row3;
char condStr[255];
char *answer;
char *kgTempDb;
char *outfileName;
FILE *outf;
int i;
char *chp;
char *acc2;
char *name, *txStart, *txEnd;
char *chrom;
char *acc, *stat;
char *frame, *start, *stop;
char *causes;
char *genomeDb;
char *geneName;
char srcType;
int alignCnt = 0;
char *candTable, *chkTable;
int orfStop, cdsGap, cdsSplice, numCdsIntrons;
boolean passed;
float ranking;
if (argc != 6) usage();
kgTempDb = argv[1];
genomeDb = argv[2];
candTable = argv[3];
chkTable = argv[4];
outfileName = argv[5];
outf = mustOpen(outfileName, "w");
conn = hAllocConn(genomeDb);
conn2= hAllocConn(genomeDb);
conn3= hAllocConn(genomeDb);
/* go through each protein */
safef(query2, sizeof(query2), "select * from %s.%s", kgTempDb, candTable);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
name = row2[0];
chrom = row2[1];
txStart = row2[3];
txEnd = row2[4];
/* retrieve gene-check results */
safef(query3, sizeof(query3),
"select * from %s.%s where acc='%s' and chrStart=%s and chrEnd = %s",
kgTempDb, chkTable, name, txStart, txEnd);
sr3 = sqlMustGetResult(conn3, query3);
row3 = sqlNextRow(sr3);
{
passed = FALSE;
ranking = 3;
acc = row3[0];
stat = row3[5];
frame = row3[6];
start = row3[7];
stop = row3[8];
orfStop = atoi(row3[9]);
cdsGap = atoi(row3[10]);
cdsSplice = atoi(row3[12]);
numCdsIntrons = atoi(row3[18]);
causes = row3[21];
ranking = 9;
/* all genes passed gene-check with status ok are considered good */
if (sameWord(stat, "ok"))
{
passed = TRUE;
ranking = 1;
}
else
{
/* frame, start, orfStop, and stop conditions must be met for KG candidates */
if ((sameWord(frame, "ok")) &&
(sameWord(start, "ok")) &&
(orfStop == 0) &&
(sameWord(stop, "ok")) )
{
ranking = 2;
/* accept cdsSplice = 0 or (cdsSplice = 1 and numCdsIntrons > 1) */
if ((cdsSplice == 0) || ((numCdsIntrons > 1) && (cdsSplice == 1)))
{
passed = TRUE;
}
/* if cdsGap > 0, degrade it ranking by 1. If cdsGap is not
a multiple of 3, degrade its ranking further */
if (cdsGap > 0)
{
ranking = ranking + 1;
if ((cdsGap - (cdsGap/3)*3) != 0) ranking = ranking + 1;
}
}
}
/* give RefSeq entries 0.5 advantage in its ranking */
safef(condStr, sizeof(condStr), "name='%s'", acc);
answer = sqlGetField(genomeDb, "refGene", "name", condStr);
if (answer != NULL)
{
ranking = ranking - 0.5;
}
else
{
chp = strstr(acc, "_");
if (chp != NULL)
{
acc2 = chp + 1;
}
else
{
acc2 = acc;
}
safef(condStr, sizeof(condStr), "name='%s'", acc2);
/* If it is an MGC gene, give it a 0.3 advantable */
answer = sqlGetField(genomeDb, "mgcGenes", "name", condStr);
if (answer != NULL)
{
ranking = ranking - 0.3;
}
}
/* print out entries, with their rankings, that passed the above criteria */
if (passed)
{
/*for (i=0; i<10; i++)
{
fprintf(outf, "%s\t", row2[i]);
}
*/
geneName = strdup(row2[0]);
chp = strstr(geneName, "_");
if (chp != NULL)
{
if (strstr(geneName, "NM_") != NULL)
{
srcType = 'R'; /* src is RefSeq */
}
else
{
chp++;
/* keep the composite name, so that kgGetCds can process correctly */
/* geneName = chp; */
srcType = 'U'; /* src is UCSC prot/mrna alignment */
}
}
else
{
srcType = 'G'; /* src is GenBank */
}
alignCnt++;
fprintf(outf, "%s\t", geneName);
for (i= 1; i<10; i++) fprintf(outf, "%s\t", row2[i]);
fprintf(outf, "%c%d\t", srcType, alignCnt);
fprintf(outf, "%.2f\n", ranking);
}
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn);
hFreeConn(&conn2);
hFreeConn(&conn3);
fclose(outf);
return(0);
}
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);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2;
struct sqlResult *sr2;
char query2[256];
char **row2;
char condStr[255];
char *uniProtDb;
char *score;
FILE *outf;
FILE *dupOutf;
char *chrom, *cdsStart, *cdsEnd;
char *displayID;
char *oldDisplayID;
char *chp, *chp1;
int i;
int isDuplicate;
char *kgTempDb;
char *infileName, *outfileName, *dupOutfileName;
if (argc != 6) usage();
kgTempDb = argv[1];
uniProtDb = argv[2];
infileName = argv[3];
outfileName = argv[4];
dupOutfileName = argv[5];
inf = mustOpen(infileName, "r");
outf = mustOpen(outfileName, "w");
dupOutf = mustOpen(dupOutfileName, "w");
conn = hAllocConn();
conn2= hAllocConn();
strcpy(oldInfo, "");
isDuplicate = 0;
oldMrnaStr = cloneString("");
oldAlignStr = cloneString("");
oldProteinStr = cloneString("");
oldDisplayID = cloneString("");
mrnaStr = cloneString("");
proteinStr = cloneString("");
while (fgets(line_in, 500, inf) != NULL)
{
strcpy(line, line_in);
strcpy(line2, line_in);
chp = strstr(line, "\t");
*chp = '\0';
mrnaStr = strdup(line);
chp ++;
chp1 = chp;
chp = strstr(chp, "\t");
*chp = '\0';
chrom = strdup(chp1);
chp ++;
chp1 = chp;
chp = strstr(chp, "\t");
*chp = '\0';
cdsStart = strdup(chp1);
chp ++;
chp1 = chp;
chp = strstr(chp, "\t");
*chp = '\0';
cdsEnd = strdup(chp1);
chp1 = line2 + (chp - line);
*chp1 = '\0';
chp ++;
chp1 = chp;
chp = strstr(chp, "\t");
*chp = '\0';
score= strdup(chp1);
chp ++;
chp1 = chp;
chp = strstr(chp, "\n");
*chp = '\0';
proteinStr= strdup(chp1);
strcpy(newInfo, line2);
if (sameString(oldInfo, newInfo))
{
isDuplicate = 1;
sqlSafefFrag(condStr, sizeof(condStr), "acc='%s'", proteinStr);
displayID = sqlGetField(uniProtDb, "displayId", "val", condStr);
if (displayID == NULL)
{
printf("!!! %s not found\n", proteinStr);fflush(stdout);
}
sqlSafefFrag(condStr, sizeof(condStr), "acc='%s'", oldProteinStr);
oldDisplayID = sqlGetField(uniProtDb, "displayId", "val", condStr);
if (oldDisplayID == NULL)
{
printf("!!! %s not found\n", oldProteinStr);fflush(stdout);
}
fprintf(dupOutf,
"%s\t%s\t%s\t%s\n", oldMrnaStr, oldDisplayID, mrnaStr, displayID);fflush(stdout);
}
else
{
/* remember previous record as old only if it is not a duplicate */
if (!isDuplicate)
{
oldMrnaStr = mrnaStr;
oldProteinStr = proteinStr;
}
strcpy(oldInfo, newInfo);
isDuplicate = 0;
sqlSafef(query2, sizeof(query2),
"select * from %s.kgCandidate2 where name='%s' and proteinID='%s' and chrom='%s' and cdsStart='%s' and cdsEnd='%s'",
kgTempDb, mrnaStr, proteinStr, chrom, cdsStart, cdsEnd);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
for (i=0; i<10; i++) fprintf(outf, "%s\t", row2[i]);
if (!sameWord(proteinStr, row2[10]))
{
printf("\n??? %s\t%s\n", proteinStr, row2[10]);fflush(stdout);
}
sqlSafefFrag(condStr, sizeof(condStr), "acc='%s'", proteinStr);
displayID = sqlGetField(uniProtDb, "displayId", "val", condStr);
if (displayID == NULL)
{
printf("!!! %s not found\n", proteinStr);fflush(stdout);
}
fprintf(outf, "%s\t", displayID);
fprintf(outf, "%s\n", row2[11]);
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
}
}
fclose(inf);
fclose(outf);
fclose(dupOutf);
return(0);
}
int main(int argc, char *argv[])
{
FILE *inf;
char *mrnaDate;
int months;
char dirName[PATH_MAX];
struct sqlConnection *conn, *conn3;
char query[256];
struct sqlResult *sr;
char **row;
char *protAcc, *mrnaAcc, *matchStr;
char *protSizeStr, *mrnaSizeStr;
int protSize, mrnaSize, match;
char *protMrnaTableName;
char condStr[255];
int score;
if (argc != 5) usage();
proteinDataDate = argv[1];
kgTempDb = argv[2];
genomeReadOnly = argv[3];
protMrnaTableName = argv[4];
sprintf(spDB, "sp%s", proteinDataDate);
sprintf(proteinsDB, "proteins%s", proteinDataDate);
sprintf(gbTempDB, "%sTemp", kgTempDb);
inf = fopen("protein.lis", "r");
if ((FILE *) NULL == inf)
errAbort("ERROR: Can not open input file: protein.lis");
o3 = fopen("kgBestMrna.out", "w");
if ((FILE *) NULL == o3)
errAbort("ERROR: Can not open output file: kgBestMrna.out");
o7 = fopen("best.lis", "w");
if ((FILE *) NULL == o7)
errAbort("ERROR: Can not open output file: best.lis");
conn = hAllocConn(genomeReadOnly);
conn3= hAllocConn(genomeReadOnly);
proteinCount = 0;
snprintf(dirName, (size_t) sizeof(dirName), "%s", "./clusterRun" );
sqlSafef(query, sizeof query,"select qName, tName, matches, qSize, tSize from %s.%s", kgTempDb, protMrnaTableName);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
while (row != NULL)
{
protAcc = row[0];
mrnaAcc = row[1];
matchStr = row[2];
protSizeStr = row[3];
mrnaSizeStr = row[4];
sscanf(matchStr, "%d", &match);
sscanf(protSizeStr, "%d", &protSize);
sscanf(mrnaSizeStr, "%d", &mrnaSize);
sscanf(matchStr, "%d", &match);
if ((float)match/(float)protSize > 0.3)
{
sqlSafefFrag(cond_str, sizeof cond_str, "acc='%s'", mrnaAcc);
mrnaDate = sqlGetField(genomeReadOnly, "gbCdnaInfo", "moddate",
condStr);
if (mrnaDate != NULL)
{
months = cal_months(mrnaDate);
score = mrnaSize + months*2 - (protSize - match) *50;
printf("%s\t%s\t%d\n", protAcc, mrnaAcc, score);fflush(stdout);
}
}
row = sqlNextRow(sr);
}
hFreeConn(&conn);
hFreeConn(&conn3);
fclose(o3);
fclose(o7);
return(0);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn2;
char condStr[500];
FILE *inf;
FILE *outf;
char line[1000];
char *chrStart;
char *inFileName, *outFileName;
char contig[100], start[100], end[100];
char num[100], code[100], id[100], oStart[100], oEnd[100], strnd[100];
char *database;
char *oldContig;
int oldNum = 0;
int lastNum = 0;
int lastEnd = 0;
if (argc != 4)usage();
database = argv[1];
inFileName = argv[2];
outFileName = argv[3];
hSetDb(database);
outf = fopen(outFileName, "w");
conn2= hAllocConn();
inf = mustOpen(inFileName, "r");
oldContig = strdup("");
while (fgets(line, 1000, inf) != NULL)
{
sscanf(line, "%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n",
contig, start, end, num, code, id, oStart, oEnd, strnd);
sqlSafefFrag(condStr, sizeof condStr, "ctg_acc='%s'", contig);
chrStart = sqlGetField(database, "seq_contig", "chr_start", condStr);
if (!sameWord(oldContig, contig))
{
if (!sameWord(oldContig, ""))
{
lastNum++;
fprintf(outf, "%s\t%d\t%d\t",
oldContig, lastEnd+1, atoi(chrStart)+atoi(start)-2);
fprintf(outf, "%d\t%s\t%d\t%s\t%s\n",
lastNum, "N", atoi(chrStart)+atoi(start)-2 - (lastEnd+1) +1, "contig", "no");
}
oldContig = strdup(contig);
oldNum = lastNum;
}
lastNum = atoi(num) + oldNum;
fprintf(outf, "%s\t%d\t%d\t",
contig, atoi(chrStart)+atoi(start)-1, atoi(chrStart)+atoi(end)-1);
fprintf(outf, "%d\t%s\t%s\t%s\t%s\t%s\n",
lastNum, code, id, oStart, oEnd, strnd);
lastEnd = atoi(chrStart)+atoi(end)-1;
}
hFreeConn(&conn2);
fclose(outf);
return(0);
}
