void doRnaFoldDisplay(struct sqlConnection *conn, char *geneId, char *geneName)
/* Show RNA folding somehow. */
{
char *table = cartString(cart, hggMrnaFoldRegion);
char *how = cartString(cart, hggDoRnaFoldDisplay);
struct rnaFold *fold = loadFold(conn, table, geneId);
if (fold == NULL)
{
warn("Couldn't load %s from %s", geneId, table);
return;
}
if (sameString(how, "text"))
{
hPrintf("<TT><PRE>");
hPrintf("%s\n%s (%1.2f)\n", fold->seq, fold->fold, fold->energy);
hPrintf("</PRE></TT>");
}
else if (sameString(how, "picture"))
{
char *psFile = cartString(cart, hggMrnaFoldPs);
char *rootName = cloneString(psFile);
char pngName[256];
char pdfName[256];
chopSuffix(rootName);
safef(pngName, sizeof(pngName), "%s.png", rootName);
safef(pdfName, sizeof(pngName), "%s.pdf", rootName);
hPrintf("<H2>%s (%s) %s energy %1.2f</H2>\n",
geneName, geneId, table, fold->energy);
if (!fileExists(pdfName))
{
char command[512];
safef(command, sizeof(command), "ps2pdf %s %s" , psFile, pdfName);
mustSystem(command);
}
hPrintf("Click <A HREF=\"%s\">here for PDF version</A><BR>", pdfName);
if (!fileExists(pngName))
{
char command[512];
safef(command, sizeof(command),
"gs -sDEVICE=png16m -sOutputFile=%s -dBATCH -dNOPAUSE -q %s"
, pngName, psFile);
mustSystem(command);
}
hPrintf("<IMG SRC=\"%s\">", pngName);
}
}
void makeTmpSai(struct sqlConnection *conn, struct cdwValidFile *vf, char *genoFile,
char **retSampleFile, char **retSaiFile)
/* Given a fastq file, make a subsample of it 100k reads long and align it with
* bwa producing a sai file of given name. */
{
/* Get fastq record */
long long fileId = vf->fileId;
struct cdwFastqFile *fqf = cdwFastqFileFromFileId(conn, fileId);
if (fqf == NULL)
errAbort("No cdwFastqFile record for file id %lld", fileId);
/* Create downsampled fastq in temp directory - downsampled more than default even. */
char sampleFastqName[PATH_LEN];
cdwMakeTempFastqSample(fqf->sampleFileName, FASTQ_SAMPLE_SIZE, sampleFastqName);
verbose(1, "downsampled %s into %s\n", vf->licensePlate, sampleFastqName);
/* Do alignment */
char cmd[3*PATH_LEN];
char *saiName = cloneString(rTempName(cdwTempDir(), "cdwPairSample", ".sai"));
safef(cmd, sizeof(cmd), "bwa aln -t 3 %s %s > %s", genoFile, sampleFastqName, saiName);
mustSystem(cmd);
/* Save return variables, clean up, and go home. */
*retSampleFile = cloneString(sampleFastqName);
*retSaiFile = saiName;
cdwFastqFileFree(&fqf);
}
void fastqRepeatQa(struct sqlConnection *conn, struct cdwFile *ef, struct cdwValidFile *vf)
/* Do repeat QA if possible on fastq file. */
{
/* First see if total repeat content is already in our table, in which case we are done. */
long long fileId = ef->id;
char query[512];
sqlSafef(query, sizeof(query),
"select count(*) from cdwQaRepeat where fileId=%lld and repeatClass='total'" , fileId);
if (sqlQuickNum(conn, query) != 0)
return; /* We've done this already */
/* Get sample file name from fastq table. */
struct cdwFastqFile *fqf = cdwFastqFileForFileId(conn, fileId);
if (fqf == NULL)
errAbort("No edqFastqRecord for %s", vf->licensePlate);
char *fastqPath = fqf->sampleFileName;
char bwaIndex[PATH_LEN];
safef(bwaIndex, sizeof(bwaIndex), "%s%s/repeatMasker/repeatMasker.fa",
cdwValDataDir, vf->ucscDb);
char cmd[3*PATH_LEN];
char *saiName = cloneString(rTempName(cdwTempDir(), "cdwQaRepeat", ".sai"));
safef(cmd, sizeof(cmd), "bwa aln %s %s > %s", bwaIndex, fastqPath, saiName);
mustSystem(cmd);
char *samName = cloneString(rTempName(cdwTempDir(), "cdwQaRepeat", ".sam"));
safef(cmd, sizeof(cmd), "bwa samse %s %s %s > %s", bwaIndex, saiName, fastqPath, samName);
mustSystem(cmd);
remove(saiName);
char *raName = cloneString(rTempName(cdwTempDir(), "cdwQaRepeat", ".ra"));
safef(cmd, sizeof(cmd), "edwSamRepeatAnalysis %s %s", samName, raName);
mustSystem(cmd);
verbose(2, "mustSystem(%s)\n", cmd);
remove(samName);
raIntoCdwRepeatQa(raName, conn, fileId);
remove(raName);
#ifdef SOON
#endif /* SOON */
freez(&saiName);
freez(&samName);
freez(&raName);
cdwFastqFileFree(&fqf);
}
void doSystem(char *command)
/* Do system call if doReal is set, otherwise just print command. */
{
printf("%s\n", command);
if (doReal)
{
mustSystem(command);
}
}
void edwAlignFastqMakeBed(struct edwFile *ef, struct edwAssembly *assembly,
char *fastqPath, struct edwValidFile *vf, FILE *bedF,
double *retMapRatio, double *retDepth, double *retSampleCoverage)
/* Take a sample fastq and run bwa on it, and then convert that file to a bed.
* bedF and all the ret parameters can be NULL. */
{
/* Hmm, tried doing this with Mark's pipeline code, but somehow it would be flaky the
* second time it was run in same app. Resorting therefore to temp files. */
char genoFile[PATH_LEN];
safef(genoFile, sizeof(genoFile), "%s%s/bwaData/%s.fa",
edwValDataDir, assembly->ucscDb, assembly->ucscDb);
char cmd[3*PATH_LEN];
char *saiName = cloneString(rTempName(edwTempDir(), "edwSample1", ".sai"));
safef(cmd, sizeof(cmd), "bwa aln -t 3 %s %s > %s", genoFile, fastqPath, saiName);
mustSystem(cmd);
char *samName = cloneString(rTempName(edwTempDir(), "ewdSample1", ".sam"));
safef(cmd, sizeof(cmd), "bwa samse %s %s %s > %s", genoFile, saiName, fastqPath, samName);
mustSystem(cmd);
remove(saiName);
/* Scan sam file to calculate vf->mapRatio, vf->sampleCoverage and vf->depth.
* and also to produce little bed file for enrichment step. */
struct genomeRangeTree *grt = genomeRangeTreeNew();
long long hitCount=0, missCount=0, totalBasesInHits=0;
scanSam(samName, bedF, grt, &hitCount, &missCount, &totalBasesInHits);
verbose(1, "hitCount=%lld, missCount=%lld, totalBasesInHits=%lld, grt=%p\n",
hitCount, missCount, totalBasesInHits, grt);
if (retMapRatio)
*retMapRatio = (double)hitCount/(hitCount+missCount);
if (retDepth)
*retDepth = (double)totalBasesInHits/assembly->baseCount
* (double)vf->itemCount/vf->sampleCount;
long long basesHitBySample = genomeRangeTreeSumRanges(grt);
if (retSampleCoverage)
*retSampleCoverage = (double)basesHitBySample/assembly->baseCount;
genomeRangeTreeFree(&grt);
remove(samName);
}
void chainSplit(char *outDir, int inCount, char *inFiles[])
/* chainSplit - Split chains up by target or query sequence. */
{
struct hash *hash = newHash(0);
int inIx;
char tpath[512];
FILE *meta ;
bool metaOpen = TRUE;
makeDir(outDir);
safef(tpath, sizeof(tpath), "%s/meta.tmp", outDir);
meta = mustOpen(tpath,"w");
for (inIx = 0; inIx < inCount; ++inIx)
{
struct lineFile *lf = lineFileOpen(inFiles[inIx], TRUE);
struct chain *chain;
FILE *f;
lineFileSetMetaDataOutput(lf, meta);
while ((chain = chainRead(lf)) != NULL)
{
char *name = (splitOnQ ? chain->qName : chain->tName);
if (lump > 0)
name = lumpName(name);
if ((f = hashFindVal(hash, name)) == NULL)
{
char path[512], cmd[512];
safef(path, sizeof(path),"%s/%s.chain", outDir, name);
if (metaOpen)
fclose(meta);
metaOpen = FALSE;
safef(cmd,sizeof(cmd), "cat %s | sort -u > %s", tpath, path);
mustSystem(cmd);
f = mustOpen(path, "a");
hashAdd(hash, name, f);
}
chainWrite(chain, f);
chainFree(&chain);
}
lineFileClose(&lf);
}
}
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 *conn2, *conn3;
char query2[256];
struct sqlResult *sr2;
char **row2;
char *proteinDataDate;
FILE *o2;
char *entrez;
char *chp;
char *hgncId, *name, *symbol, *refSeqIds, *uniProt;
int j;
char *locusType;
char *refseq;
boolean gotRefseq;
if (argc != 2) usage();
proteinDataDate = argv[1];
o2 = fopen("j.dat", "w");
conn2= hAllocConn(hDefaultDb());
conn3= hAllocConn(hDefaultDb());
sprintf(query2,
"select hgncId, symbol, name, refSeqMapped, refSeqIds, uniProt, entrezMapped, locusType from proteins%s.hgnc where status not like '%cWithdrawn%c'",
proteinDataDate, '%', '%');
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
j=0;
hgncId = row2[j];j++;
symbol = row2[j];j++;
name = row2[j];j++;
refseq = row2[j];j++;
refSeqIds = row2[j];j++;
uniProt = row2[j];j++;
entrez = row2[j];j++;
locusType = row2[j];j++;
chp = strstr(hgncId, "HGNC:");
hgncId = chp+5;
gotRefseq = FALSE;
/* process refSeqMapped first */
if (!sameWord(refseq, ""))
{
fprintf(o2, "%s\t%s\t%s\t%s\t%s\t%s\n", symbol, refseq, uniProt, hgncId, entrez, name);
gotRefseq = TRUE;
}
/* process refSeqIds next */
chp = strstr(refSeqIds, ",");
if (chp != NULL)
{
*chp = '\0';
while (chp != NULL)
{
fprintf(o2, "%s\t%s\t%s\t%s\t%s\t%s\n", symbol, refseq, uniProt, hgncId, entrez, name);
chp++;
while (*chp == ' ') chp++;
refseq = chp;
chp = strstr(refseq, ",");
if (chp != NULL) *chp = '\0';
}
fprintf(o2, "%s\t%s\t%s\t%s\t%s\t%s\n", symbol, refseq, uniProt, hgncId, entrez, name);
gotRefseq = TRUE;
}
else
{
if (!sameWord(refseq,""))
{
fprintf(o2, "%s\t%s\t%s\t%s\t%s\t%s\n", symbol, refseq, uniProt, hgncId, entrez, name);
}
else
{
/* output the record if no RefSeq in either refSeqIds or refSeqMapped */
if (!gotRefseq)
{
fprintf(o2,
"%s\t%s\t%s\t%s\t%s\t%s\n", symbol, refseq, uniProt, hgncId, entrez, name);
}
}
}
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn2);
fclose(o2);
mustSystem("cat j.dat |sort -u >hgncXref.tab");
return(0);
}
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);
}
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);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn2, *conn3;
char query2[256], query3[256];
struct sqlResult *sr2, *sr3;
char **row2, **row3;
char *accession;
char *displayID;
char *division;
char *extDB;
char *extAC;
char *proteinDataDate;
char *genomeRelease;
FILE *o2, *o3;
char *name, *chrom, *strand, *txStart, *txEnd, *cdsStart, *cdsEnd,
*exonCount, *exonStarts, *exonEnds;
char *bioDBID, *bioentryID;
if (argc != 3) usage();
proteinDataDate = argv[1];
genomeRelease = argv[2];
o2 = fopen("jj.dat", "w");
o3 = fopen("j.dat", "w");
conn2= hAllocConn(hDefaultDb());
conn3= hAllocConn(hDefaultDb());
sqlSafef(query2, sizeof query2, "select * from %sTemp.refGene;", genomeRelease);
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];
sqlSafef(query3, sizeof query3, "select * from proteins%s.spXref2 where extAC='%s' and extDB='EMBL';",
proteinDataDate, name);
sr3 = sqlMustGetResult(conn3, query3);
row3 = sqlNextRow(sr3);
while (row3 != NULL)
{
accession = row3[0];
displayID = row3[1];
division = row3[2];
extDB = row3[3];
extAC = row3[4];
bioentryID= row3[5];
bioDBID = row3[6];
if (! ( (strcmp(bioDBID, "1") == 0) ||
(strcmp(bioDBID, "2") == 0) ||
(strcmp(bioDBID, "3") == 0)
)
)
{
printf("non-recognized bioDB index %s encountered.\n", bioDBID);
printf("displayId=%s bioDBID=%s\n", displayID, bioDBID);
fflush(stdout);
exit(1);
}
fprintf(o2, "%s\n", displayID);
fprintf(o3, "%s\t%s\n", displayID, extAC);
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn2);
hFreeConn(&conn3);
fclose(o2);
fclose(o3);
mustSystem("cat j.dat |sort|uniq >proteinMrna.tab");
mustSystem("cat jj.dat|sort|uniq >protein.lis");
mustSystem("rm j.dat");
mustSystem("rm jj.dat");
return(0);
}
void pairedEndQa(struct sqlConnection *conn, struct cdwFile *ef, struct cdwValidFile *vf)
/* Look for other end, do a pairwise alignment, and save results in database. */
{
verbose(2, "pairedEndQa on %u %s %s\n", ef->id, ef->cdwFileName, ef->submitFileName);
/* Get other end, return if not found. */
struct cdwValidFile *otherVf = cdwOppositePairedEnd(conn, ef, vf);
if (otherVf == NULL)
return;
if (otherVf->fileId > vf->fileId)
return;
struct cdwValidFile *vf1, *vf2;
struct cdwQaPairedEndFastq *pair = cdwQaPairedEndFastqFromVfs(conn, vf, otherVf, &vf1, &vf2);
if (pair != NULL)
{
cdwValidFileFree(&otherVf);
return;
}
/* Get target assembly and figure out path for BWA index. */
struct cdwAssembly *assembly = cdwAssemblyForUcscDb(conn, vf->ucscDb);
assert(assembly != NULL);
char genoFile[PATH_LEN];
safef(genoFile, sizeof(genoFile), "%s%s/bwaData/%s.fa",
cdwValDataDir, assembly->ucscDb, assembly->ucscDb);
verbose(1, "aligning subsamples on %u vs. %u paired reads\n", vf1->fileId, vf2->fileId);
/* Make alignments of subsamples. */
char *sample1 = NULL, *sample2 = NULL, *sai1 = NULL, *sai2 = NULL;
makeTmpSai(conn, vf1, genoFile, &sample1, &sai1);
makeTmpSai(conn, vf2, genoFile, &sample2, &sai2);
/* Make paired end alignment */
char *tmpSam = cloneString(rTempName(cdwTempDir(), "cdwPairSample", ".sam"));
char command[6*PATH_LEN];
safef(command, sizeof(command),
"bwa sampe -n 1 -N 1 -f %s %s %s %s %s %s"
, tmpSam, genoFile, sai1, sai2, sample1, sample2);
mustSystem(command);
/* Make ra file with pairing statistics */
char *tmpRa = cloneString(rTempName(cdwTempDir(), "cdwPairSample", ".ra"));
safef(command, sizeof(command),
"edwSamPairedEndStats -maxInsert=%d %s %s", maxInsert, tmpSam, tmpRa);
mustSystem(command);
/* Read RA file into variables. */
struct cdwQaPairedEndFastq *pe = cdwQaPairedEndFastqOneFromRa(tmpRa);
/* Update database with record. */
struct sqlConnection *freshConn = cdwConnectReadWrite();
char query[256];
sqlSafef(query, sizeof(query),
"insert into cdwQaPairedEndFastq "
"(fileId1,fileId2,concordance,distanceMean,distanceStd,distanceMin,distanceMax,recordComplete) "
" values (%u,%u,%g,%g,%g,%g,%g,1)"
, vf1->fileId, vf2->fileId, pe->concordance, pe->distanceMean
, pe->distanceStd, pe->distanceMin, pe->distanceMax);
sqlUpdate(conn, query);
sqlDisconnect(&freshConn);
/* Clean up and go home. */
cdwValidFileFree(&otherVf);
remove(sample1);
remove(sample2);
remove(sai1);
remove(sai2);
remove(tmpSam);
remove(tmpRa);
#ifdef SOON
#endif /* SOON */
freez(&sample1);
freez(&sample2);
freez(&sai1);
freez(&sai2);
freez(&tmpSam);
freez(&tmpRa);
cdwQaPairedEndFastqFree(&pe);
cdwValidFileFree(&otherVf);
}
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);
}
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);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2, *conn3;
struct sqlConnection *connCentral = hConnectCentral();
char query[256], query2[256], query3[256];
struct sqlResult *sr, *sr2;
char **row, **row2;
char buf[128];
char *answer;
char *kgID, *chrom, *txStart, *txEnd;
char *mRNA;
int i;
int geneCnt = 0;
int pageNum = 0;
int topLevel = 1;
char *geneSymbol, *proteinID, *spID, *desc;
FILE *outf, *outf2;
char fileName[255];
database = strdup("hg17");
boolean newPage;
int totalKgId, totalKgCnt;
int totalKgPage;
int kgIdCnt = 0;
if (argc != 2) usage();
database = argv[1];
sqlSafef(query, sizeof query, "select genome from dbDb where name = '%s'", database);
answer = sqlQuickQuery(connCentral, query, buf, sizeof(buf));
if (answer == NULL)
{
fprintf(stderr,"'%s' is not a valid genome database name.", database);
exit(1);
}
else
{
genome = strdup(answer);
}
if (!hTableExists(database, "knownGene"))
{
fprintf(stderr,"Database %s currently does not have UCSC Known Genes.", database);
exit(1);
}
sqlSafef(query, sizeof query, "select description from dbDb where name = '%s'", database);
genomeDesc = strdup(sqlQuickQuery(connCentral, query, buf, sizeof(buf)));
hDisconnectCentral(&connCentral);
/* create first top level subdirectory */
safef(command, sizeof(command), "mkdir -p knownGeneList/%s/%d", database, topLevel);
mustSystem(command);
conn = hAllocConn(database);
conn2= hAllocConn(database);
conn3= hAllocConn(database);
newPage = TRUE;
currentPage = 0;
/* put this in to avoid compiler complaining */
outf = NULL;
geneSymbol = NULL;
char *protAcc = NULL;
/* figure out how many pages in total */
sqlSafef(query2, sizeof(query2), "select count(k.name) from %s.knownGene k, %s.kgXref x where k.name=x.kgId and geneSymbol != ''", database, database);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
totalKgCnt = atoi(row2[0]);
sqlFreeResult(&sr2);
/* figure out how many KG IDs in total */
sqlSafef(query2, sizeof(query2), "select count(*) from %s.kgXref where geneSymbol !=''", database);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
totalKgId = atoi(row2[0]);
sqlFreeResult(&sr2);
totalKgPage = totalKgId/LINKSPERPAGE + 1;
sqlSafef(query2, sizeof(query2),
"select kgID, geneSymbol, description from %s.kgXref where geneSymbol!= '' order by geneSymbol",
database);
/* for debugging */
/* "select kgID, geneSymbol, description from %s.kgXref order by geneSymbol limit %d",
database, TESTSIZE);*/
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
/* for debugging */
/* while (kgIdCnt < TESTSIZE) */
while (kgIdCnt < totalKgId)
{
kgIdCnt++;
kgID = row2[0];
geneSymbol = strdup(row2[1]);
desc = row2[2];
sqlSafef(query, sizeof(query),
"select chrom,txSTart,txEnd,proteinID from %s.knownGene where name='%s'", database, kgID);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
if (row != NULL)
{
geneCnt++;
chrom = row[0];
txStart = row[1];
txEnd = row[2];
proteinID = row[3];
if (newPage)
{
/* create a KG links page */
pageNum++;
currentPage++;
/* use mkdir -p to make sure the subdirectory exists */
safef(command, sizeof(command), "mkdir -p knownGeneList/%s/%d", database, topLevel);
mustSystem(command);
safef(fileName, sizeof(fileName),
"knownGeneList/%s/%d/kgList%d.html", database, topLevel, pageNum);
outf = fopen(fileName, "w");
printHtmlHead(outf);
fprintf(outf,"<H2>UCSC %s Known Genes List (page %d of %d)</H2>\n",
genome, pageNum, totalKgPage);
fprintf(outf, "<TABLE BORDER=1=CELLSPACING=1 CELLPADDING=3 BGCOLOR=\"#D9F8E4\"><TR>\n");
fprintf(outf,
"<TR><TH>Gene Symbol</TH><TH>Known Gene ID</TH><TH>mRNA</TH><TH>UniProt</TH><TH>RefSeq Protein</TH><TH>Description</TH>\n");
strcpy(startSymbol[pageNum], geneSymbol);
strcpy(pageStartSymbol[currentPage], geneSymbol);
newPage = FALSE;
}
fprintf(outf,"<TR>");
fprintf(outf,"<TD>%s</TD>", geneSymbol);
/*fprintf(outf,"<TD>%d:%s</TD>", geneCnt, geneSymbol);*/
fprintf(outf,"<TD>");
fprintf(outf,"<A href=\"/cgi-bin/hgGene?db=%s&hgg_gene=%s", database, kgID);
fprintf(outf,"&hgg_chrom=%s&hgg_start=%s&hgg_end=%s\">", chrom, txStart, txEnd);
fprintf(outf,"%s", kgID);
fprintf(outf,"</A>");
fprintf(outf,"</TD>\n");
sqlSafef(query3,sizeof(query3),"select spID from %s.kgXref where kgID = '%s'", database, kgID);
spID = cloneString(sqlQuickQuery(conn3, query3, buf, sizeof(buf)));
if (spID == NULL)
{
spID = emptyString;
}
else
{
if (sameWord(spID,"")) spID = emptyString;
}
sqlSafef(query3,sizeof(query3),"select mRNA from %s.kgXref where kgID = '%s'", database, kgID);
mRNA = cloneString(sqlQuickQuery(conn3, query3, buf, sizeof(buf)));
if (mRNA == NULL)
{
mRNA = emptyString;
}
else
{
if (sameWord(mRNA,"")) mRNA = emptyString;
}
sqlSafef(query3,sizeof(query3),"select protAcc from %s.kgXref where kgID = '%s'", database, kgID);
protAcc = sqlQuickQuery(conn3, query3, buf, sizeof(buf));
if (protAcc == NULL)
{
protAcc = emptyString;
}
else
{
if (sameWord(protAcc,"")) protAcc = emptyString;
}
fprintf(outf,"<TD>%s</TD>", mRNA);
fprintf(outf,"<TD>%s</TD>", spID);
fprintf(outf,"<TD>%s</TD>", protAcc);
fprintf(outf,"<TD>%s</TD>", desc );
fprintf(outf,"</TR>\n");
if ((geneCnt % LINKSPERPAGE) == 0)
{
/* flush out and close the page if a page is filled, and start a new page */
fprintf(outf,"</TABLE>");
strcpy(endSymbol[pageNum], geneSymbol);
strcpy(pageEndSymbol[currentPage], endSymbol[pageNum]);
fprintf(outf, "<BR>");
fprintf(outf, "<A href=\"/knownGeneList/%s/%d/kgIndex%d.html\">",
database, topLevel,topLevel);
fprintf(outf, "Up");
fprintf(outf,"</A><BR>\n");
printHtmlEnd(outf);
newPage = TRUE;
fclose(outf);
outf = NULL;
if ((pageNum % LINKSPERPAGE) == 0 )
{
printf("Processing topLevel %d ...\n", topLevel);fflush(stdout);
safef(fileName, sizeof(fileName),
"knownGeneList/%s/%d/kgIndex%d.html", database, topLevel, topLevel);
outf2 = fopen(fileName, "w");
printHtmlHead(outf2);
//fprintf(outf2,"<H2>UCSC %s Known Genes List</H2>\n", genome);
fprintf(outf2,"<H2>UCSC %s Known Genes List (Group %d)</H2>\n", genome, topLevel);
for (i=1; i<= currentPage; i++)
{
fprintf(outf2, "Page %d: ", (topLevel-1)*LINKSPERPAGE+i);
fprintf(outf2,
"<A href=\"/knownGeneList/%s/%d/kgList%d.html\">",
database, topLevel, (topLevel-1)*LINKSPERPAGE+i);
fprintf(outf2, "%s to %s", pageStartSymbol[i], pageEndSymbol[i]);
fprintf(outf2,"</A><BR>\n");
}
fprintf(outf2, "<BR>");
fprintf(outf2, "<A href=\"/knownGeneList/%s/top.html\">",database);
fprintf(outf2, "Up");
fprintf(outf2,"</A><BR>\n");
printHtmlEnd(outf2);
fclose(outf2);
strcpy(topStartSymbol[topLevel], pageStartSymbol[1]);
strcpy( topEndSymbol[topLevel], pageEndSymbol[currentPage]);
currentPage = 0;
topLevel++;
}
}
row = sqlNextRow(sr);
}
sqlFreeResult(&sr);
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
/* flush out and close the last list page */
if (outf != NULL)
{
fprintf(outf,"</TABLE>");
strcpy(endSymbol[pageNum], geneSymbol);
strcpy(pageEndSymbol[currentPage], endSymbol[pageNum]);
fprintf(outf, "<BR>");
fprintf(outf, "<A href=\"/knownGeneList/%s/%d/kgIndex%d.html\">",
database, topLevel,topLevel);
fprintf(outf, "Up");
fprintf(outf,"</A><BR>\n");
printHtmlEnd(outf);
fclose(outf);
}
/* generate the last index page */
safef(command, sizeof(command), "mkdir -p knownGeneList/%s/%d", database, topLevel);
mustSystem(command);
safef(fileName, sizeof(fileName),
"knownGeneList/%s/%d/kgIndex%d.html", database, topLevel, topLevel);
outf2 = fopen(fileName, "w");
printHtmlHead(outf2);
fprintf(outf2,"<H2>UCSC %s Known Genes List (Group %d)</H2>\n", genome, topLevel);
for (i=1; i<= currentPage; i++)
{
fprintf(outf2, "Page %d: ", (topLevel-1)*LINKSPERPAGE+i);
fprintf(outf2, "<A href=\"/knownGeneList/%s/%d/kgList%d.html\">",
database, topLevel, (topLevel-1)*LINKSPERPAGE+i);
fprintf(outf2, "%s to %s", pageStartSymbol[i], pageEndSymbol[i]);
fprintf(outf2,"</A><BR>\n");
fflush(outf2);
}
fprintf(outf2, "<BR>");
fprintf(outf2, "<A href=\"/knownGeneList/%s/top.html\">",database);
fprintf(outf2, "Up");
fprintf(outf2,"</A><BR>\n");
strcpy(topStartSymbol[topLevel], pageStartSymbol[1]);
strcpy( topEndSymbol[topLevel], pageEndSymbol[currentPage]);
fclose(outf2);
currentPage = 0;
/* generate the top HTML page */
safef(fileName, sizeof(fileName), "knownGeneList/%s/top.html", database);
outf2 = fopen(fileName, "w");
printHtmlHead(outf2);
fprintf(outf2,"<H2>UCSC %s Known Genes List</H2>\n", genome);
for (i=1; i<= topLevel; i++)
{
fprintf(outf2, "Group %d: ", i);
fprintf(outf2, "<A href=\"/knownGeneList/%s/%d/kgIndex%d.html\">", database, i, i);
fprintf(outf2, " %s to %s", topStartSymbol[i], topEndSymbol[i]);
fprintf(outf2,"</A><BR>\n");
fflush(outf2);
}
fprintf(outf2, "<BR>");
fprintf(outf2, "<A href=\"/knownGeneLists.html\">");
fprintf(outf2, "Up");
fprintf(outf2,"</A><BR>\n");
printHtmlEnd(outf2);
printHtmlEnd(outf2);
fclose(outf2);
return(0);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2, *conn3;
char query[256], query3[256];
struct sqlResult *sr, *sr3;
char **row, **row3;
FILE *o1, *o2;
char *locusID; /* LocusLink ID */
char *refAC; /* Refseq accession.version */
char *kgTempDbName, *roDbName;
char cond_str[200];
char *kgID;
char *mapID;
char *desc;
if (argc != 3) usage();
kgTempDbName = argv[1];
roDbName = argv[2];
conn = hAllocConn(roDbName);
conn2= hAllocConn(roDbName);
conn3= hAllocConn(roDbName);
o1 = fopen("j.dat", "w");
o2 = fopen("jj.dat", "w");
sqlSafef(query, sizeof query, "select kgID, refseq from %s.kgXref", roDbName);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
while (row != NULL)
{
kgID = row[0];
refAC = row[1];
sqlSafefFrag(cond_str, sizeof cond_str, "refseq='%s'", refAC);
locusID = sqlGetField("entrez", "entrezRefProt", "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);fflush(o1);
fprintf(o2, "%s\t%s\n", mapID, desc);
row3 = sqlNextRow(sr3);
}
sqlFreeResult(&sr3);
}
row = sqlNextRow(sr);
}
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 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;
FILE *o1;
char *chp0, *chp;
char *genomeDBname;
char refseqID[40], mapID[40];
char *kgID, *geneSymbol;
if (argc != 3) usage();
infileName = argv[1];
genomeDBname = argv[2];
conn= hAllocConn(genomeDBname);
o1 = fopen("j.dat", "w");
inf = mustOpen(infileName, "r");
while (fgets(line_in, 1000, inf) != NULL)
{
strcpy(line, line_in);
chp = strstr(line, "\t");
*chp = '\0';
strcpy(refseqID, line);
again:
chp ++;
chp0 = chp;
chp = strtok(chp, "\r\t\n");
if (chp == NULL) continue;
sprintf(cond_str, "alias='%s'", refseqID);
kgID=sqlGetField(genomeDBname, "kgAlias", "kgID", cond_str);
// check with refLink if not found in kgAlias
if (kgID == NULL)
{
sprintf(cond_str, "mrnaAcc='%s'", refseqID);
geneSymbol=sqlGetField(genomeDBname, "refLink", "name", cond_str);
sprintf(cond_str, "alias='%s'", geneSymbol);
kgID=sqlGetField(genomeDBname, "kgAlias", "kgID", cond_str);
}
strcpy(mapID, chp);
if (kgID != NULL)
{
fprintf(o1, "%s\t%s\t%s\n", kgID, refseqID, mapID);fflush(stdout);
}
else
{
printf("%s not found in kgAlias nor in refLink\n", refseqID);
}
chp = chp + strlen(mapID);
// process remaing refeqID(s)
goto again;
}
fclose(o1);
mustSystem("cat j.dat|sort|uniq >bioCycPathway.tab");
mustSystem("rm j.dat");
return 0;
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2, *conn3;
char query[256], query2[256];
struct sqlResult *sr, *sr2;
char **row, **row2;
char *chp;
FILE *o1;
char *locusID; /* LocusLink ID */
char *gbAC; /* GenBank accession.version */
char *giNCBI; /* NCBI gi for the protein record associated with the CDS */
char *seqType; /* sequence type m=mRNA g=genomic u=undefined */
char *proteinAC; /* protein accession.version */
char *taxID; /* tax id */
char *locusID2; /* 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 *dbName;
if (argc != 2) usage();
dbName = argv[1];
conn = hAllocConn(dbName);
conn2= hAllocConn(dbName);
conn3= hAllocConn(dbName);
o1 = fopen("j.dat", "w");
sqlSafef(query2, sizeof query2, "select * from %sTemp.locus2Ref0;", dbName);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
locusID2 = row2[0];
refAC = row2[1];
giNCBI2 = row2[2];
revStatus = row2[3];
proteinAC2 = row2[4];
taxID2 = row2[5];
sqlSafef(query, sizeof 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];
giNCBI = row[2];
seqType = row[3];
proteinAC = row[4];
taxID = row[5];
chp = strstr(gbAC, ".");
if (chp != NULL) *chp = '\0';
chp = strstr(refAC, ".");
if (chp != NULL) *chp = '\0';
fprintf(o1, "%s\t%s\n", gbAC, refAC);
row = sqlNextRow(sr);
}
row2 = sqlNextRow(sr2);
}
fclose(o1);
hFreeConn(&conn);
hFreeConn(&conn2);
sqlFreeResult(&sr2);
mustSystem("cat j.dat|sort|uniq >mrnaRefseq.tab");
printf("mrnaRefseq.tab created.\n");
mustSystem("rm j.dat");
return(0);
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2;
char query[256], query2[256];
struct sqlResult *sr, *sr2;
char **row, **row2;
char *chp;
char *kgID;
FILE *o2;
char cond_str[256];
char *database;
char *ro_db;
char *proteinID;
char *proteinAC;
if (argc != 3) usage();
database = cloneString(argv[1]);
ro_db = cloneString(argv[2]);
conn = hAllocConn(database);
conn2= hAllocConn(database);
o2 = fopen("jj.dat", "w");
sprintf(query2,"select name, proteinID from %s.knownGene;", database);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
kgID = row2[0];
proteinID = row2[1];
// get RefSeq protein AC numbers (NP_xxxxx) if they exist
sprintf(cond_str, "kgID='%s'", kgID);
proteinAC = sqlGetField(database, "kgXref", "protAcc", cond_str);
if (proteinAC != NULL)
{
if (strlen(proteinAC) > 0)
{
fprintf(o2, "%s\t%s\t%s\n", kgID, proteinID, proteinAC);
}
}
// get Genbank protein accession numbers
if (strstr(kgID, "NM_") != NULL)
{
sprintf(query,"select protAcc from %s.refLink where mrnaAcc = '%s';", ro_db, kgID);
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
while (row != NULL)
{
proteinAC = row[0];
fprintf(o2, "%s\t%s\t%s\n", kgID, proteinID, proteinAC);
row = sqlNextRow(sr);
}
sqlFreeResult(&sr);
}
else
{
sprintf(query,"select proteinAC from %sTemp.locus2Acc0 where gbAC like '%s%c';", database, kgID, '%');
sr = sqlMustGetResult(conn, query);
row = sqlNextRow(sr);
while (row != NULL)
{
proteinAC = row[0];
chp = strstr(proteinAC, ".");
if (chp != NULL)
{
*chp = '\0';
}
if (proteinAC[0] != '-')
{
fprintf(o2, "%s\t%s\t%s\n", kgID, proteinID, proteinAC);
}
row = sqlNextRow(sr);
}
sqlFreeResult(&sr);
}
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
fclose(o2);
hFreeConn(&conn);
hFreeConn(&conn2);
mustSystem("cat jj.dat|sort|uniq >kgProtAliasNCBI.tab");
mustSystem("rm jj.dat");
return(0);
}
void gsBig(char *faName, char *gtfName,
char *suboptName,
char *transName,
char *exeName,
char *parName,
char *tmpDirName)
/* gsBig - Run Genscan on big input and produce GTF files. */
{
struct dnaSeq seq;
struct lineFile *lf = lineFileOpen(faName, TRUE);
FILE *gtfFile = mustOpen(gtfName, "w");
FILE *subFile = NULL;
FILE *transFile = NULL;
ZeroVar(&seq);
if (suboptName != NULL)
subFile = mustOpen(suboptName, "w");
if (transName != NULL)
transFile = mustOpen(transName, "w");
if (exeName != NULL)
exePath = cloneString(exeName);
if (parName != NULL)
parPath = cloneString(parName);
if (tmpDirName != NULL)
tmpDir = cloneString(tmpDirName);
if (optionExists("prerun"))
{
char *preFileName = optionVal("prerun", NULL);
char seqName[128];
struct segment *seg = parseSegment(preFileName, 0, 100000000, seqName);
writeSeg(seqName, seg, gtfFile, subFile, transFile);
}
else
{
struct dyString *dy = newDyString(1024);
char tempFa[512], tempGs[512];
char dir1[256], root1[128], ext1[64];
int myPid = (int)getpid();
splitPath(faName, dir1, root1, ext1);
while (faSpeedReadNext(lf, &seq.dna, &seq.size, &seq.name))
{
int offset, sizeOne;
struct segment *segList = NULL, *seg;
char *seqName = cloneString(seq.name);
int chunkNum = 0;
for (offset = 0; offset < seq.size; offset += stepSize)
{
boolean allN = TRUE;
int i;
safef(tempFa, sizeof(tempFa), "%s/temp_gsBig_%d_%s_%d.fa",
tmpDir, myPid, seqName, chunkNum);
safef(tempGs, sizeof(tempGs), "%s/temp_gsBig_%d_%s_%d.genscan",
tmpDir, myPid, seqName, chunkNum);
sizeOne = seq.size - offset;
if (sizeOne > winSize) sizeOne = winSize;
/* Genscan hangs forever if a chunk is all-N's... if so,
* then skip this chunk. */
for (i=offset; i < (offset+sizeOne); i++)
{
if (seq.dna[i] != 'N' && seq.dna[i] != 'n')
{
allN = FALSE;
break;
}
}
if (allN)
{
printf("\ngsBig: skipping %s[%d:%d] -- it's all N's.\n\n",
seqName, offset, (offset+sizeOne-1));
}
else
{
faWrite(tempFa, "split", seq.dna + offset, sizeOne);
dyStringClear(dy);
dyStringPrintf(dy, "%s %s %s", exePath, parPath, tempFa);
if (suboptName != NULL)
dyStringPrintf(dy, " -subopt");
dyStringPrintf(dy, " > %s", tempGs);
verbose(3, "%s\n", dy->string);
mustSystem(dy->string);
seg = parseSegment(tempGs, offset, offset+sizeOne, NULL);
slAddHead(&segList, seg);
}
chunkNum++;
}
slReverse(&segList);
seg = mergeSegs(segList);
writeSeg(seqName, seg, gtfFile, subFile, transFile);
freez(&seqName);
}
if (! optionExists("noRemove"))
{
remove(tempFa);
remove(tempGs);
}
}
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2;
char query2[256];
struct sqlResult *sr2;
char **row2;
char cond_str[256];
char *kgID;
char *proteinID;
char *seq;
char *acc;
char protDbName[100];
char spDbName[100];
char *dbName;
char *ro_dbName;
FILE *o1, *o2;
struct dnaSeq *kgSeq;
if (argc != 4) usage();
o1 = fopen("j.dat", "w");
o2 = fopen("jj.dat", "w");
dbName = argv[1];
ro_dbName = argv[3];
sprintf(protDbName, "proteins%s", argv[2]);
sprintf(spDbName, "sp%s", argv[2]);
conn= hAllocConn(ro_dbName);
conn2= hAllocConn(ro_dbName);
sqlSafef(query2, sizeof query2, "select name from %s.knownGene;", dbName);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
while (row2 != NULL)
{
kgID = row2[0];
sqlSafefFrag(cond_str, sizeof cond_str, "name = '%s';", kgID);
seq = sqlGetField(dbName, "knownGenePep", "seq", cond_str);
if (seq != NULL)
{
fprintf(o1, "%s\t%s\n", kgID, seq);fflush(o1);
}
else
{
sqlSafefFrag(cond_str, sizeof cond_str, "name = '%s';", kgID);
proteinID=sqlGetField(dbName, "knownGene", "proteinID", cond_str);
if (proteinID != NULL)
{
sqlSafefFrag(cond_str, sizeof cond_str, "val = '%s';", proteinID);
acc = sqlGetField(spDbName, "displayId", "acc", cond_str);
if (acc == NULL)
{
fprintf(stderr, "NO acc.displayId.%s: %s from name.knownGene.%s: %s\n", spDbName, proteinID, dbName, kgID);
fflush(stderr);
}
else
{
sqlSafefFrag(cond_str, sizeof cond_str, "acc = '%s';", acc);
seq = sqlGetField(spDbName, "protein", "val", cond_str);
if (seq == NULL)
{
fprintf(stderr, "NO protein seq for %s\n", kgID);
fprintf(stderr, "proteinID.knownGene.%s: %s, acc.displayID.%s: %s\n", dbName, proteinID, spDbName, acc);
fflush(stderr);
}
else
{
fprintf(o1, "%s\t%s\n", kgID, seq);
}
}
} else {
fprintf(stderr, "kgID: %s not in knownGenePep or knownGene\n", kgID);
}
}
sqlSafefFrag(cond_str, sizeof cond_str, "name = '%s';", kgID);
seq = sqlGetField(dbName, "knownGeneMrna", "seq", cond_str);
if (seq != NULL)
{
fprintf(o2, "%s\t%s\n", kgID, seq);fflush(o1);
}
else
{
kgSeq = hGenBankGetMrna(dbName, kgID, NULL);
if (kgSeq != NULL)
{
fprintf(o2, "%s\t%s\n", kgID, kgSeq->dna);fflush(o1);
}
else
{
fprintf(stderr, "NO mRNA seq for %s\n", kgID);fflush(stderr);
}
}
row2 = sqlNextRow(sr2);
}
sqlFreeResult(&sr2);
hFreeConn(&conn);
hFreeConn(&conn2);
fclose(o1);
fclose(o2);
mustSystem("cat j.dat |sort|uniq > knownGenePep.tab");
mustSystem("cat jj.dat|sort|uniq > knownGeneMrna.tab");
mustSystem("rm j.dat");
mustSystem("rm jj.dat");
return(0);
}
