int main(int argc, char *argv[])
/* read ContigLocFilter, writing to individual chrom tables */
{
struct hashCookie cookie;
struct hashEl *hel;
char *chromName;
if (argc != 3)
usage();
snpDb = argv[1];
contigGroup = argv[2];
hSetDb(snpDb);
/* check for needed tables */
if(!hTableExistsDb(snpDb, "ContigLocFilter"))
errAbort("no ContigLocFilter table in %s\n", snpDb);
if(!hTableExistsDb(snpDb, "ContigInfo"))
errAbort("no ContigInfo table in %s\n", snpDb);
chromHash = loadChroms(contigGroup);
if (chromHash == NULL)
{
verbose(1, "couldn't get chrom info\n");
return 1;
}
writeSplitTables();
verbose(1, "closing files...\n");
cookie = hashFirst(chromHash);
while (hel = hashNext(&cookie))
fclose(hel->val);
verbose(1, "creating tables...\n");
cookie = hashFirst(chromHash);
while ((chromName = hashNextName(&cookie)) != NULL)
createTable(chromName);
verbose(1, "loading database...\n");
cookie = hashFirst(chromHash);
while ((chromName = hashNextName(&cookie)) != NULL)
{
verbose(1, "chrom = %s\n", chromName);
loadDatabase(chromName);
}
return 0;
}
char *identifierWhereClause(char *idField, struct hash *idHash)
/* If the number of pasted IDs is reasonably low, return a where-clause component for the IDs. */
{
if (idHash == NULL || idField == NULL)
return NULL;
int numIds = hashNumEntries(idHash);
int maxIdsInWhere = cartUsualInt(cart, "hgt_maxIdsInWhere", DEFAULT_MAX_IDS_IN_WHERE);
if (numIds > 0 && numIds <= maxIdsInWhere)
{
struct dyString *dy = dyStringNew(16 * numIds);
dyStringPrintf(dy, "%s in (", idField);
struct hashCookie hc = hashFirst(idHash);
boolean first = TRUE;
char *id;
while ((id = hashNextName(&hc)) != NULL)
{
if (first)
first = FALSE;
else
dyStringAppend(dy, ", ");
dyStringPrintf(dy, "'%s'", id);
}
dyStringAppend(dy, ")");
return dyStringCannibalize(&dy);
}
return NULL;
}
void doLog()
{
FILE *logFileHandle = mustOpen("snpGetSeqDup.log", "w");
struct hashCookie cookie = hashFirst(uniqHash);
char *rsId = NULL;
int count = 0;
struct hashEl *hel = NULL;
char *fileName = NULL;
struct dyString *dy = newDyString(1024);
while ((rsId = hashNextName(&cookie)) != NULL)
{
count = 0;
for (hel = hashLookup(snpHash, rsId); hel != NULL; hel = hashLookupNext(hel))
count++;
if (count == 1) continue;
for (hel = hashLookup(snpHash, rsId); hel != NULL; hel = hashLookupNext(hel))
{
fileName = (char *)hel->val;
dyStringAppend(dy, fileName);
dyStringAppend(dy, " ");
}
fprintf(logFileHandle, "%s\t%s\n", rsId, dy->string);
dyStringClear(dy);
}
carefulClose(&logFileHandle);
}
void writeResults()
/* loop through nameHash */
/* print all coords from coordHash to outputFileHandle */
/* also print count per SNP to logFileHandle */
{
struct hashCookie cookie;
struct hashEl *hel= NULL;
struct coords *cel = NULL;
int count = 0;
char *name;
verbose(1, "writing results...\n");
cookie = hashFirst(nameHash);
while ((name = hashNextName(&cookie)) != NULL)
{
count = 0;
for (hel = hashLookup(coordHash, name); hel != NULL; hel= hashLookupNext(hel))
{
cel = (struct coords *)hel->val;
fprintf(outputFileHandle, "%s\t%d\t%d\t%s\tMultipleAlignments\n", cel->chrom, cel->start, cel->end, name);
count++;
}
fprintf(logFileHandle, "%s\t%d\n", name, count);
}
}
int main(int argc, char *argv[])
{
char *snpDb = NULL;
char *snpTableName = NULL;
struct hashCookie cookie;
char *chromName = NULL;
if (argc != 3)
usage();
snpDb = argv[1];
hSetDb(snpDb);
snpTableName = argv[2];
/* check that tables exist */
if (!hTableExists(snpTableName))
errAbort("no %s table in %s\n", snpTableName, snpDb);
if (!hTableExists("chromInfo"))
errAbort("no chromInfo table in %s\n", snpDb);
loadChroms();
getSnps(snpTableName);
verbose(1, "creating tables...\n");
cookie = hashFirst(chromHash);
while ((chromName = hashNextName(&cookie)) != NULL)
createTable(chromName);
verbose(1, "loading database...\n");
cookie = hashFirst(chromHash);
while ((chromName = hashNextName(&cookie)) != NULL)
{
verbose(1, "chrom = %s\n", chromName);
loadDatabase(chromName);
}
return 0;
}
void checkIds(char *inputFileName, char *outputFileName)
/* report if duplicate ID found */
/* put all ids in idHash */
{
struct chain *chainEl;
struct lineFile *lf = lineFileOpen(inputFileName, TRUE);
FILE *outputFileHandle = NULL;
char idString[64];
char *idString2 = NULL;
struct hashEl *hel = NULL;
struct hashEl *hel2 = NULL;
int chainCount = 0;
int dupCount = 0;
struct hashCookie cookie;
idHash = newHash(0);
duplicateHash = newHash(0);
while ((chainEl = chainRead(lf)) != NULL)
{
chainCount++;
safef(idString, sizeof(idString), "%d", chainEl->id);
hel = hashLookup(idHash, idString);
if (hel == NULL)
hashAdd(idHash, cloneString(idString), NULL);
else
{
hel2 = hashLookup(duplicateHash, idString);
if (hel2 == NULL)
hashAdd(duplicateHash, cloneString(idString), NULL);
}
}
verbose(1, "chain count = %d\n", chainCount);
// freeHash(&idHash);
/* print contents of duplicateHash */
outputFileHandle = mustOpen(outputFileName, "w");
cookie = hashFirst(duplicateHash);
while ((idString2 = hashNextName(&cookie)) != NULL)
{
dupCount++;
fprintf(outputFileHandle, "%s\n", idString2);
}
verbose(1, "count of duplicate IDs = %d\n", dupCount);
carefulClose(&outputFileHandle);
// freeHash(&duplicateHash);
}
void processSnps(char *oldTableName, char *newTableName)
{
static struct hash *contigLocHash = NULL;
static struct hash *mapInfoHash = NULL;
static struct hash *oldNameHash = NULL;
static struct hash *newNameHash = NULL;
struct hashCookie cookie;
struct hashEl *helNameNew = NULL;
struct hashEl *contigLocElement = NULL;
struct hashEl *mapInfoElement = NULL;
char *name;
FILE *outputFileHandle = mustOpen("snpMissing.out", "w");
int count = 0;
verbose(1, "creating hashes...\n");
contigLocHash = getUniqueStringHash("snp_id", "ContigLoc");
mapInfoHash = getUniqueStringHash("snp_id", "MapInfo");
oldNameHash = getUniqueStringHash("name", oldTableName);
newNameHash = getUniqueStringHash("name", newTableName);
verbose(1, "writing results...\n");
cookie = hashFirst(oldNameHash);
while ((name = hashNextName(&cookie)) != NULL)
{
count++;
helNameNew = hashLookup(newNameHash, name);
if (helNameNew == NULL)
{
fprintf(outputFileHandle, "rs%s in %s but not in %s\n", name, oldTableName, newTableName);
contigLocElement = hashLookup(contigLocHash, name);
if (contigLocElement != NULL)
fprintf(outputFileHandle, "found in ContigLoc\n");
mapInfoElement = hashLookup(mapInfoHash, name);
if (mapInfoElement != NULL)
fprintf(outputFileHandle, "found in MapInfo\n");
}
if (count == 100000) break;
}
carefulClose(&outputFileHandle);
}
struct locus *readSs(char *pbFile, char *strandFile)
/* determine which allele matches assembly and store in details file */
{
struct hash *strandHash = readStrand(strandFile);
struct strand *strand = NULL;
struct hash *missingHugoIdHash = newHash(16);
struct hashCookie hashPtr;
char *missingName;
struct locus *l = NULL, *lPtr = NULL;
struct alleleInfo *aPtr = NULL;
struct lineFile *lf = lineFileOpen(pbFile, TRUE); /* input file */
char *row[4], *row2[3]; /* number of fields in input file */
char *pbName;
char chrom[32];
int chromStart;
int chromEnd;
char name[32];
char *allele;
while (lineFileRow(lf, row)) /* process one snp at a time */
{
struct alleleInfo *ai1 = NULL, *ai2 = NULL, *aiPtr;
struct locus *m = NULL;
chopString(row[0], "-", row2, 3);
chromStart = sqlUnsigned(row2[0]);
chromEnd = chromStart+1;
safef(chrom, sizeof(chrom), "chr%s", row2[2]);
if(l==NULL||l->chrom==NULL||l->chromStart!=chromStart||!(sameString(l->chrom,chrom)))
{
AllocVar(m);
safef(name, sizeof(name), "%s_%d", row2[1], ++ssnpId);
m->chrom = cloneString(chrom);
m->chromStart = chromStart;
m->chromEnd = chromEnd;
m->name = cloneString(name);
m->hugoId = cloneString(row2[1]);
m->strictSnp = TRUE;
slAddHead(&l, m);
}
allele=cloneString(row[2]);
convertToUppercase(allele);
if ( sameString(allele,"A") || sameString(allele,"C") ||
sameString(allele,"G") || sameString(allele,"T") )
{
for (aiPtr=l->alleles; aiPtr!=NULL; aiPtr=aiPtr->next)
if (sameString(aiPtr->allele, allele))
break;
if (aiPtr==NULL)
{
AllocVar(ai1);
ai1->allele=cloneString(allele);
slAddHead(&(l->alleles), ai1);
l->alleleCount++;
aiPtr=l->alleles;
}
aiPtr->count++;
l->sampleSize++;
}
allele=cloneString(row[3]);
convertToUppercase(allele);
if ( sameString(allele,"A") || sameString(allele,"C") ||
sameString(allele,"G") || sameString(allele,"T") )
{
for (aiPtr=l->alleles; aiPtr!=NULL; aiPtr=aiPtr->next)
if (sameString(aiPtr->allele, allele))
break;
if (aiPtr==NULL)
{
AllocVar(ai2);
ai2->allele=cloneString(allele);
slAddHead(&(l->alleles), ai2);
l->alleleCount++;
aiPtr=l->alleles;
}
aiPtr->count++;
l->sampleSize++;
}
}
slReverse(&l);
for(lPtr=l; lPtr!=NULL; lPtr=lPtr->next)
{
strand = hashFindVal(strandHash, lPtr->hugoId);
if (strand == NULL)
{
hashStore(missingHugoIdHash, lPtr->hugoId);
slRemoveEl(l, lPtr);
continue;
}
lPtr->strand = cloneString(strand->strand);
}
freeHash(&strandHash);
hashPtr = hashFirst(missingHugoIdHash);
while ( (missingName=hashNextName(&hashPtr)) != NULL )
printf("HUGO ID was not found in strand.txt (usually from proteome.hgncXref): %s\n", missingName);
freeHash(&missingHugoIdHash);
return l;
}
static void mafOrAxtClick2(struct sqlConnection *conn, struct sqlConnection *conn2, struct trackDb *tdb, char *axtOtherDb, char *fileName)
/* Display details for MAF or AXT tracks. */
{
hgBotDelay();
if (winEnd - winStart > 30000)
{
printf("Zoom so that window is 30,000 bases or less to see alignments and conservation statistics\n");
}
else
{
struct mafAli *mafList = NULL, *maf, *subList = NULL;
int aliIx = 0, realCount = 0;
char dbChrom[64];
char option[128];
char *capTrack;
struct consWiggle *consWig, *consWiggles;
struct hash *speciesOffHash = NULL;
char *speciesOrder = NULL;
char *speciesTarget = trackDbSetting(tdb, SPECIES_TARGET_VAR);
char buffer[1024];
int useTarg = FALSE;
int useIrowChains = FALSE;
safef(option, sizeof(option), "%s.%s", tdb->track, MAF_CHAIN_VAR);
if (cartCgiUsualBoolean(cart, option, FALSE) &&
trackDbSetting(tdb, "irows") != NULL)
useIrowChains = TRUE;
safef(buffer, sizeof(buffer), "%s.vis",tdb->track);
if (useIrowChains)
{
if (!cartVarExists(cart, buffer) && (speciesTarget != NULL))
useTarg = TRUE;
else
{
char *val;
val = cartUsualString(cart, buffer, "useCheck");
useTarg = sameString("useTarg",val);
}
}
if (sameString(tdb->type, "bigMaf"))
{
char *fileName = trackDbSetting(tdb, "bigDataUrl");
struct bbiFile *bbi = bigBedFileOpen(fileName);
mafList = bigMafLoadInRegion(bbi, seqName, winStart, winEnd);
}
else
mafList = mafOrAxtLoadInRegion2(conn,conn2, tdb, seqName, winStart, winEnd,
axtOtherDb, fileName);
safef(dbChrom, sizeof(dbChrom), "%s.%s", hubConnectSkipHubPrefix(database), seqName);
safef(option, sizeof(option), "%s.speciesOrder", tdb->track);
speciesOrder = cartUsualString(cart, option, NULL);
if (speciesOrder == NULL)
speciesOrder = trackDbSetting(tdb, "speciesOrder");
for (maf = mafList; maf != NULL; maf = maf->next)
{
int mcCount = 0;
struct mafComp *mc;
struct mafAli *subset;
struct mafComp *nextMc;
/* remove empty components and configured off components
* from MAF, and ignore
* the entire MAF if all components are empty
* (solely for gap annotation) */
if (!useTarg)
{
for (mc = maf->components->next; mc != NULL; mc = nextMc)
{
char buf[64];
char *organism;
mafSrcDb(mc->src, buf, sizeof buf);
organism = hOrganism(buf);
if (!organism)
organism = buf;
nextMc = mc->next;
safef(option, sizeof(option), "%s.%s", tdb->track, buf);
if (!cartUsualBoolean(cart, option, TRUE))
{
if (speciesOffHash == NULL)
speciesOffHash = newHash(4);
hashStoreName(speciesOffHash, organism);
}
if (!cartUsualBoolean(cart, option, TRUE))
slRemoveEl(&maf->components, mc);
else
mcCount++;
}
}
if (mcCount == 0)
continue;
if (speciesOrder)
{
int speciesCt;
char *species[2048];
struct mafComp **newOrder, *mcThis;
int i;
mcCount = 0;
speciesCt = chopLine(cloneString(speciesOrder), species);
newOrder = needMem((speciesCt + 1) * sizeof (struct mafComp *));
newOrder[mcCount++] = maf->components;
for (i = 0; i < speciesCt; i++)
{
if ((mcThis = mafMayFindCompSpecies(maf, species[i], '.')) == NULL)
continue;
newOrder[mcCount++] = mcThis;
}
maf->components = NULL;
for (i = 0; i < mcCount; i++)
{
newOrder[i]->next = 0;
slAddHead(&maf->components, newOrder[i]);
}
slReverse(&maf->components);
}
subset = mafSubsetE(maf, dbChrom, winStart, winEnd, TRUE);
if (subset != NULL)
{
/* Reformat MAF if needed so that sequence from current
* database is the first component and on the
* plus strand. */
mafMoveComponentToTop(subset, dbChrom);
if (subset->components->strand == '-')
mafFlipStrand(subset);
subset->score = mafScoreMultiz(subset);
slAddHead(&subList, subset);
++realCount;
}
}
slReverse(&subList);
mafAliFreeList(&mafList);
if (subList != NULL)
{
char *showVarName = "hgc.showMultiBase";
char *showVarVal = cartUsualString(cart, showVarName, "all");
boolean onlyDiff = sameWord(showVarVal, "diff");
#ifdef ADDEXONCAPITAL
char *codeVarName = "hgc.multiCapCoding";
char *codeVarVal = cartUsualString(cart, codeVarName, "coding");
boolean onlyCds = sameWord(codeVarVal, "coding");
#endif
/* add links for conservation score statistics */
consWiggles = wigMafWiggles(database, tdb);
int wigCount = slCount(consWiggles);
if (wigCount == 1)
{
conservationStatsLink(tdb, "Conservation score statistics", consWiggles->table);
}
else if (wigCount > 1)
{
/* multiple wiggles. List all that have been turned on with
* checkboxes */
/* Scan for cart variables -- do any exist, are any turned on ? */
boolean wigSet = FALSE;
boolean wigOn = FALSE;
for (consWig = consWiggles; consWig != NULL;
consWig = consWig->next)
{
char *wigVarSuffix = NULL;
(void)wigMafWiggleVar(tdb->track, consWig, &wigVarSuffix);
if (cartVarExistsAnyLevel(cart, tdb, FALSE, wigVarSuffix))
{
wigSet = TRUE;
if (cartBooleanClosestToHome(cart, tdb, FALSE, wigVarSuffix))
wigOn = TRUE;
}
}
/* If there are no cart vars, turn on the first (default) wig */
if (!wigSet)
{
char *prefix = tdb->track; // use when setting things to the cart
if (tdbIsContainerChild(tdb))
prefix = tdbGetContainer(tdb)->track;
cartSetBoolean(cart, wigMafWiggleVar(prefix, consWiggles, NULL), TRUE);
wigOn = TRUE;
}
if (wigOn)
{
boolean first = TRUE;
for (consWig = consWiggles; consWig != NULL;
consWig = consWig->next)
{
if (first)
{
printf("Conservation score statistics:");
first = FALSE;
}
char *wigVarSuffix = NULL;
(void)wigMafWiggleVar(tdb->track, consWig, &wigVarSuffix);
if (cartUsualBooleanClosestToHome(cart, tdb, FALSE, wigVarSuffix,FALSE))
{
printf(" ");
subChar(consWig->uiLabel, '_', ' ');
conservationStatsLink(tdb,
consWig->uiLabel, consWig->table);
}
}
}
}
puts("</P>\n");
/* no alignment to display when in visibilities where only wiggle is shown */
char *vis = cartOptionalString(cart, tdb->track);
if (vis)
{
enum trackVisibility tv = hTvFromStringNoAbort(vis);
if (tv == tvSquish || tv == tvDense)
return;
}
#ifdef ADDEXONCAPITAL
puts("<FORM ACTION=\"../cgi-bin/hgc\" NAME=\"gpForm\" METHOD=\"GET\">");
cartSaveSession(cart);
cgiContinueHiddenVar("g");
cgiContinueHiddenVar("c");
cgiContinueHiddenVar("i");
printf("Capitalize ");
cgiMakeDropListFull(codeVarName, codeAll, codeAll,
ArraySize(codeAll), codeVarVal, autoSubmit);
printf("exons based on ");
capTrack = genePredDropDown(cart, trackHash,
"gpForm", "hgc.multiCapTrack");
#endif
printf("show ");
cgiMakeDropListFull(showVarName, showAll, showAll,
ArraySize(showAll), showVarVal, autoSubmit);
printf("bases");
printf("<BR>\n");
printf("</FORM>\n");
#ifdef REVERSESTRAND
/* notify if bases are complemented (hgTracks is on reverse strand) */
if (cartCgiUsualBoolean(cart, COMPLEMENT_BASES_VAR, FALSE))
puts("<EM>Alignment displayed on reverse strand</EM><BR>");
#endif
puts("Place cursor over species for alignment detail. Click on 'B' to link to browser ");
puts("for aligned species, click on 'D' to get DNA for aligned species.<BR>");
printf("<TT><PRE>");
/* notify if species removed from alignment */
if (speciesOffHash)
{
char *species;
struct hashCookie hc = hashFirst(speciesOffHash);
puts("<B>Components not displayed:</B> ");
while ((species = hashNextName(&hc)) != NULL)
printf("%s ", species);
puts("<BR>");
}
for (maf = subList; maf != NULL; maf = maf->next)
{
mafLowerCase(maf);
#ifdef ADDEXONCAPITAL
if (capTrack != NULL)
capMafOnTrack(maf, capTrack, onlyCds);
#endif
printf("<B>Alignment block %d of %d in window, %d - %d, %d bps </B>\n",
++aliIx,realCount,maf->components->start + 1,
maf->components->start + maf->components->size, maf->components->size);
mafPrettyOut(stdout, maf, 70,onlyDiff, aliIx);
}
mafAliFreeList(&subList);
}
else
{
printf("No multiple alignment in browser window");
}
printf("</PRE></TT>");
}
}
