long long currentVmPeak()
/* return value of peak Vm memory usage (if /proc/ business exists) */
{
long long vmPeak = 0;
pid_t pid = getpid();
char temp[256];
safef(temp, sizeof(temp), "/proc/%d/status", (int) pid);
struct lineFile *lf = lineFileMayOpen(temp, TRUE);
if (lf)
{
char *line;
while (lineFileNextReal(lf, &line))
{ // typical line: 'VmPeak: 62646196 kB'
// seems to always be kB
if (stringIn("VmPeak", line))
{
char *words[3];
chopByWhite(line, words, 3);
vmPeak = sqlLongLong(words[1]); // assume always 2nd word
break;
}
}
lineFileClose(&lf);
}
return vmPeak;
}
struct lineFile *lineFileOpen(char *fileName, bool zTerm)
/* Open up a lineFile or die trying. */
{
struct lineFile *lf = lineFileMayOpen(fileName, zTerm);
if (lf == NULL)
errAbort("Couldn't open %s , %s", fileName, strerror(errno));
return lf;
}
void liftGl(char *destFile, struct hash *liftHash, int sourceCount, char *sources[])
/* Lift up coordinates in .gl file. */
{
char dirBuf[256], chromName[256];
int i;
char *source;
char *contig;
FILE *dest = mustOpen(destFile, "w");
struct lineFile *lf = NULL;
int lineSize, wordCount;
char *line, *words[32];
struct liftSpec *spec;
int offset;
if (how == carryMissing)
warn("'carry' doesn't work for .gl files, ignoring");
for (i=0; i<sourceCount; ++i)
{
source = sources[i];
verbose(1, "Processing %s\n", source);
contig = contigInDir(source, dirBuf);
verbose(2,"#\tcontig: %s, source: %s, dirBuf: %s\n", contig, source, dirBuf);
if (!startsWith("ctg", contig) &&
!startsWith("NC_", contig) &&
!startsWith("NT_", contig) &&
!startsWith("NG_", contig))
{
sprintf(chromName, "chr%s", contig);
contig = chromName;
verbose(2,"#\tcontig: %s, chromName: %s\n", contig, chromName);
}
spec = findLift(liftHash, contig, lf);
if (spec == NULL)
continue;
cantHandleSpecRevStrand(spec);
offset = spec->offset;
lf = lineFileMayOpen(source, TRUE);
if (lf == NULL)
{
warn("%s doesn't exist, skipping", source);
continue;
}
while (lineFileNext(lf, &line, &lineSize))
{
int s, e;
if ((wordCount = chopLine(line, words)) != 4)
errAbort("Bad line %d of %s", lf->lineIx, lf->fileName);
s = atoi(words[1]);
e = atoi(words[2]);
fprintf(dest, "%s\t%d\t%d\t%s\n", words[0], s+offset, e+offset, words[3]);
}
lineFileClose(&lf);
if (dots)
verbose(1, "\n");
}
}
struct vcfFile *vcfFileMayOpen(char *fileOrUrl, int maxErr, int maxRecords)
/* Parse a VCF file into a vcfFile object. If maxErr not zero, then
* continue to parse until this number of error have been reached. A maxErr
* less than zero does not stop and reports all errors. */
{
struct lineFile *lf = NULL;
if (startsWith("http://", fileOrUrl) || startsWith("ftp://", fileOrUrl) ||
startsWith("https://", fileOrUrl))
lf = netLineFileOpen(fileOrUrl);
else
lf = lineFileMayOpen(fileOrUrl, TRUE);
struct vcfFile *vcff = vcfFileHeaderFromLineFile(lf, maxErr);
vcfParseData(vcff, maxRecords);
return vcff;
}
struct mouseChromCache *newMouseChromCache(char *chrom, int chromSize,
char *ratMouseDir)
/* Create a new chromCache. */
{
struct mouseChromCache *mcc;
char fileName[512];
struct lineFile *lf;
char *row[3];
int start,end;
long long *pPos;
/* Open up file with actual alignments. Warn and return NULL
* if it doesn't exist. */
sprintf(fileName, "%s/%s.axt", ratMouseDir, chrom);
lf = lineFileMayOpen(fileName, TRUE);
/* Allocate structure and store basic info in it. */
AllocVar(mcc);
mcc->name = cloneString(chrom);
mcc->size = chromSize;
mcc->lf = lf;
if (lf == NULL)
{
warn("%s doesn't exist", fileName);
if (!noDieMissing)
noWarnAbort();
return mcc;
}
/* Read index file into bk. */
sprintf(fileName, "%s/%s.axt.ix", ratMouseDir, chrom);
mcc->bk = binKeeperNew(0, chromSize);
lf = lineFileOpen(fileName, TRUE);
verbose(1, "Reading %s\n", fileName);
while (lineFileRow(lf, row))
{
start = lineFileNeedNum(lf, row, 0);
end = lineFileNeedNum(lf, row, 1) + start;
AllocVar(pPos);
*pPos = atoll(row[2]);
binKeeperAdd(mcc->bk, start, end, pPos);
}
lineFileClose(&lf);
/* Return initialized object. */
return mcc;
}
void printMem(FILE *f)
/* Print out memory used and other stuff from linux. */
{
struct lineFile *lf = lineFileMayOpen("/proc/self/stat", TRUE);
if (lf != NULL)
{
char *line, *words[50];
int wordCount;
if (lineFileNext(lf, &line, NULL))
{
wordCount = chopLine(line, words);
if (wordCount >= 23)
fprintf(f, "memory usage %s, utime %s s/100, stime %s\n",
words[22], words[13], words[14]);
}
lineFileClose(&lf);
}
}
static struct visiMatch *readMatchFile(char *fileName)
/* Read in match file */
{
struct visiMatch *matchList = NULL, *match;
struct lineFile *lf = lineFileMayOpen(fileName, TRUE);
if (lf != NULL)
{
char *row[2];
while (lineFileRow(lf, row))
{
AllocVar(match);
match->imageId = lineFileNeedNum(lf, row, 0);
match->weight = lineFileNeedDouble(lf, row, 1);
slAddHead(&matchList, match);
}
lineFileClose(&lf);
slReverse(&matchList);
}
return matchList;
}
struct vcfFile *vcfFileMayOpen(char *fileOrUrl, int maxErr, int maxRecords, boolean parseAll)
/* Open fileOrUrl and parse VCF header; return NULL if unable.
* If parseAll, then read in all lines, parse and store in
* vcff->records; if maxErr >= zero, then continue to parse until
* there are maxErr+1 errors. A maxErr less than zero does not stop
* and reports all errors. Set maxErr to VCF_IGNORE_ERRS for silence */
{
struct lineFile *lf = NULL;
if (startsWith("http://", fileOrUrl) || startsWith("ftp://", fileOrUrl) ||
startsWith("https://", fileOrUrl))
lf = netLineFileOpen(fileOrUrl);
else
lf = lineFileMayOpen(fileOrUrl, TRUE);
struct vcfFile *vcff = vcfFileHeaderFromLineFile(lf, maxErr);
if (parseAll)
{
vcff->records = vcfParseData(vcff, maxRecords);
lineFileClose(&(vcff->lf)); // Not sure why it is closed. Angie?
}
return vcff;
}
void setAliBits(char *axtBestDir, char *chrom, int chromSize,
Bits *aliBits, Bits *matchBits)
/* Set bits where there are alignments and matches. */
{
char axtFileName[512];
struct axt *axt;
struct lineFile *lf;
sprintf(axtFileName, "%s/%s.axt", axtBestDir, chrom);
if ((lf = lineFileMayOpen(axtFileName, TRUE)) == NULL)
{
warn("Couldn't open %s", axtFileName);
return;
}
while ((axt = axtRead(lf)) != NULL)
{
axtSetBits(axt, chromSize, aliBits, matchBits);
axtFree(&axt);
}
lineFileClose(&lf);
}
struct blackListRange *genbankBlackListParse(char *blackList)
/* parse a black list file into blackListRange data structure */
{
struct lineFile *lf = lineFileMayOpen(blackList, TRUE);
if (lf == NULL)
errAbort("Could not open black list file %s. ", blackList);
struct blackListRange *ranges = NULL;
char *words[2];
while(lineFileRow(lf, words))
{
char *prefix1 = cloneString(words[0]);
genbankDropVer(prefix1, prefix1);
char *number1 = skipToNumeric(prefix1);
int begin = atoi(number1);
*number1 = 0; // null so now prefix1 points to only the prefix
char *prefix2 = cloneString(words[1]);
genbankDropVer(prefix2, prefix2);
char *number2 = skipToNumeric(prefix2);
int end = atoi(number2);
*number2 = 0; // null so now prefix2 points to only the prefix
if (differentString(prefix1, prefix2))
errAbort("blackList file %s has accesions with different prefixes on line %d\n",
lf->fileName, lf->lineIx);
if (begin > end)
errAbort("blackList file %s has end before begin on line %d\n",
lf->fileName, lf->lineIx);
struct blackListRange *range;
AllocVar(range);
range->prefix = prefix1;
range->begin = begin;
range->end = end;
slAddHead(&ranges, range);
}
return ranges;
}
void raFoldIn(char *fileName, struct hash *hashOfHash)
/* Read ra's in file name and fold them into hashOfHash.
* This will add ra's and ra fields to whatever already
* exists in the hashOfHash, overriding fields of the
* same name if they exist already. */
{
struct lineFile *lf = lineFileMayOpen(fileName, TRUE);
if (lf != NULL)
{
struct hash *uniqHash = hashNew(0);
char *name;
while ((name = raFoldInOneRetName(lf, hashOfHash)) != NULL)
{
if (hashLookup(uniqHash, name))
errAbort("%s duplicated in record ending line %d of %s", name,
lf->lineIx, lf->fileName);
hashAdd(uniqHash, name, NULL);
}
lineFileClose(&lf);
hashFree(&uniqHash);
}
}
struct segFile *segMayOpen(char *fileName)
/* Open up a segment file for reading. Read header and verify. Prepare
* for subsequent calls to segNext(). Return NULL if file does not exist. */
{
struct segFile *sf;
struct lineFile *lf;
char *line, *name, *val, *word;
char *sig = "##seg";
/* Open fileName. */
if ((lf = lineFileMayOpen(fileName, TRUE)) == NULL)
return NULL;
AllocVar(sf);
sf->lf = lf;
/* Check for a valid signature. */
lineFileNeedNext(lf, &line, NULL);
if (!startsWith(sig, line))
errAbort("%s does not start with %s", fileName, sig);
line += strlen(sig);
/* parse name=val. */
while ((word = nextWord(&line)) != NULL)
{
name = word;
val = strchr(word, '=');
if (val == NULL)
errAbort("Missing = after %s line 1 of %s", name, fileName);
*val++ = 0;
if (sameString(name, "version"))
sf->version = atoi(val);
}
if (sf->version == 0)
errAbort("No version line 1 of %s", fileName);
return sf;
}
struct mafFile *mafMayOpen(char *fileName)
/* Open up a maf file and verify header. */
{
struct mafFile *mf;
struct lineFile *lf;
char *line, *word;
char *sig = "##maf";
if ((lf = lineFileMayOpen(fileName, TRUE)) == NULL)
return NULL;
AllocVar(mf);
mf->lf = lf;
lineFileNeedNext(lf, &line, NULL);
if (!startsWith(sig, line))
{
errAbort("%s does not start with %s", fileName, sig);
}
line += strlen(sig);
while ((word = nextWord(&line)) != NULL)
{
/* Parse name=val. */
char *name = word;
char *val = strchr(word, '=');
if (val == NULL)
errAbort("Missing = after %s line 1 of %s\n", name, fileName);
*val++ = 0;
if (sameString(name, "version"))
mf->version = atoi(val);
else if (sameString(name, "scoring"))
mf->scoring = cloneString(val);
}
if (mf->version == 0)
errAbort("No version line 1 of %s\n", fileName);
return mf;
}
struct slName *randomVcfIds(char *table, struct sqlConnection *conn, int count, boolean isTabix)
/* Return some semi-random IDs from a VCF file. */
{
/* Read 10000 items from vcf file, or if they ask for a big list, then 4x what they ask for. */
struct trackDb *tdb = hashFindVal(fullTableToTdbHash, table);
char *fileName = vcfFileName(tdb, conn, table, hDefaultChrom(database));
struct lineFile *lf = isTabix ? lineFileTabixMayOpen(fileName, TRUE) :
lineFileMayOpen(fileName, TRUE);
if (lf == NULL)
noWarnAbort();
int orderedCount = count * 4;
if (orderedCount < 100)
orderedCount = 100;
struct slName *idList = NULL;
char *words[4];
int i;
for (i = 0; i < orderedCount && lineFileChop(lf, words); i++)
{
// compress runs of identical ID, in case most are placeholder
if (i == 0 || !sameString(words[2], idList->name))
slAddHead(&idList, slNameNew(words[2]));
}
lineFileClose(&lf);
/* Shuffle list and trim it to count if necessary. */
shuffleList(&idList);
struct slName *sl;
for (sl = idList, i = 0; sl != NULL; sl = sl->next, i++)
{
if (i+1 >= count)
{
slNameFreeList(&(sl->next));
break;
}
}
freez(&fileName);
return idList;
}
void printVmPeak()
/* print to stderr peak Vm memory usage (if /proc/ business exists) */
{
pid_t pid = getpid();
char temp[256];
safef(temp, sizeof(temp), "/proc/%d/status", (int) pid);
struct lineFile *lf = lineFileMayOpen(temp, TRUE);
if (lf)
{
char *line;
while (lineFileNextReal(lf, &line))
{
if (stringIn("VmPeak", line))
{
fprintf(stderr, "# pid=%d: %s\n", pid, line);
break;
}
}
lineFileClose(&lf);
}
else
fprintf(stderr, "# printVmPeak: %s - not available\n", temp);
fflush(stderr);
}
void liftAgp(char *destFile, struct hash *liftHash, int sourceCount, char *sources[])
/* Lift up coordinates in .agp file. */
{
FILE *dest = mustOpen(destFile, "w");
char *source;
int i;
struct lineFile *lf;
int lineSize, wordCount;
char *line, *words[32];
char *s;
struct liftSpec *spec;
int start = 0;
int end = 0;
int ix = 0;
char newDir[256], newName[128], newExt[64];
struct bigInsert *bi;
struct chromInserts *chromInserts;
struct hash *insertHash = newHash(8);
struct hash *contigsHash = newHash(10);
boolean firstContig = TRUE;
char lastContig[256];
char *contig;
int lastEnd = 0;
if (sourceCount < 2)
usage();
if (how == carryMissing)
warn("'carry' doesn't work for .agp files, ignoring");
splitPath(destFile, newDir, newName, newExt);
/* Read in inserts file and process it. */
chromInsertsRead(sources[0], insertHash);
chromInserts = hashFindVal(insertHash, newName);
strcpy(lastContig, "");
for (i=1; i<sourceCount; ++i)
{
source = sources[i];
verbose(1, "Lifting %s\n", source);
lf = lineFileMayOpen(source, TRUE);
if (lf != NULL)
{
while (lineFileNext(lf, &line, &lineSize))
{
/* Check for comments and just pass them through. */
s = skipLeadingSpaces(line);
if (s[0] == '#')
{
fprintf(dest, "%s\n", line);
continue;
}
/* Parse line, adjust offsets, write */
wordCount = chopLine(line, words);
if (wordCount != 8 && wordCount != 9)
malformedAgp(lf);
contig = words[0];
if (!sameString(contig, lastContig))
{
char *gapType = "contig";
char *ctg = rmChromPrefix(contig);
int gapSize = chromInsertsGapSize(chromInserts,
ctg, firstContig);
if (hashLookup(contigsHash, contig))
errAbort("Contig repeated line %d of %s", lf->lineIx, lf->fileName);
hashAdd(contigsHash, contig, NULL);
if (gapSize != 0)
{
if ((bi = bigInsertBeforeContig(chromInserts, ctg)) != NULL)
{
gapType = bi->type;
}
fprintf(dest, "%s\t%d\t%d\t%d\tN\t%d\t%s\tno\n",
newName, end+1, end+gapSize, ++ix, gapSize, gapType);
}
firstContig = FALSE;
strcpy(lastContig, contig);
}
spec = findLift(liftHash, contig, lf);
cantHandleSpecRevStrand(spec);
start = numField(words, 1, 0, lf) + spec->offset;
end = numField(words, 2, 0, lf) + spec->offset;
if (end > lastEnd) lastEnd = end;
if (!sameString(newName, spec->newName))
errAbort("Mismatch in new name between %s and %s", newName, spec->newName);
fprintf(dest, "%s\t%d\t%d\t%d\t%s\t%s\t%s\t%s",
newName, start, end, ++ix,
words[4], words[5], words[6], words[7]);
if (wordCount == 9)
fprintf(dest, "\t%s", words[8]);
fputc('\n', dest);
}
lineFileClose(&lf);
if (dots)
verbose(1, "\n");
}
}
if (chromInserts != NULL)
{
if ((bi = chromInserts->terminal) != NULL)
{
fprintf(dest, "%s\t%d\t%d\t%d\tN\t%d\t%s\tno\n",
newName, lastEnd+1, lastEnd+bi->size, ++ix, bi->size, bi->type);
}
}
if (ferror(dest))
errAbort("error writing %s", destFile);
fclose(dest);
}
void liftGap(char *destFile, struct hash *liftHash, int sourceCount, char *sources[])
/* Lift up coordinates in .gap file (just the gaps from .agp). Negative strad allowed */
{
FILE *dest = mustOpen(destFile, "w");
char *source;
int i;
struct lineFile *lf;
int lineSize, wordCount;
char *line, *words[32];
char *s;
struct liftSpec *spec;
int start = 0;
int end = 0;
int ix = 0;
char newDir[256], newName[128], newExt[64];
char lastContig[256];
char *contig;
int lastEnd = 0;
int fragStart, fragEnd;
if (how == carryMissing)
warn("'carry' doesn't work for .gap files, ignoring");
splitPath(destFile, newDir, newName, newExt);
strcpy(lastContig, "");
for (i=0; i<sourceCount; ++i)
{
source = sources[i];
verbose(1, "Lifting %s\n", source);
lf = lineFileMayOpen(source, TRUE);
if (lf != NULL)
{
while (lineFileNext(lf, &line, &lineSize))
{
/* Check for comments and just pass them through. */
s = skipLeadingSpaces(line);
if (s[0] == '#')
{
fprintf(dest, "%s\n", line);
continue;
}
/* Parse line, adjust offsets, write */
wordCount = chopLine(line, words);
if (wordCount != 8 && wordCount != 9)
malformedAgp(lf);
if (words[4][0] != 'N' && words[4][0] != 'U')
errAbort("Found non-gap in .gap file: %s", words[4]);
contig = words[0];
spec = findLift(liftHash, contig, lf);
start = fragStart = numField(words, 1, 0, lf);
end = fragEnd = numField(words, 2, 0, lf);
end = fragEnd;
if (spec->strand == '-')
{
start = spec->oldSize - fragEnd + 1;
end = spec->oldSize - fragStart + 1;
}
start += spec->offset;
end += spec->offset;
if (end > lastEnd) lastEnd = end;
fprintf(dest, "%s\t%d\t%d\t%d\t%s\t%s\t%s\t%s",
spec->newName, start, end, ++ix,
words[4], words[5], words[6], words[7]);
if (wordCount == 9)
fprintf(dest, "\t%s", words[8]);
fputc('\n', dest);
}
lineFileClose(&lf);
if (dots)
verbose(1, "\n");
}
}
if (ferror(dest))
errAbort("error writing %s", destFile);
fclose(dest);
}
void configMultiRegionPage()
/* Do multi-region config page after setting track visibility. If vis is -2, then visibility
* is unchanged. If -1 then set visibility to default, otherwise it should
* be tvHide, tvDense, etc. */
{
char *groupTarget;
struct track *trackList;
struct track *ideoTrack;
struct group *groupList;
int vis = -2;
configInitTrackList(vis, &groupTarget, &trackList, &ideoTrack, &groupList);
hPrintf("<FORM ACTION=\"%s\" NAME=\"mainForm\" METHOD=%s>\n", hgTracksName(),
cartUsualString(cart, "formMethod", "POST"));
webStartWrapperDetailedNoArgs(cart, database, "", "", FALSE, FALSE, FALSE, FALSE);
cartSaveSession(cart);
hPrintf("<BR>\n");
hTableStart();
virtModeType = cartUsualString(cart, "virtModeType", virtModeType);
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "default", sameWord("default", virtModeType));
hPrintf("</TD><TD>");
hPrintf("Exit multi-region mode");
hPrintf("</TD></TR>\n");
struct sqlConnection *conn = NULL;
if (!trackHubDatabase(database)) // no db conn for assembly hubs
conn = hAllocConn(database);
// Do we have a gene table for exonMostly?
findBestEMGeneTable(trackList);
if (emGeneTable)
{
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "exonMostly", sameWord("exonMostly", virtModeType));
hPrintf("</TD><TD>");
hPrintf("Show exons using %s. Use padding of: ", emGeneTrack->shortLabel);
hIntVar("emPadding", cartUsualInt(cart, "emPadding", emPadding), 3);
hPrintf(" bases.");
hPrintf("</TD></TR>\n");
}
if (emGeneTable)
{
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "geneMostly", sameWord("geneMostly", virtModeType));
hPrintf("</TD><TD>");
hPrintf("Show genes using %s. Use padding of: ", emGeneTrack->shortLabel);
hIntVar("gmPadding", cartUsualInt(cart, "gmPadding", gmPadding), 3);
hPrintf(" bases.");
hPrintf("</TD></TR>\n");
}
/* obsolete
if (conn && sqlTableExists(conn,"knownCanonical"))
{
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "kcGenes", sameWord("kcGenes", virtModeType));
hPrintf("</TD><TD>");
hPrintf("Show gene regions genome-wide.");
hPrintf("</TD></TR>\n");
}
*/
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "customUrl", sameWord("customUrl", virtModeType));
hPrintf("</TD><TD>");
hPrintf("Enter Custom regions as BED, or a URL to them:<br>");
multiRegionsBedUrl = cartUsualString(cart, "multiRegionsBedUrl", multiRegionsBedUrl);
struct dyString *dyMultiRegionsBedInput = dyStringNew(256);
if (strstr(multiRegionsBedUrl,"://"))
{
dyStringAppend(dyMultiRegionsBedInput, multiRegionsBedUrl);
}
else
{
if (fileExists(multiRegionsBedUrl))
{
struct lineFile *lf = lineFileMayOpen(multiRegionsBedUrl, TRUE);
char *line;
int lineSize;
while (lineFileNext(lf, &line, &lineSize))
{
dyStringPrintf(dyMultiRegionsBedInput, "%s\n", line);
}
lineFileClose(&lf);
}
}
hPrintf("<TEXTAREA NAME='multiRegionsBedInput' ID='multiRegionsBedInput' rows='4' cols='58' style='white-space: pre;'>%s</TEXTAREA>",
dyMultiRegionsBedInput->string);
hPrintf("</TD></TR>\n");
/* The AllChroms option will be released in future
if (emGeneTable && sqlTableExists(conn, emGeneTable))
{
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "singleTrans", sameWord("singleTrans", virtModeType));
hPrintf("</TD><TD>");
hPrintf("Show only one transcript using an ID from %s : ", emGeneTrack->shortLabel);
char *trans = cartUsualString(cart, "singleTransId", singleTransId);
char sql[1024];
sqlSafef(sql, sizeof sql, "select name from %s where name='%s'", emGeneTable, trans);
char *result = sqlQuickString(conn, sql);
if (!result)
{
sqlSafef(sql, sizeof sql, "select name from %s limit 1", emGeneTable);
trans = sqlQuickString(conn, sql);
}
hTextVar("singleTransId", trans, 20);
hPrintf("</TD></TR>\n");
}
*/
if (conn)
{
boolean altLocExists = sqlTableExists(conn, "altLocations");
boolean fixLocExists = sqlTableExists(conn, "fixLocations");
if (altLocExists || fixLocExists)
{
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "singleAltHaplo",
sameWord("singleAltHaplo", virtModeType));
hPrintf("</TD><TD>");
hPrintf("Show one alternate haplotype");
if (fixLocExists)
hPrintf(" or fix patch");
hPrintf(", placed on its chromosome, using ID: ");
char *haplo = cartUsualString(cart, "singleAltHaploId", singleAltHaploId);
char *foundHaplo = NULL;
char sql[1024];
if (altLocExists)
{
sqlSafef(sql, sizeof sql,
"select name from altLocations where name rlike '^%s(:[0-9-]+)?'", haplo);
foundHaplo = sqlQuickString(conn, sql);
}
if (!foundHaplo && fixLocExists)
{
sqlSafef(sql, sizeof sql,
"select name from fixLocations where name rlike '^%s(:[0-9-]+)?'", haplo);
foundHaplo = sqlQuickString(conn, sql);
}
if (!foundHaplo)
{
if (altLocExists)
sqlSafef(sql, sizeof sql, "select name from altLocations limit 1");
else
sqlSafef(sql, sizeof sql, "select name from fixLocations limit 1");
haplo = sqlQuickString(conn, sql);
chopSuffixAt(haplo, ':');
}
hTextVar("singleAltHaploId", haplo, 60);
hPrintf("</TD></TR>\n");
}
}
/* disable demo for now
if (sameString(database,"hg19") || sameString(database, "hg38"))
{
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "demo1", sameWord("demo1", virtModeType));
hPrintf("</TD><TD>");
hPrintf("demo1 two windows on two chroms (default pos on chr21, and same loc on chr22)");
hPrintf("</TD></TR>\n");
}
*/
/* Disabled for now
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "demo2", sameWord("demo2", virtModeType));
hPrintf("</TD><TD>");
hPrintf("demo2 multiple ");
hIntVar("demo2NumWindows", cartUsualInt(cart, "demo2NumWindows", demo2NumWindows), 3);
hPrintf(" windows on one chrom chr21 def posn, window size ");
hIntVar("demo2WindowSize", cartUsualInt(cart, "demo2WindowSize", demo2WindowSize), 3);
hPrintf(" and step size ");
hIntVar("demo2StepSize", cartUsualInt(cart, "demo2StepSize", demo2StepSize), 3);
hPrintf(" exon-like");
hPrintf("</TD></TR>\n");
*/
/* The AllChroms option will be released in future
if (conn) // requires chromInfo from database.
{ // TODO allow it to use assembly hubs via trackHubAllChromInfo() ?
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "allChroms", sameWord("allChroms", virtModeType));
hPrintf("</TD><TD>");
hPrintf("<br>Show all chromosomes.<br><span style='color:red'>Warning:</span> Turn off all tracks except bigBed, bigWig, and very sparse tracks.<br>Press Hide All to hide all tracks.");
hPrintf("</TD></TR>\n");
}
*/
/* Disabled for now
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "demo4", sameWord("demo4", virtModeType));
hPrintf("</TD><TD>");
hPrintf("demo4 multiple (311) windows showing exons from TITIN gene uc031rqd.1.");
hPrintf("</TD></TR>\n");
*/
/* Disabled for now
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "demo5", sameWord("demo5", virtModeType));
hPrintf("</TD><TD>");
hPrintf("demo5 alt locus on hg38. Shows alt chrom surrounded by regions of same size from reference genome.");
hPrintf("</TD></TR>\n");
*/
/* Disabled for now
hPrintf("<TR><TD>");
cgiMakeRadioButton("virtModeType", "demo6", sameWord("demo6", virtModeType));
hPrintf("</TD><TD>");
hPrintf("demo6 shows zoomed in exon-exon junction from SOD1 gene, between exon1 and exon2.");
hPrintf("</TD></TR>\n");
*/
hTableEnd();
hPrintf("<BR>\n");
hPrintf("<TABLE style=\"border:0px; \">\n");
hPrintf("<TR><TD>");
hCheckBox("emAltHighlight", cartUsualBoolean(cart, "emAltHighlight", FALSE));
hPrintf("</TD><TD>");
hPrintf("Highlight alternating regions in multi-region view");
hPrintf("</TD></TR>\n");
hPrintf("</TABLE>\n");
hPrintf("<BR>\n");
hPrintf("<TABLE style=\"border:0px;width:650px \">\n");
hPrintf("<TR><TD>");
cgiMakeButton("topSubmit", "submit");
hPrintf("</TD><TD align=right>");
hPrintf("<A HREF=\"../goldenPath/help/multiRegionHelp.html\" target=_blank>Help</A>\n");
hPrintf("</TD></TR>\n");
hPrintf("</TABLE>\n");
hFreeConn(&conn);
cgiDown(0.9);
freez(&groupTarget);
webEndSectionTables();
hPrintf("</FORM>");
}
void showSchemaVcf(char *table, struct trackDb *tdb, boolean isTabix)
/* Show schema on vcf. */
{
struct sqlConnection *conn = hAllocConn(database);
char *fileName = vcfFileName(tdb, conn, table, hDefaultChrom(database));
struct asObject *as = vcfAsObj();
hPrintf("<B>Database:</B> %s", database);
hPrintf(" <B>Primary Table:</B> %s<br>", table);
hPrintf("<B>VCF File:</B> %s", fileName);
hPrintf("<BR>\n");
hPrintf("<B>Format description:</B> %s<BR>", as->comment);
hPrintf("See the <A HREF=\"%s\" target=_blank>Variant Call Format specification</A> for more details<BR>\n",
"http://www.1000genomes.org/wiki/analysis/vcf4.0");
/* Put up table that describes fields. */
hTableStart();
hPrintf("<TR><TH>field</TH>");
hPrintf("<TH>description</TH> ");
puts("</TR>\n");
struct asColumn *col;
int colCount = 0;
for (col = as->columnList; col != NULL; col = col->next)
{
hPrintf("<TR><TD><TT>%s</TT></TD>", col->name);
hPrintf("<TD>%s</TD></TR>", col->comment);
++colCount;
}
hTableEnd();
/* Put up another section with sample rows. */
webNewSection("Sample Rows");
hTableStart();
/* Fetch sample rows. */
struct lineFile *lf = isTabix ? lineFileTabixMayOpen(fileName, TRUE) :
lineFileMayOpen(fileName, TRUE);
if (lf == NULL)
noWarnAbort();
char *row[VCF_MAX_SCHEMA_COLS];
int i;
for (i = 0; i < 10; i++)
{
int colCount = lineFileChop(lf, row);
int colIx;
if (i == 0)
{
// Print field names as column headers, using colCount to compute genotype span
hPrintf("<TR>");
for (colIx = 0, col = as->columnList; col != NULL && colIx < colCount;
colIx++, col = col->next)
{
if (sameString("genotypes", col->name) && colCount > colIx+1)
hPrintf("<TH colspan=%d>%s</TH>", colCount - colIx, col->name);
else
hPrintf("<TH>%s</TH>", col->name);
}
hPrintf("</TR>\n");
}
hPrintf("<TR>");
for (colIx=0; colIx < colCount; ++colIx)
{
if (colCount > VCFDATALINE_NUM_COLS && colIx == colCount - 1)
hPrintf("<TD>...</TD>");
else
writeHtmlCell(row[colIx]);
}
hPrintf("</TR>\n");
}
hTableEnd();
printTrackHtml(tdb);
/* Clean up and go home. */
lineFileClose(&lf);
freeMem(fileName);
hFreeConn(&conn);
}
void sortGenes(struct sqlConnection *conn)
/* Put up sort gene page. */
{
cartWebStart(cart, database, "Finding Candidate Genes for Gene Sorter");
if (!hgNearOk(database))
errAbort("Sorry, gene sorter not available for this database.");
/* Get list of regions. */
struct genoGraph *gg = ggFirstVisible();
double threshold = getThreshold();
struct bed3 *bed, *bedList = regionsOverThreshold(gg);
/* Figure out what table and column are the sorter's main gene set. */
struct hash *genomeRa = hgReadRa(genome, database, "hgNearData",
"genome.ra", NULL);
char *geneTable = hashMustFindVal(genomeRa, "geneTable");
char *idColumn = hashMustFindVal(genomeRa, "idColumn");
/* if marker labels were present when the file was uploaded, they are saved here */
char cgmName[256];
safef(cgmName, sizeof(cgmName), "%s.cgm", gg->binFileName);
struct lineFile *m = lineFileMayOpen(cgmName, TRUE);
char *cgmRow[4];
cgmRow[0] = ""; /* dummy row */
cgmRow[1] = "";
cgmRow[2] = "0";
cgmRow[3] = "0";
FILE *g = NULL;
int markerCount = 0;
struct tempName snpTn;
if (m)
{
/* Create custom column output file. */
trashDirFile(&snpTn, "hgg", "marker", ".mrk");
g = mustOpen(snpTn.forCgi, "w");
fprintf(g,
"column name=\"%s Markers\" shortLabel=\"%s Markers over threshold\" longLabel=\"%s Markers in regions over threshold\" "
"visibility=on priority=99 "
"\n"
, gg->shortLabel
, gg->shortLabel
, gg->shortLabel
);
}
/*** Build up hash of all transcriptHash that are in region. */
struct hash *transcriptHash = hashNew(16);
/* This loop handles one chromosome at a time. It depends on
* the bedList being sorted by chromosome. */
for (bed = bedList; bed != NULL; )
{
/* Make binKeeper and stuff in all regions in this chromosome into it. */
char *chrom = bed->chrom;
int chromSize = hChromSize(database, chrom);
struct binKeeper *bk = binKeeperNew(0, chromSize);
while (bed != NULL && sameString(chrom, bed->chrom))
{
binKeeperAdd(bk, bed->chromStart, bed->chromEnd, bed);
bed = bed->next;
}
struct binKeeper *bkGenes = NULL;
if (m)
bkGenes = binKeeperNew(0, chromSize);
/* Query database to find out bounds of all genes on this chromosome
* and if they overlap any of the regions then put them in the hash. */
char query[512];
safef(query, sizeof(query),
"select name,txStart,txEnd from %s where chrom='%s'", geneTable, chrom);
struct sqlResult *sr = sqlGetResult(conn, query);
char **row;
while ((row = sqlNextRow(sr)) != NULL)
{
char *name = row[0];
int start = sqlUnsigned(row[1]);
int end = sqlUnsigned(row[2]);
if (binKeeperAnyOverlap(bk, start, end))
{
hashStore(transcriptHash, name);
if (m)
binKeeperAdd(bkGenes, start, end, cloneString(name));
}
}
sqlFreeResult(&sr);
if (m)
{
/* Read cgm file if it exists, looking at all markers on this chromosome
* and if they overlap any of the regions and genes then output them. */
do
{
// marker, chrom, chromStart, val
char *marker = cgmRow[0];
char *chr = cgmRow[1];
int start = sqlUnsigned(cgmRow[2]);
int end = start+1;
double val = sqlDouble(cgmRow[3]);
int cmp = strcmp(chr,chrom);
if (cmp > 0)
break;
if (cmp == 0)
{
if (val >= threshold)
{
struct binElement *el, *bkList = binKeeperFind(bkGenes, start, end);
for (el = bkList; el; el=el->next)
{
/* output to custom column trash file */
fprintf(g, "%s %s\n", (char *)el->val, marker);
}
if (bkList)
{
++markerCount;
slFreeList(&bkList);
}
}
}
}
while (lineFileRow(m, cgmRow));
}
/* Clean up for this chromosome. */
binKeeperFree(&bk);
if (m)
{
/* For speed, we do not free up the values (cloned the kg names earlier) */
binKeeperFree(&bkGenes);
}
}
/* Get list of all transcripts in regions. */
struct hashEl *el, *list = hashElListHash(transcriptHash);
/* Create file with all matching gene IDs. */
struct tempName keyTn;
trashDirFile(&keyTn, "hgg", "key", ".key");
FILE *f = mustOpen(keyTn.forCgi, "w");
for (el = list; el != NULL; el = el->next)
fprintf(f, "%s\n", el->name);
carefulClose(&f);
/* Print out some info. */
hPrintf("Thresholding <i>%s</i> at %g. ", gg->shortLabel, threshold);
hPrintf("There are %d regions covering %lld bases.<BR>\n",
slCount(bedList), bedTotalSize((struct bed*)bedList) );
hPrintf("Installed a Gene Sorter filter that selects only genes in these regions.<BR>\n");
if (m)
{
hPrintf("There are %d markers in the regions over threshold that overlap knownGenes.<BR>\n", markerCount);
hPrintf("Installed a Gene Sorter custom column called \"%s Markers\" with these markers.<BR>\n", gg->shortLabel);
}
/* close custom column output file */
if (m)
{
lineFileClose(&m);
carefulClose(&g);
}
/* Stuff cart variable with name of file. */
char keyCartName[256];
safef(keyCartName, sizeof(keyCartName), "%s%s.keyFile",
advFilterPrefix, idColumn);
cartSetString(cart, keyCartName, keyTn.forCgi);
cartSetString(cart, customFileVarName, snpTn.forCgi);
char snpVisCartNameTemp[256];
char *snpVisCartName = NULL;
safef(snpVisCartNameTemp, sizeof(snpVisCartNameTemp), "%s%s Markers.vis",
colConfigPrefix, gg->shortLabel);
snpVisCartName = replaceChars(snpVisCartNameTemp, " ", "_");
cartSetString(cart, snpVisCartName, "1");
freeMem(snpVisCartName);
hPrintf("<FORM ACTION=\"../cgi-bin/hgNear\" METHOD=GET>\n");
cartSaveSession(cart);
hPrintf("<CENTER>");
cgiMakeButton("submit", "go to gene sorter");
hPrintf("</CENTER>");
hPrintf("</FORM>");
cartWebEnd();
}
