void makeHtml(char *fileName)
/* Make html page with links for Tom. */
{
struct tomRough *tr;
int start, end;
char chrom[256];
FILE *f = mustOpen(fileName, "w");
htmStart(f, "roughly mapped disease genes");
fprintf(f, "<PRE>");
for (tr = roughList; tr != NULL; tr = tr->next)
{
sprintf(chrom, "chr%s", tr->chromosome);
bandRange(chrom, tr->startBand, tr->endBand, &start, &end);
fprintf(f,
"<A HREF=\"http://www.ncbi.nlm.nih.gov/entrez/dispomim.cgi?id=%d\">",
tr->omimId);
fprintf(f, "OMIM %d</A>\t", tr->omimId);
fprintf(f,
"<A HREF=\"http://genome.ucsc.edu/cgi-bin/hgTracks?position=%s:%d-%d&pix=800\">", chrom, start, end);
fprintf(f, "%s:%d-%d</A>\t", chrom, start, end);
fprintf(f, "%s,%s\t", tr->startBand, tr->endBand);
fprintf(f, "%s\n", tr->description);
}
htmEnd(f);
fclose(f);
}
int main(int argc, char *argv[])
{
char *inName, *outName;
FILE *in, *out;
char line[256];
struct ernaHit *list = NULL, *el;
struct ernaClump *clumpList, *clump;
int clumpCount = 0;
if (argc != 3)
{
errAbort("sorterna - Process raw editRna file into a nice html.\n"
"usage: sorterna editrna.raw editrna.html");
}
inName = argv[1];
outName = argv[2];
in = mustOpen(inName, "r");
out = mustOpen(outName, "w");
/* Read in whole file. */
while (fgets(line, sizeof(line), in))
{
el = parseHit(line);
if (el->totalCdna >= el->commonErr + 2)
slAddHead(&list, el);
}
fclose(in);
printf("Got %d raw elements\n", slCount(list));
clumpList = clumpHits(list);
printf("Merged into %d clumps\n", slCount(clumpList));
slSort(&clumpList, cmpScore);
htmStart(out, "RNA Editing Candidates");
fprintf(out, "<H2>RNA Editing Candidates</H2>\n");
fprintf(out, "<TT><PRE>");
for (clump = clumpList; clump != NULL; clump = clump->next)
{
++clumpCount;
printf("Analysing clump %d\n", clumpCount);
fprintf(out, "<P><HR ALIGN=CENTER></P>");
fprintf(out, "<A HREF=\"../cgi-bin/tracks.exe?where=%s:%d-%d&hilite=%d-%d\">%s:%d-%d</A> ",
clump->chrom, clump->start-1000, clump->end+1000,
clump->start, clump->end+1,
clump->chrom, clump->start, clump->end+1);
fprintf(out, " score %f elements %d\n", clump->score, slCount(clump->hits));
showClump(clump, out);
}
htmEnd(out);
return 0;
}
void outputPicks(struct scoredWindow *winList, char *database,
struct hash *chromLimitHash, struct stats *stats, FILE *f)
/* Output picked regions. */
{
struct scoredWindow *strata[strataCount][strataCount];
double geneCuts[strataCount], consCuts[strataCount];
struct scoredWindow *win, *next;
int geneIx, consIx, pickIx;
FILE *html = NULL;
struct region *avoidList = NULL, *avoid;
/* Get list of regions to avoid */
if (avoidFile != NULL)
avoidList = loadRegionFile(database, avoidFile);
if (htmlOutput != NULL)
{
html = mustOpen(htmlOutput, "w");
htmStart(html, "Random Regions for 1% Project");
}
fprintf(f, "Cuts at:\n");
calcCuts(stats->consCounts, histSize,
stats->totalConsCount, 1.0, consCuts, strataCount);
uglyf("cons %f %f %f\n", consCuts[0], consCuts[1], consCuts[2]);
fprintf(f, "cons %f %f %f\n", consCuts[0], consCuts[1], consCuts[2]);
calcCuts(stats->geneCounts, histSize,
stats->totalGeneCount, 1.0/geneScale, geneCuts, strataCount);
uglyf("gene %f %f %f\n", geneCuts[0], geneCuts[1], geneCuts[2]);
uglyf("gene %f %f %f\n", geneCuts[0], geneCuts[1], geneCuts[2]);
fprintf(f, "\n");
/* Move winList to strata. */
zeroBytes(strata, sizeof(strata));
for (win = winList; win != NULL; win = next)
{
/* Calculate appropriate strata and move. */
next = win->next;
consIx = cutIx(consCuts, strataCount, win->consRatio);
geneIx = cutIx(geneCuts, strataCount, win->geneRatio);
slAddHead(&strata[consIx][geneIx], win);
}
/* Shuffle strata and output first picks in each. */
srand(randSeed);
for (consIx=strataCount-1; consIx>=0; --consIx)
{
for (geneIx=strataCount-1; geneIx>=0; --geneIx)
{
int cs=0, gs=0;
if (geneIx>0)
gs = round(100*genoCuts[geneIx-1]);
if (consIx>0)
cs = round(100*genoCuts[consIx-1]);
fprintf(f, "consNonTx %d%%-%d%%, gene %d%%-%d%%\n",
cs, round(100*genoCuts[consIx]),
gs, round(100*genoCuts[geneIx]));
if (html)
{
fprintf(html, "<H2>consNonTx %d%% - %d%%, gene %d%% - %d%%</H3>\n",
cs, round(100*genoCuts[consIx]),
gs, round(100*genoCuts[geneIx]));
}
shuffleList(&strata[consIx][geneIx]);
pickIx = 0;
for (win = strata[consIx][geneIx]; win != NULL; win = win->next)
{
int end = win->start + bigWinSize;
if (!hitsRegions(win->chrom, win->start, end, avoidList))
{
if (withinChromLimits(chromLimitHash, win->chrom))
{
AllocVar(avoid);
avoid->chrom = cloneString(win->chrom);
avoid->start = win->start;
avoid->end = end;
slAddHead(&avoidList, avoid);
fprintf(f, "%s:%d-%d\t", win->chrom, win->start+1, end);
fprintf(f, "consNonTx %4.1f%%, gene %4.1f%%\n",
100*win->consRatio, 100*win->geneRatio);
if (html)
{
fprintf(html, "<A HREF=\"http://genome.ucsc.edu/cgi-bin/");
fprintf(html, "hgTracks?db=%s&position=%s:%d-%d\">",
database, win->chrom, win->start+1, end);
fprintf(html, "%s:%d-%d</A>", win->chrom, win->start+1, end);
fprintf(html, "\tconsNonTx %4.1f%%, gene %4.1f%%<BR>\n",
100*win->consRatio, 100*win->geneRatio);
}
if (++pickIx >= picksPer)
break;
}
}
}
fprintf(f, "\n");
}
}
if (html)
{
htmEnd(html);
carefulClose(&html);
}
}
int main(int argc, char *argv[])
{
char *genoListName;
char *cdnaListName;
char *oocFileName;
char *pairFileName;
struct patSpace *patSpace;
long startTime, endTime;
char **genoList;
int genoListSize;
char *genoListBuf;
char **cdnaList;
int cdnaListSize;
char *cdnaListBuf;
char *genoName;
int i;
int estIx = 0;
struct dnaSeq **seqListList = NULL, *seq;
static char hitFileName[512], mergerFileName[512], okFileName[512];
char *outRoot;
struct hash *pairHash;
if (dumpMe)
{
bigHtmlFile = mustOpen("C:\\inetpub\\wwwroot\\test\\patAli.html", "w");
littleHtmlFile = mustOpen("C:\\inetpub\\wwwroot\\test\\patSpace.html", "w");
htmStart(bigHtmlFile, "PatSpace Alignments");
htmStart(littleHtmlFile, "PatSpace Index");
}
if ((hostName = getenv("HOST")) == NULL)
hostName = "";
if (argc != 6)
usage();
pushWarnHandler(patSpaceWarnHandler);
startTime = clock1000();
dnaUtilOpen();
makePolys();
genoListName = argv[1];
cdnaListName = argv[2];
oocFileName = argv[3];
pairFileName = argv[4];
outRoot = argv[5];
sprintf(hitFileName, "%s.hit", outRoot);
sprintf(mergerFileName, "%s.glu", outRoot);
sprintf(okFileName, "%s.ok", outRoot);
readAllWords(genoListName, &genoList, &genoListSize, &genoListBuf);
readAllWords(cdnaListName, &cdnaList, &cdnaListSize, &cdnaListBuf);
pairHash = makePairHash(pairFileName);
hitOut = mustOpen(hitFileName, "w");
mergerOut = mustOpen(mergerFileName, "w");
dumpOut = mustOpen("dump.out", "w");
seqListList = needMem(genoListSize*sizeof(seqListList[0]) );
fprintf(hitOut, "Pattern space 0.2 cDNA matcher\n");
fprintf(hitOut, "cDNA files: ", cdnaListSize);
for (i=0; i<cdnaListSize; ++i)
fprintf(hitOut, " %s", cdnaList[i]);
fprintf(hitOut, "\n");
fprintf(hitOut, "%d genomic files\n", genoListSize);
for (i=0; i<genoListSize; ++i)
{
genoName = genoList[i];
if (!startsWith("//", genoName) )
{
seqListList[i] = seq = faReadAllDna(genoName);
fprintf(hitOut, "%d els in %s ", slCount(seq), genoList[i]);
for (; seq != NULL; seq = seq->next)
fprintf(hitOut, "%d ", seq->size);
fprintf(hitOut, "\n");
}
}
patSpace = makePatSpace(seqListList, genoListSize, oocFileName);
for (i=0; i<cdnaListSize; ++i)
{
FILE *f;
char *estFileName;
DNA *dna;
char *estName;
int size;
int c;
int maxSizeForFuzzyFind = 20000;
int dotCount = 0;
estFileName = cdnaList[i];
if (startsWith("//", estFileName) )
continue;
f = mustOpen(estFileName, "rb");
while ((c = fgetc(f)) != EOF)
if (c == '>')
break;
printf("%s", cdnaList[i]);
fflush(stdout);
while (fastFaReadNext(f, &dna, &size, &estName))
{
aliSeqName = estName;
if (size < maxSizeForFuzzyFind) /* Some day need to fix this somehow... */
{
struct hashEl *hel;
struct cdnaAliList *calList = NULL;
hel = hashLookup(pairHash, estName);
if (hel != NULL) /* Do pair processing. */
{
struct estPair *ep;
struct seq *thisSeq, *otherSeq;
ep = hel->val;
if (hel->name == ep->name3)
{
thisSeq = &ep->seq3;
otherSeq = &ep->seq5;
}
else
{
thisSeq = &ep->seq5;
otherSeq = &ep->seq3;
}
if (otherSeq->dna == NULL) /* First in pair - need to save sequence. */
{
thisSeq->size = size;
thisSeq->dna = needMem(size);
memcpy(thisSeq->dna, dna, size);
}
else /* Second in pair - do gluing and free partner. */
{
char mergedName[64];
thisSeq->dna = dna;
thisSeq->size = size;
sprintf(mergedName, "%s_AND_%s", ep->name5, ep->name3);
patSpaceFindOne(patSpace, ep->seq5.dna, ep->seq5.size,
'+', '5', ep->name5, &calList);
reverseComplement(ep->seq5.dna, ep->seq5.size);
patSpaceFindOne(patSpace, ep->seq5.dna, ep->seq5.size,
'-', '5', ep->name5, &calList);
patSpaceFindOne(patSpace, ep->seq3.dna, ep->seq3.size,
'+', '3', ep->name3, &calList);
reverseComplement(ep->seq3.dna, ep->seq3.size);
patSpaceFindOne(patSpace, ep->seq3.dna, ep->seq3.size,
'-', '3', ep->name3, &calList);
slReverse(&calList);
writeMergers(calList, mergedName, genoList);
freez(&otherSeq->dna);
thisSeq->dna = NULL;
thisSeq->size =otherSeq->size = 0;
}
}
else
{
patSpaceFindOne(patSpace, dna, size, '+', '5', estName, &calList);
reverseComplement(dna, size);
patSpaceFindOne(patSpace, dna, size, '-', '5', estName, &calList);
slReverse(&calList);
writeMergers(calList, estName, genoList);
}
++estIx;
if ((estIx & 0xfff) == 0)
{
printf(".");
++dotCount;
fflush(stdout);
}
}
}
printf("\n");
}
aliSeqName = "";
printf("ffSubmitted %3d ffAccepted %3d ffOkScore %3d ffSolidMatch %2d\n",
ffSubmitted, ffAccepted, ffOkScore, ffSolidMatch);
endTime = clock1000();
printf("Total time is %4.2f\n", 0.001*(endTime-startTime));
/* Write out file who's presense say's we succeeded */
{
FILE *f = mustOpen(okFileName, "w");
fputs("ok", f);
fclose(f);
}
if (dumpMe)
{
htmEnd(bigHtmlFile);
htmEnd(littleHtmlFile);
}
return 0;
}
int main(int argc, char *argv[])
{
char *genoListName;
char *cdnaListName;
char *oocFileName;
char *hitFileName;
char *mergerFileName;
struct patSpace *patSpace;
long startTime, endTime;
char **genoList;
int genoListSize;
char *genoListBuf;
char **cdnaList;
int cdnaListSize;
char *cdnaListBuf;
char *genoName;
int i;
int estIx = 0;
struct dnaSeq **seqListList = NULL, *seq;
if (dumpMe)
{
bigHtmlFile = mustOpen("C:\\inetpub\\wwwroot\\test\\patAli.html", "w");
littleHtmlFile = mustOpen("C:\\inetpub\\wwwroot\\test\\patSpace.html", "w");
htmStart(bigHtmlFile, "PatSpace Alignments");
htmStart(littleHtmlFile, "PatSpace Index");
}
if (argc != 6)
usage();
startTime = clock1000();
dnaUtilOpen();
makePolys();
genoListName = argv[1];
cdnaListName = argv[2];
oocFileName = argv[3];
hitFileName = argv[4];
mergerFileName = argv[5];
readAllWords(genoListName, &genoList, &genoListSize, &genoListBuf);
readAllWords(cdnaListName, &cdnaList, &cdnaListSize, &cdnaListBuf);
hitOut = mustOpen(hitFileName, "w");
mergerOut = mustOpen(mergerFileName, "w");
dumpOut = mustOpen("dump.out", "w");
seqListList = needMem(genoListSize*sizeof(seqListList[0]) );
fprintf(hitOut, "Pattern space 0.2 cDNA matcher\n");
fprintf(hitOut, "cDNA files: ", cdnaListSize);
for (i=0; i<cdnaListSize; ++i)
fprintf(hitOut, " %s", cdnaList[i]);
fprintf(hitOut, "\n");
fprintf(hitOut, "%d genomic files\n", genoListSize);
for (i=0; i<genoListSize; ++i)
{
genoName = genoList[i];
if (!startsWith("//", genoName) )
{
seqListList[i] = seq = faReadAllDna(genoName);
fprintf(hitOut, "%d els in %s ", slCount(seq), genoList[i]);
for (; seq != NULL; seq = seq->next)
fprintf(hitOut, "%d ", seq->size);
fprintf(hitOut, "\n");
}
}
patSpace = makePatSpace(seqListList, genoListSize, oocFileName);
for (i=0; i<cdnaListSize; ++i)
{
FILE *f;
char *estFileName;
DNA *dna;
char *estName;
int size;
int c;
int maxSizeForFuzzyFind = 20000;
int dotCount = 0;
estFileName = cdnaList[i];
if (startsWith("//", estFileName) )
continue;
f = mustOpen(estFileName, "rb");
while ((c = fgetc(f)) != EOF)
if (c == '>')
break;
printf("%s", cdnaList[i]);
fflush(stdout);
while (fastFaReadNext(f, &dna, &size, &estName))
{
if (size < maxSizeForFuzzyFind) /* Some day need to fix this somehow... */
{
struct cdnaAliList *calList = NULL;
patSpaceFindOne(patSpace, dna, size, '+', estName, estIx, &calList);
reverseComplement(dna, size);
patSpaceFindOne(patSpace, dna, size, '-', estName, estIx, &calList);
slReverse(&calList);
writeMergers(calList, estName, size, genoList);
++estIx;
if ((estIx & 0xfff) == 0)
{
printf(".");
++dotCount;
fflush(stdout);
}
}
}
printf("\n");
}
printf("raw %4d ffSubmitted %3d ffAccepted %3d ffOkScore %3d ffSolidMatch %2d\n",
grandTotalHits, ffSubmitted, ffAccepted, ffOkScore, ffSolidMatch);
endTime = clock1000();
printf("Total time is %4.2f\n", 0.001*(endTime-startTime));
if (dumpMe)
{
htmEnd(bigHtmlFile);
htmEnd(littleHtmlFile);
}
return 0;
}
void writeTableHtml(struct sqlConnection *conn, struct eapGraph *eg,
struct fullExperiment *expList, char *assembly, char *outFile)
/* Write out a table in simple html based on eeList*/
{
char *lightGreen = "#D8FFD8";
char *brightWhite = "#FFFFFF";
FILE *f = mustOpen(outFile, "w");
htmStart(f, "DNase-seq experiment table");
fprintf(f, "<TABLE>\n");
fprintf(f, "<TR>\n");
fprintf(f, "<TH></TH>");
fprintf(f, "<TH>promo</TH>");
fprintf(f, "<TH>promo</TH>");
fprintf(f, "<TH>open</TH>");
fprintf(f, "<TH>open</TH>");
fprintf(f, "<TH>cross</TH>");
fprintf(f, "<TH> </TH>");
fprintf(f, "<TH></TH>");
fprintf(f, "<TH></TH>");
fprintf(f, "<TH></TH>");
fprintf(f, "<TH></TH>");
fprintf(f, "<TH></TH>");
fprintf(f, "<TH></TH>");
fprintf(f, "<TH></TH>");
fprintf(f, "</TR>\n");
fprintf(f, "<TR>\n");
fprintf(f, "<TH>Biosample</TH>");
fprintf(f, "<TH>enrich</TH>");
fprintf(f, "<TH>cover</TH>");
fprintf(f, "<TH>enrich</TH>");
fprintf(f, "<TH>cover</TH>");
fprintf(f, "<TH>enrich</TH>");
fprintf(f, "<TH> </TH>");
fprintf(f, "<TH BGCOLOR='%s'>reads(1)</TH>", lightGreen);
fprintf(f, "<TH BGCOLOR='%s'>map(1)</TH>", lightGreen);
fprintf(f, "<TH BGCOLOR='%s'>umap(1)</TH>", lightGreen);
fprintf(f, "<TH BGCOLOR='%s'>U4M(1)</TH>", lightGreen);
fprintf(f, "<TH BGCOLOR='%s'>Spot(1)</TH>", lightGreen);
fprintf(f, "<TH BGCOLOR='%s'>SpotX(1)</TH>", lightGreen);
fprintf(f, "<TH BGCOLOR='%s'>NSC(1)</TH>", lightGreen);
fprintf(f, "<TH BGCOLOR='%s'>RSC(1)</TH>", lightGreen);
fprintf(f, "<TH> </TH>");
fprintf(f, "<TH>reads(2)</TH>");
fprintf(f, "<TH>map(2)</TH>");
fprintf(f, "<TH>umap(2)</TH>");
fprintf(f, "<TH>U4M(2)</TH>");
fprintf(f, "<TH>Spot(2)</TH>");
fprintf(f, "<TH>SpotX(2)</TH>");
fprintf(f, "<TH>NSC(2)</TH>");
fprintf(f, "<TH>RSC(2)</TH>");
fprintf(f, "</TR>\n");
struct fullExperiment *exp;
for (exp = expList; exp != NULL; exp = exp->next)
{
struct edwExperiment *ee = exp->exp;
fprintf(f, "<TR>\n");
fprintf(f, "<TD>%s</TD>\n", ee->biosample);
long long pooledPeakId = pooledBroadPeaksForExp(exp);
if (pooledPeakId > 0)
{
struct edwBamFile *bamFileList = bamFilesAncestralToFile(conn, eg, pooledPeakId);
long long readCount = 0, mappedCount = 0, uniqueMappedCount = 0;
double u4mUniqueRatio = 0;
bamListStats(bamFileList, &readCount, &mappedCount, &uniqueMappedCount, &u4mUniqueRatio);
printEnrichment(conn, pooledPeakId, "promoter", f);
printEnrichment(conn, pooledPeakId, "open", f);
}
else
{
fprintf(f, "<TD>n/a</TD>\n");
fprintf(f, "<TD>n/a</TD>\n");
fprintf(f, "<TD>n/a</TD>\n");
fprintf(f, "<TD>n/a</TD>\n");
fprintf(f, "<TD>n/a</TD>\n");
fprintf(f, "<TD>n/a</TD>\n");
fprintf(f, "<TD>n/a</TD>\n");
}
uglyOne = (pooledPeakId == 74961);
struct edwQaPairSampleOverlap *so = findSampleOverlap(conn, eg, pooledPeakId);
if (so != NULL)
{
fprintf(f, "<TD>%4.2fx</TD>\n", so->sampleSampleEnrichment);
}
else
{
uglyf("Trouble finding sample/overlap for file %lld exp %s\n", pooledPeakId, exp->name);
fprintf(f, "<TD>n/a</TD>\n");
}
fprintf(f, "<TD> </TD>\n");
printOneRep(conn, eg, exp, mustFindReplicateInList(exp->repList, "1"), lightGreen, f);
fprintf(f, "<TH> </TH>");
printOneRep(conn, eg, exp, mustFindReplicateInList(exp->repList, "2"), brightWhite, f);
fprintf(f, "</TR>\n");
}
fprintf(f, "</TABLE>\n");
htmEnd(f);
carefulClose(&f);
}
int main(int argc, char *argv[])
{
char *outName;
char xaFileName[512];
char region[64];
FILE *xaFile, *out;
struct xaAli *xaList = NULL, *xa;
char *sortBy;
char *subtitle;
int (*cmp)(const void *va, const void *vb);
if (argc != 3)
{
usage();
}
sortBy = argv[1];
outName = argv[2];
if (sameWord(sortBy, "score"))
{
cmp = cmpXaScore;
subtitle = "(sorted by alignment score)";
}
else if (sameWord(sortBy, "briggsae"))
{
cmp = cmpXaQuery;
subtitle = "(sorted by <I>C. briggsae</I> region)";
}
else if (sameWord(sortBy, "elegans"))
{
cmp = cmpXaTarget;
subtitle = "(sorted by <I>C. elegans</I> region)";
}
else
usage();
/* Read in alignment file. */
sprintf(xaFileName, "%s%s/all%s", wormXenoDir(), "cbriggsae",
xaAlignSuffix());
printf("Scanning %s\n", xaFileName);
xaFile = xaOpenVerify(xaFileName);
while ((xa = xaReadNext(xaFile, FALSE)) != NULL)
{
xa->milliScore = round(0.001 * xa->milliScore * (xa->tEnd - xa->tStart));
freeMem(xa->qSym);
freeMem(xa->tSym);
freeMem(xa->hSym);
slAddHead(&xaList, xa);
}
/* Sort by score. */
printf("Sorting...");
slSort(&xaList, cmp);
printf(" best score %d\n", xaList->milliScore);
/* Write out .html */
printf("Writing %s\n", outName);
out = mustOpen(outName, "w");
htmStart(out, "C. briggsae/C. elegans Homologies");
fprintf(out, "<H2>Regions with Sequenced <I>C. briggsae</I> Homologs</H2>\n");
fprintf(out, "<H3>%s</H3>\n", subtitle);
fprintf(out, "<TT><PRE><B>");
fprintf(out, "Score <I>C. elegans Region</I> <I>C. briggsae</I> Region </B>\n");
fprintf(out, "--------------------------------------------------------\n");
for (xa = xaList; xa != NULL; xa = xa->next)
{
fprintf(out, "%6d ", xa->milliScore);
sprintf(region, "%s:%d-%d", xa->target, xa->tStart, xa->tEnd);
fprintf(out, "<A HREF=\"../cgi-bin/tracks.exe?where=%s\">%21s</A> %s:%d-%d %c",
region, region, xa->query, xa->qStart, xa->qEnd, xa->qStrand);
fprintf(out, "\n");
}
htmEnd(out);
return 0;
}
