void copyPslXaToTab(char *pslFile, char *tabFile)
/* copy a single PSL XA to the tab file */
{
struct xAli *xa;
char *row[23];
struct lineFile *lf = lineFileOpen(pslFile, TRUE);
struct pipeline *pl = NULL;
FILE *tabFh = NULL;
if (noSort)
tabFh = mustOpen(tabFile, "w");
else
{
if (pslCreateOpts & PSL_WITH_BIN)
pl = pipelineOpen(outPipeBin, pipelineWrite, tabFile, NULL);
else
pl = pipelineOpen(outPipeNoBin, pipelineWrite, tabFile, NULL);
tabFh = pipelineFile(pl);
}
while (lineFileRow(lf, row))
{
xa = xAliLoad(row);
if (pslCreateOpts & PSL_WITH_BIN)
fprintf(tabFh, "%u\t", hFindBin(xa->tStart, xa->tEnd));
xAliTabOut(xa, tabFh);
xAliFree(&xa);
}
lineFileClose(&lf);
if (noSort)
carefulClose(&tabFh);
else
{
pipelineWait(pl);
pipelineFree(&pl);
}
}
void writeResults(char *tableName)
/* read through preliminary table with chimp and macaque alleles. */
/* lookup into snpHash. */
{
struct hashEl *hel= NULL;
char query[512];
struct sqlConnection *conn = hAllocConn();
struct sqlResult *sr;
char **row;
struct snpSubset *subsetElement = NULL;
int bin = 0;
int start = 0;
int end = 0;
verbose(1, "write results...\n");
safef(query, sizeof(query), "select chrom, chromStart, chromEnd, name, species, strand, allele from %s", tableName);
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
hel = hashLookup(snpHash, row[3]);
if (hel == NULL) continue;
subsetElement = (struct snpSubset *)hel->val;
start = subsetElement->start;
end = subsetElement->end;
bin = hFindBin(start, end);
fprintf(outputFileHandle, "%d\t%s\t%d\t%d\t%s\t", bin, subsetElement->chrom, start, end, row[3]);
fprintf(outputFileHandle, "0\t%s\t%s\t%s\t", subsetElement->strand, subsetElement->refUCSC, subsetElement->observed);
fprintf(outputFileHandle, "%s\t%s\t%s\t%s\t%s\t%s\n", row[4], row[0], row[1], row[2], row[5], row[6]);
}
}
void outputSummary(FILE *f, struct mafSummary *ms)
/* output to .tab file */
{
fprintf(f, "%u\t", hFindBin(ms->chromStart, ms->chromEnd));
mafSummaryTabOut(ms, f);
summaryCount++;
}
void copyPslToTab(char *pslFile, char *tabFile)
/* copy a single PSL to the tab file */
{
struct psl *psl;
struct lineFile *lf = pslFileOpen(pslFile);
struct pipeline *pl = NULL;
FILE *tabFh = NULL;
if (noSort)
tabFh = mustOpen(tabFile, "w");
else
{
if (pslCreateOpts & PSL_WITH_BIN)
pl = pipelineOpen(outPipeBin, pipelineWrite, tabFile, NULL);
else
pl = pipelineOpen(outPipeNoBin, pipelineWrite, tabFile, NULL);
tabFh = pipelineFile(pl);
}
while ((psl = pslNext(lf)) != NULL)
{
if (pslCreateOpts & PSL_WITH_BIN)
fprintf(tabFh, "%u\t", hFindBin(psl->tStart, psl->tEnd));
pslTabOut(psl, tabFh);
pslFree(&psl);
}
lineFileClose(&lf);
if (noSort)
carefulClose(&tabFh);
else
{
pipelineWait(pl);
pipelineFree(&pl);
}
}
void hgPhMouse(char *database, char *track, int fileCount, char *fileNames[])
/* hgPhMouse - Load phMouse track. */
{
int i;
char *fileName;
char *tabName = "phMouse.tab";
FILE *f = mustOpen(tabName, "w");
struct lineFile *lf;
char *words[32], *s, c;
int wordCount;
int oneSize, totalSize = 0;
for (i=0; i<fileCount; ++i)
{
struct bed *bedList = NULL, *bed;
fileName = fileNames[i];
lf = lineFileOpen(fileName, TRUE);
printf("Reading %s ", fileName);
fflush(stdout);
while ((wordCount = lineFileChop(lf, words)) > 0)
{
if (wordCount < 7)
errAbort("Expecting at least 7 words line %d of %s",
lf->lineIx, fileName);
AllocVar(bed);
bed->chrom = cloneString(words[0]);
bed->chromStart = lineFileNeedNum(lf, words, 1);
bed->chromEnd = lineFileNeedNum(lf, words, 2);
bed->score = lineFileNeedNum(lf, words, 6);
s = strrchr(words[3], '|');
c = s[1];
s[0] = 0;
if (c != '+' && c != '-')
errAbort("Misformed strandless trace name line %d of %s",
lf->lineIx, lf->fileName);
bed->name = cloneString(words[3]);
bed->strand[0] = c;
slAddHead(&bedList, bed);
}
oneSize = slCount(bedList);
printf("%d alignments ", oneSize);
totalSize += oneSize;
fflush(stdout);
slSort(&bedList, bedCmp);
printf("sorted ");
fflush(stdout);
for (bed = bedList; bed != NULL; bed = bed->next)
{
int bin = hFindBin(bed->chromStart, bed->chromEnd);
fprintf(f, "%d\t", bin);
bedTabOutN(bed, 6, f);
}
printf("tabbed out\n");
bedFreeList(&bedList);
}
carefulClose(&f);
printf("Loading %d items into %s.%s\n", totalSize, database, track);
loadDatabase(database, track, tabName);
remove(tabName);
}
void writeBedTab(char *fileName, struct bedStub *bedList, int bedSize)
/* Write out bed list to tab-separated file. */
{
struct bedStub *bed;
FILE *f = mustOpen(fileName, "w");
char *words[64];
int i, wordCount;
for (bed = bedList; bed != NULL; bed = bed->next)
{
if (!noBin)
if (fprintf(f, "%u\t", hFindBin(bed->chromStart, bed->chromEnd)) <= 0)
writeFailed(fileName);
if (strictTab)
wordCount = chopTabs(bed->line, words);
else
wordCount = chopLine(bed->line, words);
for (i=0; i<wordCount; ++i)
{
/* new definition for old "reserved" field, now itemRgb */
/* and when itemRgb, it is a comma separated string r,g,b */
if (itemRgb && (i == 8))
{
char *comma;
/* Allow comma separated list of rgb values here */
comma = strchr(words[8], ',');
if (comma)
{
int itemRgb = 0;
if (-1 == (itemRgb = bedParseRgb(words[8])))
errAbort("ERROR: expecting r,g,b specification, "
"found: '%s'", words[8]);
else
if (fprintf(f, "%d", itemRgb) <= 0)
writeFailed(fileName);
verbose(2, "itemRgb: %s, rgb: %#x\n", words[8], itemRgb);
}
else
if (fputs(words[i], f) == EOF)
writeFailed(fileName);
}
else
if (fputs(words[i], f) == EOF)
writeFailed(fileName);
if (i == wordCount-1)
{
if (fputc('\n', f) == EOF)
writeFailed(fileName);
}
else
if (fputc('\t', f) == EOF)
writeFailed(fileName);
}
}
fclose(f);
}
void gapFileToTable(struct sqlConnection *conn, char *gapFileName,
char *gapTableName)
/* Build a single gap table from a single gap file. */
{
struct lineFile *lf = lineFileOpen(gapFileName, TRUE);
char tabFileName[256];
FILE *tabFile = NULL;
char *words[16];
int wordCount;
safef(tabFileName, sizeof(tabFileName), "%s.tab", gapTableName);
tabFile = mustOpen(tabFileName, "w");
while ((wordCount = lineFileChop(lf, words)) > 0)
{
if (wordCount < 5)
errAbort("Short line %d of %s", lf->lineIx, lf->fileName);
if (words[4][0] == 'N' || words[4][0] == 'U')
{
int len = strlen(words[0]);
if (len > maxChromNameSize)
{
maxChromNameSize = len;
if (maxChromNameSize > 254)
errAbort("ERROR: chrom name size is over 254(%d) characters: "
"'%s'", maxChromNameSize, words[0]);
}
struct agpGap gap;
agpGapStaticLoad(words, &gap);
gap.chromStart -= 1;
fprintf(tabFile, "%u\t", hFindBin(gap.chromStart, gap.chromEnd));
agpGapTabOut(&gap, tabFile);
}
}
lineFileClose(&lf);
fclose(tabFile);
if (! noLoad)
{
struct dyString *ds = newDyString(2048);
if (unsplit)
sqlDyStringPrintf(ds, createGapUnsplit, gapTableName,
maxChromNameSize, maxChromNameSize);
else
sqlDyStringPrintf(ds, createGapSplit, gapTableName);
char query[1024];
sqlRemakeTable(conn, gapTableName, ds->string);
sqlSafef(query, sizeof(query), "LOAD data local infile '%s' into table %s",
tabFileName, gapTableName);
sqlUpdate(conn, query);
remove(tabFileName);
freeDyString(&ds);
}
}
void readFreq()
{
FILE *outputFileHandle = mustOpen("snpFreq.tab", "w");
FILE *logFileHandle = mustOpen("snpFreq.log", "w");
char query[512];
struct sqlConnection *conn = hAllocConn();
struct sqlResult *sr;
char **row;
struct hashEl *alleleHashEl = NULL;
struct hashEl *snpHashEl = NULL;
struct coords *coordsInstance = NULL;
char snpName[32];
int bin = 0;
safef(query, sizeof(query), "select snp_id, allele_id, freq from SNPAlleleFreq");
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
alleleHashEl = hashLookup(alleleHash, row[1]);
if (alleleHashEl == NULL)
{
fprintf(logFileHandle, "couldn't find allele_id %s\n", row[1]);
continue;
}
safef(snpName, sizeof(snpName), "rs%s", row[0]);
snpHashEl = hashLookup(snpHash, snpName);
if (snpHashEl == NULL)
{
fprintf(logFileHandle, "skipping snp_id %s\n", row[0]);
continue;
}
coordsInstance = (struct coords *)snpHashEl->val;
/* could add bin here */
bin = hFindBin(coordsInstance->start, coordsInstance->end);
fprintf(outputFileHandle, "%d\t%s\t%d\t%d\t%s\t%s\t%f\n",
bin, coordsInstance->chrom, coordsInstance->start, coordsInstance->end,
snpName, (char *)alleleHashEl->val, sqlFloat(row[2]));
}
carefulClose(&outputFileHandle);
carefulClose(&logFileHandle);
sqlFreeResult(&sr);
hFreeConn(&conn);
}
void addGlBin(char *in, char *out)
/* Copy in to out, but adding bin field in first column. */
{
char *row[4];
int i, start, end;
struct lineFile *lf = lineFileOpen(in, TRUE);
FILE *f = mustOpen(out, "w");
while (lineFileRow(lf, row))
{
start = sqlUnsigned(row[1]);
end = sqlUnsigned(row[2]);
fprintf(f, "%u", hFindBin(start, end));
for (i=0; i<ArraySize(row); ++i)
fprintf(f, "\t%s", row[i]);
fprintf(f, "\n");
}
carefulClose(&f);
lineFileClose(&lf);
}
static void gbGeneTblWriteGene(struct gbGeneTbl *ggt, struct gbStatus* status,
struct psl* psl, struct sqlConnection *conn)
/* write genePred row */
{
struct genePred* gp
= genePredFromPsl3(psl, &status->cds,
(ggt->hasExtCols ? genePredAllFlds : 0), genePredPslCdsMod3,
genePredStdInsertMergeSize, genePredStdInsertMergeSize);
FILE *fh = gbGeneTblGetTabFh(ggt, conn);
if (ggt->hasExtCols)
{
/* add gene name */
freeMem(gp->name2);
gp->name2 = cloneString(status->geneName);
}
if (ggt->hasBin)
fprintf(fh, "%u\t", hFindBin(gp->txStart, gp->txEnd));
genePredTabOut(gp, fh);
genePredFree(&gp);
}
void copyGene(char *db, struct genePred *gene, FILE *tabFh)
/* copy one gene to the tab file */
{
unsigned holdOptFields = gene->optFields;
unsigned optFields = (genePredScoreFld|genePredName2Fld|genePredCdsStatFld|genePredExonFramesFld);
if (gGenePredExt && ((optFields & optFields) != optFields))
errAbort("genePred %s doesn't have fields required for -genePredExt", gene->name);
if (gNoValidate || checkGene(db, gene))
{
if (!gGenePredExt)
gene->optFields = 0; /* omit optional fields */
if (gBin)
fprintf(tabFh, "%u\t", hFindBin(gene->txStart, gene->txEnd));
genePredTabOut(gene, tabFh);
gene->optFields = holdOptFields; /* restore optional fields */
}
}
void getSnps()
{
char query[512];
struct sqlConnection *conn = hAllocConn();
struct sqlResult *sr;
char **row;
int start = 0;
int end = 0;
int bin = 0;
sqlSafef(query, sizeof(query), "select * from chrX_snp126 where chromEnd < 2709520");
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
fprintf(outputFileHandle, "%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t",
row[0], "chrY", row[2], row[3], row[4], row[5], row[6], row[7], row[8]);
fprintf(outputFileHandle, "%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n",
row[9], row[10], row[11], row[12], row[13], row[14], row[15], row[16], row[17]);
}
sqlFreeResult(&sr);
sqlSafef(query, sizeof(query), "select * from chrX_snp126 where chromEnd > 154584237");
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
start = sqlUnsigned(row[2]);
end = sqlUnsigned(row[3]);
start = start - 97140800;
end = end - 97140800;
bin = hFindBin(start, end);
fprintf(outputFileHandle, "%d\t%s\t%d\t%d\t%s\t%s\t%s\t%s\t%s\t",
bin, "chrY", start, end, row[4], row[5], row[6], row[7], row[8]);
fprintf(outputFileHandle, "%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n",
row[9], row[10], row[11], row[12], row[13], row[14], row[15], row[16], row[17]);
}
sqlFreeResult(&sr);
hFreeConn(&conn);
}
void readOneOut(char *rmskFile)
/* Read .out file rmskFile, check each line, and print OK lines to .tab. */
{
struct lineFile *lf;
char *line, *words[24];
int lineSize, wordCount;
/* Open .out file and process header. */
lf = lineFileOpen(rmskFile, TRUE);
if (!lineFileNext(lf, &line, &lineSize))
errAbort("Empty %s", lf->fileName);
if (!startsWith(" SW perc perc", line))
{
if (!startsWith(" SW perc perc", line))
errAbort("%s doesn't seem to be a RepeatMasker .out file, first "
"line seen:\n%s", lf->fileName, line);
}
lineFileNext(lf, &line, &lineSize);
lineFileNext(lf, &line, &lineSize);
/* Process line oriented records of .out file. */
while (lineFileNext(lf, &line, &lineSize))
{
static struct rmskOut2 r;
char *s;
wordCount = chopLine(line, words);
if (wordCount < 14)
errAbort("Expecting 14 or 15 words line %d of %s",
lf->lineIx, lf->fileName);
r.swScore = atoi(words[0]);
r.milliDiv = makeMilli(words[1], lf);
r.milliDel = makeMilli(words[2], lf);
r.milliIns = makeMilli(words[3], lf);
r.genoName = words[4];
r.genoStart = atoi(words[5])-1;
r.genoEnd = atoi(words[6]);
r.genoLeft = parenInt(words[7], lf);
r.strand[0] = (words[8][0] == '+' ? '+' : '-');
r.repName = words[9];
r.repClass = words[10];
char *repClassTest = cloneString(r.repClass);
stripChar(repClassTest, '(');
stripChar(repClassTest, ')');
int nonDigitCount = countLeadingNondigits(repClassTest);
int wordOffset = 0;
// this repClass is only digits, (or only (digits) with surrounding parens)
// this is the sign of an empty field here
// due to custom library in use that has no class/family indication
if (0 == nonDigitCount)
{
wordOffset = 1;
r.repClass = cloneString("Unspecified");
r.repFamily = cloneString("Unspecified");
}
else
{
s = strchr(r.repClass, '/');
if (s == NULL)
r.repFamily = r.repClass;
else
{
*s++ = 0;
r.repFamily = s;
}
}
r.repStart = parenInt(words[11-wordOffset], lf);
r.repEnd = atoi(words[12-wordOffset]);
r.repLeft = parenInt(words[13-wordOffset], lf);
r.id = atoi(words[14-wordOffset]);
if (words[8][0] == 'C')
{
r.repLeft = parenInt(words[11-wordOffset], lf);
r.repStart = parenInt(words[13-wordOffset], lf);
}
if (checkRepeat(&r, lf))
{
FILE *f = getFileForChrom(r.genoName);
if (!noBin)
fprintf(f, "%u\t", hFindBin(r.genoStart, r.genoEnd));
rmskOut2TabOut(&r, f);
}
}
}
void writeBedTab(char *fileName, struct bedStub *bedList)
/* Write out bed list to tab-separated file. */
{
struct bedStub *bed;
FILE *f = mustOpen(fileName, "w");
char *words[64];
int i, wordCount;
for (bed = bedList; bed != NULL; bed = bed->next)
{
if (!noBin)
{
// allow for zero-length at start of seq [bin code can't handle 0-0]
unsigned end = (bed->chromEnd > 0) ? bed->chromEnd : 1;
if (fprintf(f, "%u\t", hFindBin(bed->chromStart, end)) <= 0)
writeFailed(fileName);
}
if (strictTab)
wordCount = chopTabs(bed->line, words);
else
wordCount = chopLine(bed->line, words);
for (i=0; i<wordCount; ++i)
{
/* new definition for old "reserved" field, now itemRgb */
/* and when itemRgb, it is a comma separated string r,g,b */
if (itemRgb && (i == 8))
{
char *comma;
/* Allow comma separated list of rgb values here */
comma = strchr(words[8], ',');
if (comma)
{
int itemRgb = 0;
if (-1 == (itemRgb = bedParseRgb(words[8])))
errAbort("ERROR: expecting r,g,b specification, "
"found: '%s'", words[8]);
else
if (fprintf(f, "%d", itemRgb) <= 0)
writeFailed(fileName);
verbose(2, "itemRgb: %s, rgb: %#x\n", words[8], itemRgb);
}
else
if (fputs(words[i], f) == EOF)
writeFailed(fileName);
}
else if ((dotIsNull > 0) && (dotIsNull == i) && sameString(words[i],"."))
/* If the . was used to represent NULL, replace with -1 in the tables */
{
if (fputs("-1", f) == EOF)
writeFailed(fileName);
}
else
if (fputs(words[i], f) == EOF)
writeFailed(fileName);
if (i == wordCount-1)
{
if (fputc('\n', f) == EOF)
writeFailed(fileName);
}
else
if (fputc('\t', f) == EOF)
writeFailed(fileName);
}
}
fclose(f);
}
void writeResults(char *tableName)
{
char query[512];
struct sqlConnection *conn = hAllocConn();
struct sqlResult *sr;
char **row;
struct hashEl *helChimp = NULL, *helMacaque = NULL;
struct orthoSnp *chimpData, *macaqueData, *missingData = NULL;
char *snpId;
char *humanObserved;
char *humanAllele;
char *humanStrand;
int humanCount = 0;
int chimpOnlyCount = 0;
int macaqueOnlyCount = 0;
int missingCount = 0;
int bothCount = 0;
int chromStart = 0;
int chromEnd = 0;
int binVal = 0;
AllocVar(missingData);
missingData->chrom = cloneString("?");
missingData->start = 0;
missingData->end = 0;
missingData->strand = cloneString("?");
missingData->allele = cloneString("?");
sqlSafef(query, sizeof(query), "select chrom, chromStart, chromEnd, name, strand, refUCSC, observed from %s", tableName);
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
chromStart = sqlUnsigned(row[1]);
chromEnd = sqlUnsigned(row[2]);
snpId = cloneString(row[3]);
humanStrand = cloneString(row[4]);
humanAllele = cloneString(row[5]);
humanObserved = cloneString(row[6]);
humanCount++;
helChimp = hashLookup(chimpHash, snpId);
if (helChimp != NULL)
chimpData = (struct orthoSnp *)helChimp->val;
else
chimpData = missingData;
helMacaque = hashLookup(macaqueHash, snpId);
if (helMacaque != NULL)
macaqueData = (struct orthoSnp *)helMacaque->val;
else
macaqueData = missingData;
if (!helChimp && !helMacaque)
{
fprintf(missingFileHandle, "%s not found\n", snpId);
missingCount++;
continue;
}
if (helChimp && helMacaque)
bothCount++;
if (!helChimp && helMacaque)
macaqueOnlyCount++;
if (helChimp && !helMacaque)
chimpOnlyCount++;
binVal = hFindBin(chromStart, chromEnd);
fprintf(outputFileHandle, "%d\t%s\t%s\t%s\t", binVal, row[0], row[1], row[2]);
fprintf(outputFileHandle, "%s\t%s\t%s\t%s\t", snpId, humanObserved, humanAllele, humanStrand);
fprintf(outputFileHandle, "%s\t%d\t%d\t", chimpData->chrom, chimpData->start, chimpData->end);
fprintf(outputFileHandle, "%s\t%s\t", chimpData->allele, chimpData->strand);
fprintf(outputFileHandle, "%s\t%d\t%d\t", macaqueData->chrom, macaqueData->start, macaqueData->end);
fprintf(outputFileHandle, "%s\t%s\n", macaqueData->allele, macaqueData->strand);
}
verbose(1, "humanCount = %d\n", humanCount);
verbose(1, "chimpOnlyCount = %d\n", chimpOnlyCount);
verbose(1, "macaqueOnlyCount = %d\n", macaqueOnlyCount);
verbose(1, "missingCount = %d\n", missingCount);
verbose(1, "bothCount = %d\n", bothCount);
}
void cnWriteTables(char *chrom, struct cnFill *fillList, FILE *f, int depth)
/* Recursively write out fill and gap lists. */
{
char qName[64];
struct cnFill *fill;
for (fill = fillList; fill != NULL; fill = fill->next)
{
if (fill->chainId != 0)
{
if (fill->type == NULL)
errAbort("No type field, please run netSyntenic on input");
if (fill->tN < 0)
{
if (warnFlag)
{
if (!warned)
{
fprintf(stderr, "Warning: missing fields\n");
warned = TRUE;
}
}
else
errAbort("Missing fields. Please run netClass on input");
}
}
if (fill->score < 0)
fill->score = 0;
if (!noBin)
fprintf(f, "%d\t", hFindBin(fill->tStart, fill->tStart + fill->tSize));
qName[0] = 0;
if (qPrefix != NULL)
{
strcat(qName, qPrefix);
strcat(qName, "-");
}
strcat(qName, fill->qName);
fprintf(f, "%d\t", depth);
fprintf(f, "%s\t", chrom);
fprintf(f, "%d\t", fill->tStart);
fprintf(f, "%d\t", fill->tStart + fill->tSize);
fprintf(f, "%c\t", fill->qStrand);
fprintf(f, "%s\t", qName);
fprintf(f, "%d\t", fill->qStart);
fprintf(f, "%d\t", fill->qStart + fill->qSize);
fprintf(f, "%d\t", fill->chainId);
fprintf(f, "%d\t", fill->ali);
fprintf(f, "%1.1f\t", fill->score);
fprintf(f, "%d\t", fill->qOver);
fprintf(f, "%d\t", fill->qFar);
fprintf(f, "%d\t", fill->qDup);
fprintf(f, "%s\t", (fill->type == NULL ? "gap" : fill->type));
fprintf(f, "%d\t", fill->tN);
fprintf(f, "%d\t", fill->qN);
fprintf(f, "%d\t", fill->tR);
fprintf(f, "%d\t", fill->qR);
fprintf(f, "%d\t", fill->tNewR);
fprintf(f, "%d\t", fill->qNewR);
fprintf(f, "%d\t", fill->tOldR);
fprintf(f, "%d\t", fill->qOldR);
fprintf(f, "%d\t", fill->tTrf);
fprintf(f, "%d\n", fill->qTrf);
if (fill->children)
cnWriteTables(chrom, fill->children, f, depth+1);
}
}
void splitAgp(char *agpName, char *goldFileName, char *gapFileName)
/* Split up agp file into gold and gap files. */
{
struct lineFile *lf;
char *words[16];
int wordCount;
FILE *goldTab, *gapTab;
/* Scan through .agp file splitting it into gold
* and gap components. */
goldTab = mustOpen(goldFileName, "w");
gapTab = mustOpen(gapFileName, "w");
lf = lineFileOpen(agpName, TRUE);
while ((wordCount = lineFileChop(lf, words)) > 0)
{
int start, end;
if (wordCount < 5)
errAbort("Short line %d of %s", lf->lineIx, lf->fileName);
int len = strlen(words[0]);
if (len > maxChromNameSize)
{
maxChromNameSize = len;
if (maxChromNameSize > 254)
errAbort("ERROR: chrom name size is over 254(%d) characters: "
"'%s'", maxChromNameSize, words[0]);
}
start = sqlUnsigned(words[1])-1;
end = sqlUnsigned(words[2]);
if (words[4][0] == 'N' || words[4][0] == 'U')
{
struct agpGap gap;
agpGapStaticLoad(words, &gap);
gap.chromStart -= 1;
fprintf(gapTab, "%u\t", hFindBin(start, end));
agpGapTabOut(&gap, gapTab);
verbose(3,"#GAP\t%s:%d-%d\n", gap.chrom, gap.chromStart, gap.chromEnd);
}
else
{
struct agpFrag gold;
agpFragStaticLoad(words, &gold);
agpFragValidate(&gold);
len = strlen(words[5]);
if (len > maxFragNameSize)
{
maxFragNameSize = len;
if (maxFragNameSize > 254)
errAbort("ERROR: fragment name size is over 254(%d) "
"characters: '%s'", maxFragNameSize, words[5]);
}
// file is 1-based. agpFragLoad() now assumes 0-based.
// and agpFragTabOut() will assume 1-based, but we will load
// the generated file straight into the database, so
// subtract 2:
gold.chromStart -= 2;
gold.fragStart -= 2;
fprintf(goldTab, "%u\t", hFindBin(start, end));
agpFragTabOut(&gold, goldTab);
}
}
lineFileClose(&lf);
carefulClose(&goldTab);
carefulClose(&gapTab);
}