int main(int argc,char **argv) {
int count;
bode::Interval *i;
std::string readsFN(argv[1]);
std::string inFN(argv[2]);
std::string outFN(argv[3]);
int target[16384];
bode::IntervalReader *reads = bode::IntervalReader::open(readsFN);
bode::BedReader *in = new bode::BedReader(inFN);
std::ofstream *out = new std::ofstream(outFN.c_str(),std::ofstream::out);
bode::IntervalSet *is = loadReads(reads);
reads->close();
while (i=in->nextI()) {
int left = i->left();
int right = i->right();
if (right - left > 16384) {
std::cerr << "over-wide interval: "<<i->chrom()<<":"<<left<<"-"<<right<<std::endl;
} else {
is->density(i->chrom(),left,right,target);
*out <<i->chrom()<<":"<<left<<"-"<<right;
for (int j=0;j<right-left;j++) {
*out << " " << target[j];
}
*out << std::endl;
}
count++;
}
in->close();
out->close();
std::cerr << "processed " << count << " intervals" << std::endl;
return 0;
}
struct read* readsGetNext(Reads* reads)
{
struct read* result;
// Do we need to load more reads?
if(reads->current % BATCH_SIZE == 0)
{
loadReads(reads);
}
result = &(reads->readData[reads->current % BATCH_SIZE]);
reads->current = reads->current + 1;
return result;
}
int
main(int argc, char **argv) {
char *gkpStoreName = NULL;
char *outPrefix = NULL;
uint32 minReadLength = 0;
uint32 firstFileArg = 0;
char errorLogName[FILENAME_MAX];
char htmlLogName[FILENAME_MAX];
char nameMapName[FILENAME_MAX];
argc = AS_configure(argc, argv);
int arg = 1;
int err = 0;
while (arg < argc) {
if (strcmp(argv[arg], "-o") == 0) {
gkpStoreName = argv[++arg];
} else if (strcmp(argv[arg], "-minlength") == 0) {
minReadLength = atoi(argv[++arg]);
} else if (strcmp(argv[arg], "--") == 0) {
firstFileArg = arg++;
break;
} else if (argv[arg][0] == '-') {
fprintf(stderr, "ERROR: unknown option '%s'\n", argv[arg]);
err++;
} else {
firstFileArg = arg;
break;
}
arg++;
}
if (gkpStoreName == NULL)
err++;
if (firstFileArg == 0)
err++;
if (err) {
fprintf(stderr, "usage: %s [...] -o gkpStore\n", argv[0]);
fprintf(stderr, " -o gkpStore create this gkpStore\n");
fprintf(stderr, " \n");
fprintf(stderr, " -minlength L discard reads shorter than L\n");
fprintf(stderr, " \n");
fprintf(stderr, " \n");
if (gkpStoreName == NULL)
fprintf(stderr, "ERROR: no gkpStore (-g) supplied.\n");
if (firstFileArg == 0)
fprintf(stderr, "ERROR: no input files supplied.\n");
exit(1);
}
gkStore *gkpStore = gkStore::gkStore_open(gkpStoreName, gkStore_extend);
gkRead *gkpRead = NULL;
gkLibrary *gkpLibrary = NULL;
uint32 gkpFileID = 0; // Used for HTML output, an ID for each file loaded.
uint32 inLineLen = 1024;
char inLine[1024] = { 0 };
validSeq['a'] = validSeq['c'] = validSeq['g'] = validSeq['t'] = validSeq['n'] = 1;
validSeq['A'] = validSeq['C'] = validSeq['G'] = validSeq['T'] = validSeq['N'] = 1;
errno = 0;
sprintf(errorLogName, "%s/errorLog", gkpStoreName);
FILE *errorLog = fopen(errorLogName, "w");
if (errno)
fprintf(stderr, "ERROR: cannot open error file '%s': %s\n", errorLogName, strerror(errno)), exit(1);
sprintf(htmlLogName, "%s/load.dat", gkpStoreName);
FILE *htmlLog = fopen(htmlLogName, "w");
if (errno)
fprintf(stderr, "ERROR: cannot open uid map file '%s': %s\n", htmlLogName, strerror(errno)), exit(1);
sprintf(nameMapName, "%s/readNames.txt", gkpStoreName);
FILE *nameMap = fopen(nameMapName, "w");
if (errno)
fprintf(stderr, "ERROR: cannot open uid map file '%s': %s\n", nameMapName, strerror(errno)), exit(1);
uint32 nERROR = 0; // There aren't any errors, we just exit fatally if encountered.
uint32 nWARNS = 0;
uint32 nLOADED = 0; // Reads loaded
uint64 bLOADED = 0; // Bases loaded
uint32 nSKIPPED = 0;
uint64 bSKIPPED = 0; // Bases not loaded, too short
#if 0
fprintf(htmlLog, "<!DOCTYPE html>\n");
fprintf(htmlLog, "<html>\n");
fprintf(htmlLog, "<head>\n");
fprintf(htmlLog, "<title>gatekeeper load statistics</title>\n");
fprintf(htmlLog, "<style type='text/css'>\n");
fprintf(htmlLog, "body { font-family: Helvetica, Verdana, sans-serif; }\n");
fprintf(htmlLog, "h1, h2 { color: #ee3e80; }\n");
fprintf(htmlLog, "p { color: #665544; }\n");
fprintf(htmlLog, "th, td { border: 1px solid #111111; padding: 2px 2px 2px 2px; }\n");
fprintf(htmlLog, "td:hover { background-color: #e4e4e4; }\n");
fprintf(htmlLog, "th:hover { background-color: #d4d4d4; }\n");
fprintf(htmlLog, "tr.details { visibility: collapse; }\n");
fprintf(htmlLog, "</style>\n");
fprintf(htmlLog, "</head>\n");
fprintf(htmlLog, "<body>\n");
fprintf(htmlLog, "<h2>Input Files</h2>\n");
fprintf(htmlLog, "<table>\n");
#endif
for (; firstFileArg < argc; firstFileArg++) {
fprintf(stderr, "\n");
fprintf(stderr, "Starting file '%s'.\n", argv[firstFileArg]);
compressedFileReader *inFile = new compressedFileReader(argv[firstFileArg]);
char *line = new char [10240];
KeyAndValue keyval;
while (fgets(line, 10240, inFile->file()) != NULL) {
chomp(line);
keyval.find(line);
if (keyval.key() == NULL) {
// No key, so must be a comment or blank line
continue;
}
if (strcasecmp(keyval.key(), "name") == 0) {
gkpLibrary = gkpStore->gkStore_addEmptyLibrary(keyval.value());
continue;
}
// We'd better have a gkpLibrary defined, if not, the .gkp input file is incorrect.
if (gkpLibrary == NULL) {
fprintf(stderr, "WARNING: no 'name' tag in gkp input; creating library with name 'DEFAULT'.\n");
gkpLibrary = gkpStore->gkStore_addEmptyLibrary(keyval.value());
nWARNS++;
}
if (strcasecmp(keyval.key(), "preset") == 0) {
gkpLibrary->gkLibrary_parsePreset(keyval.value());
} else if (strcasecmp(keyval.key(), "qv") == 0) {
gkpLibrary->gkLibrary_setDefaultQV(keyval.value_double());
} else if (strcasecmp(keyval.key(), "isNonRandom") == 0) {
gkpLibrary->gkLibrary_setIsNonRandom(keyval.value_bool());
} else if (strcasecmp(keyval.key(), "trustHomopolymerRuns") == 0) {
gkpLibrary->gkLibrary_setTrustHomopolymerRuns(keyval.value_bool());
} else if (strcasecmp(keyval.key(), "removeDuplicateReads") == 0) {
gkpLibrary->gkLibrary_setRemoveDuplicateReads(keyval.value_bool());
} else if (strcasecmp(keyval.key(), "finalTrim") == 0) {
gkpLibrary->gkLibrary_setFinalTrim(keyval.value());
} else if (strcasecmp(keyval.key(), "removeSpurReads") == 0) {
gkpLibrary->gkLibrary_setRemoveSpurReads(keyval.value_bool());
} else if (strcasecmp(keyval.key(), "removeChimericReads") == 0) {
gkpLibrary->gkLibrary_setRemoveChimericReads(keyval.value_bool());
} else if (strcasecmp(keyval.key(), "checkForSubReads") == 0) {
gkpLibrary->gkLibrary_setCheckForSubReads(keyval.value_bool());
} else if (AS_UTL_fileExists(keyval.key(), false, false)) {
loadReads(gkpStore,
gkpLibrary,
gkpFileID++,
minReadLength,
nameMap,
htmlLog,
errorLog,
keyval.key(),
nWARNS, nLOADED, bLOADED, nSKIPPED, bSKIPPED);
} else {
fprintf(stderr, "ERROR: option '%s' not recognized, and not a file of reads.\n", line);
exit(1);
}
}
delete inFile;
delete [] line;
}
#if 0
fprintf(htmlLog, "</table>\n");
#endif
gkpStore->gkStore_close();
fclose(nameMap);
fclose(errorLog);
fprintf(stderr, "\n");
fprintf(stderr, "Finished with:\n");
fprintf(stderr, " "F_U32" warnings (bad base or qv)\n", nWARNS);
fprintf(stderr, "\n");
fprintf(stderr, "Read from inputs:\n");
fprintf(stderr, " "F_U64" bp.\n", bLOADED);
fprintf(stderr, " "F_U32" reads.\n", nLOADED);
fprintf(stderr, "\n");
fprintf(stderr, "Loaded into store:\n");
fprintf(stderr, " "F_U64" bp.\n", bLOADED);
fprintf(stderr, " "F_U32" reads.\n", nLOADED);
fprintf(stderr, "\n");
fprintf(stderr, "Skipped (too short):\n");
fprintf(stderr, " "F_U64" bp (%.4f%%).\n", bSKIPPED, 100.0 * bSKIPPED / (bSKIPPED + bLOADED));
fprintf(stderr, " "F_U32" reads (%.4f%%).\n", nSKIPPED, 100.0 * nSKIPPED / (nSKIPPED + nLOADED));
fprintf(stderr, "\n");
fprintf(stderr, "\n");
#if 0
fprintf(htmlLog, "\n");
fprintf(htmlLog, "<h2>Final Store</h2>\n");
fprintf(htmlLog, "<table>\n");
fprintf(htmlLog, "<tr><td colspan='2'>%s</td></tr>\n", gkpStoreName);
fprintf(htmlLog, "<tr><td>readsLoaded</td><td>"F_U32" reads ("F_U64" bp)</td></tr>\n", nLOADED, bLOADED);
fprintf(htmlLog, "<tr><td>readsSkipped</td><td>"F_U32" reads ("F_U64" bp) (read was too short)</td></tr>\n", nSKIPPED, bSKIPPED);
fprintf(htmlLog, "<tr><td>warnings</td><td>"F_U32" warnings (invalid base or quality value)</td></tr>\n", nWARNS);
fprintf(htmlLog, "</table>\n");
fprintf(htmlLog, "\n");
fprintf(htmlLog, "<script type='text/javascript'>\n");
fprintf(htmlLog, "var toggleOne = function() {\n");
fprintf(htmlLog, " var table = this.closest('table');\n");
fprintf(htmlLog, " var elts = table.querySelectorAll('.details');\n");
fprintf(htmlLog, "\n");
fprintf(htmlLog, " for (var i=0; i<elts.length; i++) {\n");
fprintf(htmlLog, " if (!elts[i].enabled) {\n");
fprintf(htmlLog, " elts[i].enabled = true;\n");
fprintf(htmlLog, " elts[i].style.visibility = 'visible';\n");
fprintf(htmlLog, " } else {\n");
fprintf(htmlLog, " elts[i].enabled = false;\n");
fprintf(htmlLog, " elts[i].style.visibility = 'collapse';\n");
fprintf(htmlLog, " }\n");
fprintf(htmlLog, " }\n");
fprintf(htmlLog, "}\n");
fprintf(htmlLog, "\n");
for (uint32 ii=0; ii<gkpFileID; ii++) {
fprintf(htmlLog, "document.getElementById('gkpload%u').onclick = toggleOne;\n", ii);
fprintf(htmlLog, "document.getElementById('gkpload%u').style = 'cursor: pointer;';\n", ii);
}
fprintf(htmlLog, "</script>\n");
fprintf(htmlLog, "\n");
fprintf(htmlLog, "</body>\n");
fprintf(htmlLog, "</html>\n");
#else
fprintf(htmlLog, "sum "F_U32" "F_U64" "F_U32" "F_U64" "F_U32"\n", nLOADED, bLOADED, nSKIPPED, bSKIPPED, nWARNS);
#endif
fclose(htmlLog);
if (nERROR > 0)
fprintf(stderr, "gatekeeperCreate did NOT finish successfully; too many errors.\n");
if (bSKIPPED > 0.25 * (bSKIPPED + bLOADED))
fprintf(stderr, "gatekeeperCreate did NOT finish successfully; too many bases skipped. Check your reads.\n");
if (nWARNS > 0.25 * (nLOADED))
fprintf(stderr, "gatekeeperCreate did NOT finish successfully; too many warnings. Check your reads.\n");
if (nSKIPPED > 0.50 * (nLOADED))
fprintf(stderr, "gatekeeperCreate did NOT finish successfully; too many short reads. Check your reads!\n");
if ((nERROR > 0) ||
(bSKIPPED > 0.25 * (bSKIPPED + bLOADED)) ||
(nWARNS > 0.25 * (nSKIPPED + nLOADED)) ||
(nSKIPPED > 0.50 * (nSKIPPED + nLOADED)))
exit(1);
fprintf(stderr, "gatekeeperCreate finished successfully.\n");
exit(0);
}
int main(int argc,char **argv) {
int count,j,s;
int norm;
bode::Interval *i;
std::string readsFN;
std::string inFN;
std::string outFN;
std::string inputFN;
int readsTotal,inputTotal;
double readsScale,inputScale;
int *target;
int bins,binwidth;
bode::Interval z;
bode::Flags *fset;
fset = processCmdLine(argc,argv);
norm = fset->getInt("norm");
bins = fset->getInt("bins");
binwidth = fset->getInt("binwidth");
inputFN = fset->getStr("input");
readsFN = fset->positionalArgs()[1];
inFN = fset->positionalArgs()[2];
outFN = fset->positionalArgs()[3];
// width = getBedWidth(inFN);
target = new int[bins];
bode::IntervalReader *reads = bode::IntervalReader::open(readsFN);
bode::CffReader *in = new bode::CffReader(inFN);
std::ofstream *out = new std::ofstream(outFN.c_str(),std::ofstream::out);
bode::IntervalReader *input = NULL;
if (inputFN != "") {
input = bode::IntervalReader::open(inputFN);
}
*out << "GID\tscore\tcoordinates";
s = 1;
while (s <= bins) {
*out << "\t" << s;
s++;
}
*out << std::endl;
fprintf(stderr,"loading reads...\n");
bode::IntervalSet *is = loadReads(reads,fset);
readsTotal = is->count();
fprintf(stderr,"loaded %d reads.\n",readsTotal);
readsScale = ((double) norm) / (double) readsTotal;
bode::IntervalSet *background = NULL;
if (input != NULL) {
fprintf(stderr,"loading input...\n");
background = loadReads(input,fset);
inputTotal = background->count();
fprintf(stderr,"loaded %d input reads.\n",inputTotal);
inputScale = ((double) norm) / (double) inputTotal;
}
reads->close();
count = 0;
fprintf(stderr,"processing intervals...\n");
while ((i=in->nextI())) {
bode::Bed *bi = static_cast<bode::Bed*>(i);
count++;
if (count % 1000 == 0) {
fprintf(stderr,"%9d\r",count);
}
for (j=0;j<bins;j++) {
target[j] = 0;
}
int left = bi->left() - (bins / 2) * binwidth;
int right = bi->left() + (bins / 2) * binwidth;
// is->density(i->chrom(),left,right,target);
(*out) <<bi->name() << "\t"<<bi->score()<<"\t"<<bi->chrom()<<":"<<bi->left();
s = left;
while (s < right) {
z.update(i->chrom(),s,s+binwidth);
int c = is->overlapping(&z);
int bk;
double ncbk;
if (background != NULL) {
bk = background->overlapping(&z);
ncbk = (c * readsScale) - (bk * inputScale);
} else {
ncbk = c * readsScale;
}
*out << "\t" << ncbk;
s += binwidth;
}
/* for (int j=0;j<right-left;j++) {
*out << "\t" << target[j];
}
*/
*out << std::endl;
}
in->close();
out->close();
if (input != NULL) {
input->close();
}
std::cerr << "processed " << count << " intervals." << std::endl;
return 0;
}
int main(int argc,char **argv) {
int count,j,s;
int norm;
bode::Interval *i;
bode::Bed *b;
std::string readsFN;
std::string inFN;
std::string outFN;
int readsTotal;
double readsScale;
int *target;
bode::Flags *fset;
fset = processCmdLine(argc,argv);
readsFN = fset->positionalArgs()[1];
inFN = fset->positionalArgs()[2];
outFN = fset->positionalArgs()[3];
norm = fset->getInt("norm");
target = new int[MAXINTERVAL];
bode::IntervalReader *reads = bode::IntervalReader::open(readsFN);
bode::BedReader *in = new bode::BedReader(inFN);
std::ofstream *out = new std::ofstream(outFN.c_str(),std::ofstream::out);
fprintf(stderr,"loading reads...\n");
bode::IntervalSet *is = loadReads(reads,fset);
readsTotal = is->count();
fprintf(stderr,"loaded %d reads.\n",readsTotal);
readsScale = ((double) norm) / (double) readsTotal;
reads->close();
count = 0;
fprintf(stderr,"processing intervals...\n");
while ((i=in->nextI())) {
count++;
if (count % 1000 == 0) {
fprintf(stderr,"%9d\r",count);
}
int width = i->right() - i->left();
for (j=0;j<width;j++) {
target[j] = 0;
}
is->density(i->chrom(),i->left(),i->right(),target);
if (i->strand() == '-') {
for (int k=0;k<width/2;k++) {
int x = target[k];
target[k] = target[width-k-1];
target[width-k-1] = x;
}
}
double nc = target[0] * readsScale;
b = static_cast<bode::Bed *>(i);
*out << b->name();
for (s=0;s<width;s++) {
nc = target[s] * readsScale;
*out << "\t" << nc;
}
*out << std::endl;
}
in->close();
out->close();
std::cerr << "processed " << count << " intervals." << std::endl;
return 0;
}
