void TBDataParser::OnCharacters(const char *characters)
{
if(_currentElement != NULL && !strcmp(_currentElement->Data(),"vector")) {
TString *string = new TString(characters);
TObjArray *values = string->Tokenize(", ");
for(Int_t i = 0; i < values->GetEntries(); i++) {
TObjString *object = (TObjString *) values->At(i);
TString value = object->GetString().ReplaceAll("\n", "").ReplaceAll("\t", "").ReplaceAll(" ", "").ReplaceAll("\0", "");
if(value.IsFloat()) {
_vector->Fill(value.Atof());
}
}
if(_vectorsStack->GetEntries() == 1) {
_motherVecEntries = _vector->GetEntries();
} else if(_currentMethod != NULL && !strcmp(_currentMethod->Data(),"all")) {
for(Int_t i = 1; i < _motherVecEntries; i++) {
TObjString *object = (TObjString *) values->First();
TString value = object->GetString().ReplaceAll("\n", "").ReplaceAll("\t", "").ReplaceAll(" ", "").ReplaceAll("\0", "");
if(value.IsFloat())
_vector->Fill(value.Atof());
}
}
values->Delete();
}
}
Double_t CalcLednicR(Double_t zr, Double_t zt, Double_t v)
{
TString cmd = Form("echo '%0.5f,%0.5f,%0.5f'", zr, zt, v);
cmd+= " | ./lednic_input.exe";
Printf("%s", cmd.Data());
cmd = gSystem->GetFromPipe(cmd.Data());
if (!cmd.IsFloat()) {
Printf("Problem with program 'lednic_input.exe'");
return 0;
}
return cmd.Atof();
}
//void main_makePlots()
int main(int argc, char* argv[])
{
// Output filenames options - for creation of a log file and a histogram file in the given directory
TString dir = "control_plots/";
TString tst = fileTimeStamp();
TString pfx = "_plots_";
TString sfx = "_NO_MET_CUT";
// Process input
if(argc<3)
{
cout << "\n ERROR: Please enter a decay chain, dataset, and (optional) a max number of events.\n" << endl;
return 1;
}
TString decayChain = argv[1];
TString dataset = argv[2];
TString maxEventsStr = "" ;
if(argc>=4)
{
maxEventsStr = argv[3];
if(!maxEventsStr.IsFloat())
{
cout << "\n ERROR: Please enter the number of events you want to process for your third option"
"\n (or nothing if you want to process them all).\n"
<< endl;
return 1;
}
}
if(decayChain!="Zuu" && decayChain!="Zee" && decayChain!="Zttuu" && decayChain!="Zttee")
{
cout << "\n ERROR: Please specify one of the following decay chains for your first option: Zuu, Zee, Zttuu, Zttee\n" << endl;
return 1;
}
// Open files and map trees
TFile* f_input;
if (dataset == "muon" ) f_input = TFile::Open(fn_muon );
else if(dataset == "elec" ) f_input = TFile::Open(fn_elec );
else if(dataset == "dy" ) f_input = TFile::Open(fn_dy );
else if(dataset == "dy1j" ) f_input = TFile::Open(fn_dy1j );
else if(dataset == "ww" ) f_input = TFile::Open(fn_ww );
else if(dataset == "wz" ) f_input = TFile::Open(fn_wz );
else if(dataset == "zz" ) f_input = TFile::Open(fn_zz );
else if(dataset == "tthad" ) f_input = TFile::Open(fn_tthad );
else if(dataset == "ttlep" ) f_input = TFile::Open(fn_ttlep );
else if(dataset == "ttsemi" ) f_input = TFile::Open(fn_ttsemi );
else if(dataset == "t_s" ) f_input = TFile::Open(fn_t_s );
else if(dataset == "t_t" ) f_input = TFile::Open(fn_t_t );
else if(dataset == "t_tw" ) f_input = TFile::Open(fn_t_tw );
else if(dataset == "tbar_s" ) f_input = TFile::Open(fn_tbar_s );
else if(dataset == "tbar_t" ) f_input = TFile::Open(fn_tbar_t );
else if(dataset == "tbar_tw" ) f_input = TFile::Open(fn_tbar_tw);
else
{
cout << "\n ERROR: Please specify one of the following datasets for your second option:"
"\n muon, elec, dy, dy1j, ww, wz, zz, tthad, ttlep, ttsemi, t(bar)_(s/t/tw)\n" << endl;
return 1;
}
// If tau tau decays specified on a non-dy set.
if( (decayChain=="Zttuu" || decayChain=="Zttee") && (dataset!="dy" && dataset!="dy1j"))
{
cout << "\n ERROR: Tau decay split only valid for DY datasets\n" << endl;
return 1;
}
TTree* t_input = (TTree*)f_input->Get("tree");
// Create an instance of the ZCControlPlotMaker and input options
ZCControlPlotMaker *plotMaker = new ZCControlPlotMaker();
TString outputFileName = dir+tst+pfx+decayChain+"_"+dataset+sfx;
TString options = decayChain;
if(dataset!="muon" && dataset!="elec") options+=",sim";
if(dataset=="dy") options+=",dy";
// Process tree and save output
if(maxEventsStr!="")
t_input->Process(plotMaker, options, maxEventsStr.Atoi() );
else t_input->Process(plotMaker, options );
cout << plotMaker->saveOutputToFile(outputFileName) << endl;
return 0;
}
int main(int argc,char *argv[])
{
string usage = "\nCNVnator ";
#ifdef CNVNATOR_VERSION
usage += CNVNATOR_VERSION;
#else
usage += "v???";
#endif
usage += "\n\nUsage:\n";
usage += argv[0];
usage += " -root out.root [-genome name] [-chrom 1 2 ...] -tree file1.bam ...\n";
usage += argv[0];
usage += " -root out.root [-genome name] [-chrom 1 2 ...] -merge file1.root ...\n";
usage += argv[0];
usage += " -root file.root [-genome name] [-chrom 1 2 ...] [-d dir] -his bin_size\n";
usage += argv[0];
usage += " -root file.root [-chrom 1 2 ...] -stat bin_size\n";
usage += argv[0];
usage += " -root file.root -eval bin_size\n";
usage += argv[0];
usage += " -root file.root [-chrom 1 2 ...] -partition bin_size [-ngc]\n";
// usage += argv[0];
//usage += " -root file.root [-chrom 1 2 ...] -spartition bin_size [-gc]\n";
usage += argv[0];
usage += " -root file.root [-chrom 1 2 ...] -call bin_size [-ngc]\n";
usage += argv[0];
usage += " -root file.root -genotype bin_size [-ngc]\n";
usage += argv[0];
usage += " -root file.root -view bin_size [-ngc]\n";
usage += argv[0];
usage += " -pe file1.bam ... -qual val(20) -over val(0.8) [-f file]\n";
usage += "\n";
usage += "Valid genomes (-genome option) are: NCBI36, hg18, GRCh37, hg19\n";
if (argc < 2) {
cerr<<"Not enough parameters."<<endl;
cerr<<usage<<endl;
return 0;
}
#ifdef USE_YEPPP
YepStatus yepStatus = yepLibrary_Init();
if (yepStatus != YepStatusOk) {
cerr<<"Yeppp library initialization failed with status "<<yepStatus<<"."<<endl;
return 1;
}
#endif
static const int OPT_TREE = 0x00001;
static const int OPT_MERGE = 0x00002;
static const int OPT_HIS = 0x00004;
static const int OPT_HISMERGE = 0x00008;
static const int OPT_STAT = 0x00010;
static const int OPT_PARTITION = 0x00020;
static const int OPT_EPARTITION = 0x00040;
static const int OPT_CALL = 0x00080;
static const int OPT_VIEW = 0x00100;
static const int OPT_GENOTYPE = 0x00200;
static const int OPT_EVAL = 0x00400;
static const int OPT_PE = 0x00800;
static const int OPT_PANEL = 0x01000;
static const int OPT_FIT = 0x02000;
static const int OPT_SPARTITION = 0x04000;
static const int OPT_HIS_NEW = 0x08000;
static const int OPT_AGGREGATE = 0x10000;
// tree, merge, his, stat, partition, spartition, call, view, genotype
int max_opts = 10000, n_opts = 0, opts[max_opts], bins[max_opts], gbin = 0;
for (int i = 0;i < n_opts;i++) bins[i] = 0;
bool useGCcorr = true,useATcorr = false;
bool forUnique = false,relaxCalling = false;
string out_root_file(""),call_file(""),group_name("");
string chroms[1000],data_files[100000],root_files[100000] = {""},dir = ".";
int n_chroms = 0,n_files = 0,n_root_files = 0,range = 128, qual = 20;
double over = 0.8;
Genome *genome = NULL;
int index = 1;
while (index < argc) {
string option = argv[index++];
if (option == "-tree" || option == "-merge" || option == "-pe") {
if (option == "-tree") opts[n_opts++] = OPT_TREE;
if (option == "-merge") opts[n_opts++] = OPT_MERGE;
if (option == "-pe") opts[n_opts++] = OPT_PE;
while (index < argc && argv[index][0] != '-')
if (strlen(argv[index++]) > 0) data_files[n_files++] = argv[index - 1];
} else if (option == "-his" || option == "-his_new" ||
option == "-hismerge" ||
option == "-stat" || option == "-eval" ||
option == "-partition" || option == "-spartition" ||
option == "-epartition" ||
option == "-call" || option == "-view" ||
option == "-genotype" || option == "-aggregate") {
int bs = 0;
if (index < argc && argv[index][0] != '-') {
TString tmp = argv[index++];
if (!tmp.IsDigit()) {
cerr<<"Bin size must be integer for option '"<<option<<"'."<<endl;
cerr<<usage<<endl;
return 0;
}
bs = tmp.Atoi();
}
if (option == "-his") opts[n_opts] = OPT_HIS;
if (option == "-hismerge") opts[n_opts] = OPT_HISMERGE;
if (option == "-stat") opts[n_opts] = OPT_STAT;
if (option == "-partition") opts[n_opts] = OPT_PARTITION;
if (option == "-epartition") opts[n_opts] = OPT_EPARTITION;
if (option == "-spartition") opts[n_opts] = OPT_SPARTITION;
if (option == "-call") opts[n_opts] = OPT_CALL;
if (option == "-view") opts[n_opts] = OPT_VIEW;
if (option == "-genotype") opts[n_opts] = OPT_GENOTYPE;
if (option == "-his_new") opts[n_opts] = OPT_HIS_NEW;
if (option == "-eval") opts[n_opts] = OPT_EVAL;
if (option == "-aggregate") opts[n_opts] = OPT_AGGREGATE;
bins[n_opts++] = bs;
} else if (option == "-panel") {
opts[n_opts++] = OPT_PANEL;
} else if (option == "-fit") {
opts[n_opts++] = OPT_FIT;
if (index < argc && argv[index][0] != '-') group_name = argv[index++];
else {
cout<<"Please provide name of a sample group."<<endl;
return 0;
}
} else if (option == "-outroot") {
if (index < argc && argv[index][0] != '-') out_root_file = argv[index++];
else {
cout<<"Please provide new root-file name."<<endl;
return 0;
}
} else if (option == "-root") {
while (index < argc && argv[index][0] != '-')
if (strlen(argv[index++]) > 0)
root_files[n_root_files++] = argv[index - 1];
if (n_root_files == 0) {
cerr<<"Please provide root-file name."<<endl;
cerr<<usage<<endl;
return 0;
}
} else if (option == "-outroot") {
if (index < argc && argv[index][0] != '-') out_root_file = argv[index++];
else {
cout<<"Please provide new root-file name."<<endl;
return 0;
}
} else if (option == "-chrom") {
while (index < argc && argv[index][0] != '-')
chroms[n_chroms++] = argv[index++];
if (n_chroms == 0) {
cerr<<"Provide chromosome names."<<endl;
cerr<<usage<<endl;
return 0;
}
} else if (option == "-ngc") {
useGCcorr = false;
} else if (option == "-at") {
useATcorr = true;
} else if (option == "-genome") {
if (index < argc) genome = Genome::get(argv[index++]);
} else if (option == "-d") {
if (index < argc && argv[index][0] != '-')
dir = argv[index++];
else cerr<<"No directory is given."<<endl;
} else if (option == "-qual") {
if (index >= argc || argv[index][0] == '-') {
cerr<<"No quality value is provided."<<endl;
cerr<<usage<<endl;
return 0;
}
TString tmp = argv[index++];
if (!tmp.IsDigit()) {
cerr<<"Quality value must be integer."<<endl;
cerr<<usage<<endl;
return 0;
}
qual = tmp.Atoi();
} else if (option == "-over") {
if (index >= argc || argv[index][0] == '-') {
cerr<<"No fraction of overlap is provided."<<endl;
cerr<<usage<<endl;
return 0;
}
TString tmp = argv[index++];
if (!tmp.IsFloat()) {
cerr<<"Fraction of overlap must be number."<<endl;
cerr<<usage<<endl;
return 0;
}
over = tmp.Atof();
} else if (option == "-f") {
if (index >= argc || argv[index][0] == '-') {
cerr<<"No file name is provided."<<endl;
cerr<<usage<<endl;
return 0;
}
call_file = argv[index++];
} else if (option == "-unique") {
forUnique = true;
} else if (option == "-range") {
range = atoi(argv[index++]);
} else if (option == "-relax") {
relaxCalling = true;
} else if (option[0] == '-') {
cerr<<"Unknown option '"<<option<<"'.\n"<<endl;
}
}
if (out_root_file.length() <= 0) out_root_file = root_files[0];
if (out_root_file.length() <= 0)
cerr<<"WARNING: no name of root-file provided."<<endl;
for (int o = 0;o < n_opts;o++) {
int option = opts[o];
int bin = bins[o]; if (bin <= 0) bin = gbin;
if (option == OPT_TREE) { // tree
HisMaker maker(out_root_file,genome);
maker.setDataDir(dir);
maker.produceTrees(chroms,n_chroms,data_files,n_files,forUnique);
}
if (option == OPT_MERGE) { // merge
HisMaker maker(out_root_file,genome);
maker.mergeTrees(chroms,n_chroms,data_files,n_files);
}
if (option == OPT_HIS ||
option == OPT_HISMERGE) { // his
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.setDataDir(dir);
maker.produceHistograms(chroms,n_chroms,root_files,n_root_files,false);
if (option == OPT_HISMERGE)
maker.produceHistograms(chroms,n_chroms,root_files,n_root_files,true);
}
if (option == OPT_STAT) { // stat
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.stat(chroms,n_chroms,useATcorr);
}
if (option == OPT_PARTITION) { // partition
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.partition(chroms,n_chroms,false,useATcorr,useGCcorr,false,range);
}
if (option == OPT_EPARTITION) { // exome partition
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.partition(chroms,n_chroms,false,useATcorr,useGCcorr,true,range);
}
if (option == OPT_CALL) { // call
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.callSVs(chroms,n_chroms,useATcorr,useGCcorr,relaxCalling);
}
if (option == OPT_VIEW) { // view
HisMaker maker(out_root_file,bin,useGCcorr,genome);
TApplication theApp("App",0,0);
maker.view(root_files,n_root_files,useATcorr,useGCcorr);
theApp.Run();
}
if (option == OPT_GENOTYPE) { // genotype
HisMaker maker(out_root_file,bin,useGCcorr,genome);
TApplication theApp("App",0,0);
maker.genotype(root_files,n_root_files,useATcorr,useGCcorr);
theApp.Run();
}
if (option == OPT_EVAL) { // eval
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.eval(root_files,n_root_files,useATcorr,useGCcorr);
}
if (option == OPT_PE) { // pe
HisMaker maker("null",genome);
if (call_file.length() > 0)
maker.pe_for_file(call_file,data_files,n_files,over,qual);
else {
TApplication theApp("App",0,0);
maker.pe(data_files,n_files,over,qual);
theApp.Run();
}
}
if (option == OPT_SPARTITION) { // spartition
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.partition(chroms,n_chroms,true,useATcorr,useGCcorr,false,range);
}
if (option == OPT_HIS_NEW) { // his_new
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.setDataDir(dir);
maker.produceHistogramsNew(chroms,n_chroms);
}
// EXOnator options
if (option == OPT_PANEL) { // panel
EXOnator exonator(out_root_file);
exonator.makeTables();
}
if (option == OPT_FIT) { // fit
EXOnator exonator(out_root_file);
exonator.fit(group_name);
}
if (option == OPT_AGGREGATE) { // aggregate
HisMaker maker(out_root_file,bin,useGCcorr,genome);
maker.setDataDir(dir);
maker.aggregate(root_files,n_root_files,chroms,n_chroms);
}
}
return 0;
}
void DrawMLPoutputMovie( TFile* file, const TString& methodType, const TString& methodTitle )
{
gROOT->SetBatch( 1 );
// define Canvas layout here!
const Int_t width = 600; // size of canvas
// this defines how many canvases we need
TCanvas* c = 0;
Float_t nrms = 4;
Float_t xmin = -1.2;
Float_t xmax = 1.2;
Float_t ymin = 0;
Float_t ymax = 0;
Float_t maxMult = 6.0;
Int_t countCanvas = 0;
Bool_t first = kTRUE;
TString dirname = methodType + "/" + methodTitle + "/" + "EpochMonitoring";
TDirectory *epochDir = (TDirectory*)file->Get( dirname );
if (!epochDir) {
cout << "Big troubles: could not find directory \"" << dirname << "\"" << endl;
exit(1);
}
// now read all evolution histograms
TIter keyItTit(epochDir->GetListOfKeys());
TKey *titkeyTit;
while ((titkeyTit = (TKey*)keyItTit())) {
if (!gROOT->GetClass(titkeyTit->GetClassName())->InheritsFrom("TH1F")) continue;
TString name = titkeyTit->GetName();
if (!name.BeginsWith("convergencetest___")) continue;
if (!name.Contains("_train_")) continue; // only for training so far
if (name.EndsWith( "_B")) continue;
// must be signal histogram
if (!name.EndsWith( "_S")) {
cout << "Big troubles with histogram: " << name << " -> should end with _S" << endl;
exit(1);
}
// create canvas
countCanvas++;
TString ctitle = Form("TMVA response %s",methodTitle.Data());
c = new TCanvas( Form("canvas%d", countCanvas), ctitle, 0, 0, width, (Int_t)width*0.78 );
TH1F* sig = (TH1F*)titkeyTit->ReadObj();
sig->SetTitle( Form("TMVA response for classifier: %s", methodTitle.Data()) );
TString dataType = (name.Contains("_train_") ? "(training sample)" : "(test sample)");
// find background
TString nbn = sig->GetName(); nbn[nbn.Length()-1] = 'B';
TH1F* bgd = dynamic_cast<TH1F*>(epochDir->Get( nbn ));
if (bgd == 0) {
cout << "Big troubles with histogram: " << bgd << " -> cannot find!" << endl;
exit(1);
}
cout << "sig = " << sig->GetName() << endl;
cout << "bgd = " << bgd->GetName() << endl;
// set the histogram style
TMVAGlob::SetSignalAndBackgroundStyle( sig, bgd );
// normalise both signal and background
TMVAGlob::NormalizeHists( sig, bgd );
// set only first time, then same for all plots
if (first) {
if (xmin == 0 && xmax == 0) {
xmin = TMath::Max( TMath::Min(sig->GetMean() - nrms*sig->GetRMS(),
bgd->GetMean() - nrms*bgd->GetRMS() ),
sig->GetXaxis()->GetXmin() );
xmax = TMath::Min( TMath::Max(sig->GetMean() + nrms*sig->GetRMS(),
bgd->GetMean() + nrms*bgd->GetRMS() ),
sig->GetXaxis()->GetXmax() );
}
ymin = 0;
ymax = TMath::Max( sig->GetMaximum(), bgd->GetMaximum() )*maxMult;
first = kFALSE;
}
// build a frame
Int_t nb = 100;
TString hFrameName(TString("frame") + methodTitle);
TObject *o = gROOT->FindObject(hFrameName);
if(o) delete o;
TH2F* frame = new TH2F( hFrameName, sig->GetTitle(),
nb, xmin, xmax, nb, ymin, ymax );
frame->GetXaxis()->SetTitle( methodTitle + " response" );
frame->GetYaxis()->SetTitle("(1/N) dN^{ }/^{ }dx");
TMVAGlob::SetFrameStyle( frame );
// find epoch number (4th token)
TObjArray* tokens = name.Tokenize("_");
TString es = ((TObjString*)tokens->At(4))->GetString();
if (!es.IsFloat()) {
cout << "Big troubles in epoch parsing: \"" << es << "\" is not float" << endl;
exit(1);
}
Int_t epoch = es.Atoi();
// eventually: draw the frame
frame->Draw();
c->GetPad(0)->SetLeftMargin( 0.105 );
frame->GetYaxis()->SetTitleOffset( 1.2 );
// Draw legend
TLegend *legend= new TLegend( c->GetLeftMargin(), 1 - c->GetTopMargin() - 0.12,
c->GetLeftMargin() + 0.5, 1 - c->GetTopMargin() );
legend->SetFillStyle( 1 );
legend->AddEntry(sig,TString("Signal ") + dataType, "F");
legend->AddEntry(bgd,TString("Background ") + dataType, "F");
legend->SetBorderSize(1);
legend->SetMargin( 0.15 );
legend->Draw("same");
TText* t = new TText();
t->SetTextSize( 0.04 );
t->SetTextColor( 1 );
t->SetTextAlign( 31 );
t->DrawTextNDC( 1 - c->GetRightMargin(), 1 - c->GetTopMargin() + 0.015, Form( "Epoch: %i", epoch) );
// overlay signal and background histograms
sig->Draw("samehist");
bgd->Draw("samehist");
// save to file
TString dirname = "movieplots";
TString foutname = dirname + "/" + name;
foutname.Resize( foutname.Length()-2 );
foutname.ReplaceAll("convergencetest___","");
foutname += ".gif";
cout << "storing file: " << foutname << endl;
c->Update();
c->Print(foutname);
}
}
