void QuantitativeExperimentalDesign::mergeConsensusMaps_(ConsensusMap & out, const String & experiment, StringList & file_paths)
{
ConsensusMap map;
LOG_INFO << "Merge consensus maps: " << endl;
UInt counter = 1;
for (StringList::Iterator file_it = file_paths.begin(); file_it != file_paths.end(); ++file_it, ++counter)
{
//load should clear the map
ConsensusXMLFile().load(*file_it, map);
for (ConsensusMap::iterator it = map.begin(); it != map.end(); ++it)
{
it->setMetaValue("experiment", DataValue(experiment));
}
out += map;
}
LOG_INFO << endl;
}
ExitCodes main_(int, const char**)
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
//file list
StringList file_list = getStringList_("in");
//file type
FileHandler fh;
FileTypes::Type force_type;
if (getStringOption_("in_type").size() > 0)
{
force_type = FileTypes::nameToType(getStringOption_("in_type"));
}
else
{
force_type = fh.getType(file_list[0]);
}
//output file names and types
String out_file = getStringOption_("out");
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
bool annotate_file_origin = getFlag_("annotate_file_origin");
if (force_type == FileTypes::FEATUREXML)
{
FeatureMap<> out;
for (Size i = 0; i < file_list.size(); ++i)
{
FeatureMap<> map;
FeatureXMLFile fh;
fh.load(file_list[i], map);
if (annotate_file_origin)
{
for (FeatureMap<>::iterator it = map.begin(); it != map.end(); ++it)
{
it->setMetaValue("file_origin", DataValue(file_list[i]));
}
}
out += map;
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(out, getProcessingInfo_(DataProcessing::FORMAT_CONVERSION));
FeatureXMLFile f;
f.store(out_file, out);
}
else if (force_type == FileTypes::CONSENSUSXML)
{
ConsensusMap out;
ConsensusXMLFile fh;
fh.load(file_list[0], out);
//skip first file
for (Size i = 1; i < file_list.size(); ++i)
{
ConsensusMap map;
ConsensusXMLFile fh;
fh.load(file_list[i], map);
if (annotate_file_origin)
{
for (ConsensusMap::iterator it = map.begin(); it != map.end(); ++it)
{
it->setMetaValue("file_origin", DataValue(file_list[i]));
}
}
out += map;
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(out, getProcessingInfo_(DataProcessing::FORMAT_CONVERSION));
ConsensusXMLFile f;
f.store(out_file, out);
}
else if (force_type == FileTypes::TRAML)
{
TargetedExperiment out;
for (Size i = 0; i < file_list.size(); ++i)
{
TargetedExperiment map;
TraMLFile fh;
fh.load(file_list[i], map);
out += map;
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
//annotate output with data processing info
Software software;
software.setName("FileMerger");
software.setVersion(VersionInfo::getVersion());
out.addSoftware(software);
TraMLFile f;
f.store(out_file, out);
}
else
{
// we might want to combine different types, thus we only
// query in_type (which applies to all files)
// and not the suffix or content of a single file
force_type = FileTypes::nameToType(getStringOption_("in_type"));
//rt
bool rt_auto_number = getFlag_("raw:rt_auto");
bool rt_filename = getFlag_("raw:rt_filename");
bool rt_custom = false;
DoubleList custom_rts = getDoubleList_("raw:rt_custom");
if (custom_rts.size() != 0)
{
rt_custom = true;
if (custom_rts.size() != file_list.size())
{
writeLog_("Custom retention time list must have as many elements as there are input files!");
printUsage_();
return ILLEGAL_PARAMETERS;
}
}
//ms level
bool user_ms_level = getFlag_("raw:user_ms_level");
MSExperiment<> out;
out.reserve(file_list.size());
UInt rt_auto = 0;
UInt native_id = 0;
std::vector<MSChromatogram<ChromatogramPeak> > all_chromatograms;
for (Size i = 0; i < file_list.size(); ++i)
{
String filename = file_list[i];
//load file
MSExperiment<> in;
fh.loadExperiment(filename, in, force_type, log_type_);
if (in.empty() && in.getChromatograms().empty())
{
writeLog_(String("Warning: Empty file '") + filename + "'!");
continue;
}
out.reserve(out.size() + in.size());
//warn if custom RT and more than one scan in input file
if (rt_custom && in.size() > 1)
{
writeLog_(String("Warning: More than one scan in file '") + filename + "'! All scans will have the same retention time!");
}
for (MSExperiment<>::const_iterator it2 = in.begin(); it2 != in.end(); ++it2)
{
//handle rt
Real rt_final = it2->getRT();
if (rt_auto_number)
{
rt_final = ++rt_auto;
}
else if (rt_custom)
{
rt_final = custom_rts[i];
}
else if (rt_filename)
{
if (!filename.hasSubstring("rt"))
{
writeLog_(String("Warning: cannot guess retention time from filename as it does not contain 'rt'"));
}
for (Size i = 0; i < filename.size(); ++i)
{
if (filename[i] == 'r' && ++i != filename.size() && filename[i] == 't' && ++i != filename.size() && isdigit(filename[i]))
{
String rt;
while (i != filename.size() && (filename[i] == '.' || isdigit(filename[i])))
{
rt += filename[i++];
}
if (rt.size() > 0)
{
// remove dot from rt3892.98.dta
// ^
if (rt[rt.size() - 1] == '.')
{
// remove last character
rt.erase(rt.end() - 1);
}
}
try
{
float tmp = rt.toFloat();
rt_final = tmp;
}
catch (Exception::ConversionError)
{
writeLog_(String("Warning: cannot convert the found retention time in a value '" + rt + "'."));
}
}
}
}
// none of the rt methods were successful
if (rt_final == -1)
{
writeLog_(String("Warning: No valid retention time for output scan '") + rt_auto + "' from file '" + filename + "'");
}
out.addSpectrum(*it2);
out.getSpectra().back().setRT(rt_final);
out.getSpectra().back().setNativeID(native_id);
if (user_ms_level)
{
out.getSpectra().back().setMSLevel((int)getIntOption_("raw:ms_level"));
}
++native_id;
}
// if we had only one spectrum, we can annotate it directly, for more spectra, we just name the source file leaving the spectra unannotated (to avoid a long and redundant list of sourceFiles)
if (in.size() == 1)
{
out.getSpectra().back().setSourceFile(in.getSourceFiles()[0]);
in.getSourceFiles().clear(); // delete source file annotated from source file (its in the spectrum anyways)
}
// copy experimental settings from first file
if (i == 0)
{
out.ExperimentalSettings::operator=(in);
}
else // otherwise append
{
out.getSourceFiles().insert(out.getSourceFiles().end(), in.getSourceFiles().begin(), in.getSourceFiles().end()); // could be emtpty if spectrum was annotated above, but that's ok then
}
// also add the chromatograms
for (std::vector<MSChromatogram<ChromatogramPeak> >::const_iterator it2 = in.getChromatograms().begin(); it2 != in.getChromatograms().end(); ++it2)
{
all_chromatograms.push_back(*it2);
}
}
// set the chromatograms
out.setChromatograms(all_chromatograms);
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(out, getProcessingInfo_(DataProcessing::FORMAT_CONVERSION));
MzMLFile f;
f.setLogType(log_type_);
f.store(out_file, out);
}
return EXECUTION_OK;
}
ExitCodes main_(int, const char**)
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
// file list
StringList file_list = getStringList_("in");
// file type
FileHandler file_handler;
FileTypes::Type force_type;
if (getStringOption_("in_type").size() > 0)
{
force_type = FileTypes::nameToType(getStringOption_("in_type"));
}
else
{
force_type = file_handler.getType(file_list[0]);
}
// output file names and types
String out_file = getStringOption_("out");
bool annotate_file_origin = getFlag_("annotate_file_origin");
rt_gap_ = getDoubleOption_("rt_concat:gap");
vector<String> trafo_out = getStringList_("rt_concat:trafo_out");
if (trafo_out.empty())
{
// resize now so we don't have to worry about indexing out of bounds:
trafo_out.resize(file_list.size());
}
else if (trafo_out.size() != file_list.size())
{
writeLog_("Error: Number of transformation output files must equal the number of input files (parameters 'rt_concat:trafo_out'/'in')!");
return ILLEGAL_PARAMETERS;
}
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
if (force_type == FileTypes::FEATUREXML)
{
FeatureMap out;
FeatureXMLFile fh;
for (Size i = 0; i < file_list.size(); ++i)
{
FeatureMap map;
fh.load(file_list[i], map);
if (annotate_file_origin)
{
for (FeatureMap::iterator it = map.begin(); it != map.end(); ++it)
{
it->setMetaValue("file_origin", DataValue(file_list[i]));
}
}
if (rt_gap_ > 0.0) // concatenate in RT
{
adjustRetentionTimes_(map, trafo_out[i], i == 0);
}
out += map;
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
// annotate output with data processing info
addDataProcessing_(out, getProcessingInfo_(DataProcessing::FORMAT_CONVERSION));
fh.store(out_file, out);
}
else if (force_type == FileTypes::CONSENSUSXML)
{
ConsensusMap out;
ConsensusXMLFile fh;
fh.load(file_list[0], out);
// skip first file
for (Size i = 1; i < file_list.size(); ++i)
{
ConsensusMap map;
fh.load(file_list[i], map);
if (annotate_file_origin)
{
for (ConsensusMap::iterator it = map.begin(); it != map.end(); ++it)
{
it->setMetaValue("file_origin", DataValue(file_list[i]));
}
}
if (rt_gap_ > 0.0) // concatenate in RT
{
adjustRetentionTimes_(map, trafo_out[i], i == 0);
}
out += map;
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
// annotate output with data processing info
addDataProcessing_(out, getProcessingInfo_(DataProcessing::FORMAT_CONVERSION));
fh.store(out_file, out);
}
else if (force_type == FileTypes::TRAML)
{
TargetedExperiment out;
TraMLFile fh;
for (Size i = 0; i < file_list.size(); ++i)
{
TargetedExperiment map;
fh.load(file_list[i], map);
out += map;
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
// annotate output with data processing info
Software software;
software.setName("FileMerger");
software.setVersion(VersionInfo::getVersion());
out.addSoftware(software);
fh.store(out_file, out);
}
else // raw data input (e.g. mzML)
{
// RT
bool rt_auto_number = getFlag_("raw:rt_auto");
bool rt_filename = getFlag_("raw:rt_filename");
bool rt_custom = false;
DoubleList custom_rts = getDoubleList_("raw:rt_custom");
if (!custom_rts.empty())
{
rt_custom = true;
if (custom_rts.size() != file_list.size())
{
writeLog_("Custom retention time list (parameter 'raw:rt_custom') must have as many elements as there are input files (parameter 'in')!");
return ILLEGAL_PARAMETERS;
}
}
// MS level
Int ms_level = getIntOption_("raw:ms_level");
MSExperiment<> out;
UInt rt_auto = 0;
UInt native_id = 0;
for (Size i = 0; i < file_list.size(); ++i)
{
String filename = file_list[i];
// load file
force_type = file_handler.getType(file_list[i]);
MSExperiment<> in;
file_handler.loadExperiment(filename, in, force_type, log_type_);
if (in.empty() && in.getChromatograms().empty())
{
writeLog_(String("Warning: Empty file '") + filename + "'!");
continue;
}
out.reserve(out.size() + in.size());
// warn if custom RT and more than one scan in input file
if (rt_custom && in.size() > 1)
{
writeLog_(String("Warning: More than one scan in file '") + filename + "'! All scans will have the same retention time!");
}
// handle special raw data options:
for (MSExperiment<>::iterator spec_it = in.begin();
spec_it != in.end(); ++spec_it)
{
float rt_final = spec_it->getRT();
if (rt_auto_number)
{
rt_final = ++rt_auto;
}
else if (rt_custom)
{
rt_final = custom_rts[i];
}
else if (rt_filename)
{
static const boost::regex re("rt(\\d+(\\.\\d+)?)");
boost::smatch match;
bool found = boost::regex_search(filename, match, re);
if (found)
{
rt_final = String(match[1]).toFloat();
}
else
{
writeLog_("Warning: could not extract retention time from filename '" + filename + "'");
}
}
// none of the rt methods were successful
if (rt_final < 0)
{
writeLog_(String("Warning: No valid retention time for output scan '") + rt_auto + "' from file '" + filename + "'");
}
spec_it->setRT(rt_final);
spec_it->setNativeID("spectrum=" + String(native_id));
if (ms_level > 0)
{
spec_it->setMSLevel(ms_level);
}
++native_id;
}
// if we have only one spectrum, we can annotate it directly, for more spectra, we just name the source file leaving the spectra unannotated (to avoid a long and redundant list of sourceFiles)
if (in.size() == 1)
{
in[0].setSourceFile(in.getSourceFiles()[0]);
in.getSourceFiles().clear(); // delete source file annotated from source file (it's in the spectrum anyways)
}
if (rt_gap_ > 0.0) // concatenate in RT
{
adjustRetentionTimes_(in, trafo_out[i], i == 0);
}
// add spectra to output
for (MSExperiment<>::const_iterator spec_it = in.begin();
spec_it != in.end(); ++spec_it)
{
out.addSpectrum(*spec_it);
}
// also add the chromatograms
for (vector<MSChromatogram<ChromatogramPeak> >::const_iterator
chrom_it = in.getChromatograms().begin(); chrom_it !=
in.getChromatograms().end(); ++chrom_it)
{
out.addChromatogram(*chrom_it);
}
// copy experimental settings from first file
if (i == 0)
{
out.ExperimentalSettings::operator=(in);
}
else // otherwise append
{
out.getSourceFiles().insert(out.getSourceFiles().end(), in.getSourceFiles().begin(), in.getSourceFiles().end()); // could be emtpty if spectrum was annotated above, but that's ok then
}
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
// annotate output with data processing info
addDataProcessing_(out, getProcessingInfo_(DataProcessing::FORMAT_CONVERSION));
MzMLFile f;
f.setLogType(log_type_);
f.store(out_file, out);
}
return EXECUTION_OK;
}