예제 #1
0
  ExitCodes main_(int, const char**)
  {
    //-------------------------------------------------------------
    // parameter handling
    //-------------------------------------------------------------
    StringList in = getStringList_("in");
    String edta = getStringOption_("pos");
    String out = getStringOption_("out");
    String out_sep = getStringOption_("out_separator");
    String out_TIC_debug = getStringOption_("auto_rt:out_debug_TIC");

    StringList in_header = getStringList_("in_header");


    // number of out_debug_TIC files and input files must be identical
    /*if (out_TIC_debug.size() > 0 && in.size() != out_TIC_debug.size())
    {
        LOG_FATAL_ERROR << "Error: number of input file 'in' and auto_rt:out_debug_TIC files must be identical!" << std::endl;
        return ILLEGAL_PARAMETERS;
    }*/

    // number of header files and input files must be identical
    if (in_header.size() > 0 && in.size() != in_header.size())
    {
      LOG_FATAL_ERROR << "Error: number of input file 'in' and 'in_header' files must be identical!" << std::endl;
      return ILLEGAL_PARAMETERS;
    }

    if (!getFlag_("auto_rt:enabled") && !out_TIC_debug.empty())
    {
      LOG_FATAL_ERROR << "Error: TIC output file requested, but auto_rt is not enabled! Either do not request the file or switch on 'auto_rt:enabled'." << std::endl;
      return ILLEGAL_PARAMETERS;
    }

    double rttol = getDoubleOption_("rt_tol");
    double mztol = getDoubleOption_("mz_tol");
    Size rt_collect = getIntOption_("rt_collect");

    //-------------------------------------------------------------
    // loading input
    //-------------------------------------------------------------
    MzMLFile mzml_file;
    mzml_file.setLogType(log_type_);
    MSExperiment<Peak1D> exp, exp_pp;

    EDTAFile ed;
    ConsensusMap cm;
    ed.load(edta, cm);

    StringList tf_single_header0, tf_single_header1, tf_single_header2; // header content, for each column

    std::vector<String> vec_single; // one line for each compound, multiple columns per experiment
    vec_single.resize(cm.size());
    for (Size fi = 0; fi < in.size(); ++fi)
    {
      // load raw data
      mzml_file.load(in[fi], exp);
      exp.sortSpectra(true);

      if (exp.empty())
      {
        LOG_WARN << "The given file does not contain any conventional peak data, but might"
                    " contain chromatograms. This tool currently cannot handle them, sorry." << std::endl;
        return INCOMPATIBLE_INPUT_DATA;
      }

      // try to detect RT peaks (only for the first input file -- all others should align!)
      // cm.size() might change in here...
      if (getFlag_("auto_rt:enabled") && fi == 0)
      {
        ConsensusMap cm_local = cm; // we might have different RT peaks for each map if 'auto_rt' is enabled
        cm.clear(false); // reset global list (about to be filled)

        // compute TIC
        MSChromatogram<> tic = exp.getTIC();
        MSSpectrum<> tics, tic_gf, tics_pp, tics_sn;
        for (Size ic = 0; ic < tic.size(); ++ic)
        { // rewrite Chromatogram to MSSpectrum (GaussFilter requires it)
          Peak1D peak;
          peak.setMZ(tic[ic].getRT());
          peak.setIntensity(tic[ic].getIntensity());
          tics.push_back(peak);
        }
        // smooth (no PP_CWT here due to efficiency reasons -- large FWHM take longer!)
        double fwhm = getDoubleOption_("auto_rt:FHWM");
        GaussFilter gf;
        Param p = gf.getParameters();
        p.setValue("gaussian_width", fwhm * 2); // wider than FWHM, just to be sure we have a fully smoothed peak. Merging two peaks is unlikely
        p.setValue("use_ppm_tolerance", "false");
        gf.setParameters(p);
        tic_gf = tics;
        gf.filter(tic_gf);
        // pick peaks
        PeakPickerHiRes pp;
        p = pp.getParameters();
        p.setValue("signal_to_noise", getDoubleOption_("auto_rt:SNThreshold"));
        pp.setParameters(p);
        pp.pick(tic_gf, tics_pp);

        if (tics_pp.size())
        {
          LOG_INFO << "Found " << tics_pp.size() << " auto-rt peaks at: ";
          for (Size ipp = 0; ipp != tics_pp.size(); ++ipp) LOG_INFO << " " << tics_pp[ipp].getMZ();
        }
        else
        {
          LOG_INFO << "Found no auto-rt peaks. Change threshold parameters!";
        }
        LOG_INFO << std::endl;

        if (!out_TIC_debug.empty()) // if debug file was given
        { // store intermediate steps for debug
          MSExperiment<> out_debug;
          out_debug.addChromatogram(toChromatogram(tics));
          out_debug.addChromatogram(toChromatogram(tic_gf));

          SignalToNoiseEstimatorMedian<MSSpectrum<> > snt;
          snt.init(tics);
          for (Size is = 0; is < tics.size(); ++is)
          {
            Peak1D peak;
            peak.setMZ(tic[is].getMZ());
            peak.setIntensity(snt.getSignalToNoise(tics[is]));
            tics_sn.push_back(peak);
          }
          out_debug.addChromatogram(toChromatogram(tics_sn));

          out_debug.addChromatogram(toChromatogram(tics_pp));
          // get rid of "native-id" missing warning
          for (Size id = 0; id < out_debug.size(); ++id) out_debug[id].setNativeID(String("spectrum=") + id);

          mzml_file.store(out_TIC_debug, out_debug);
          LOG_DEBUG << "Storing debug AUTO-RT: " << out_TIC_debug << std::endl;
        }

        // add target EICs: for each m/z with no/negative RT, add all combinations of that m/z with auto-RTs
        // duplicate m/z entries will be ignored!
        // all other lines with positive RT values are copied unaffected
        //do not allow doubles
        std::set<double> mz_doubles;
        for (ConsensusMap::Iterator cit = cm_local.begin(); cit != cm_local.end(); ++cit)
        {
          if (cit->getRT() < 0)
          {
            if (mz_doubles.find(cit->getMZ()) == mz_doubles.end())
            {
              mz_doubles.insert(cit->getMZ());
            }
            else
            {
              LOG_INFO << "Found duplicate m/z entry (" << cit->getMZ() << ") for auto-rt. Skipping ..." << std::endl;
              continue;
            }

            ConsensusMap cm_RT_multiplex;
            for (MSSpectrum<>::ConstIterator itp = tics_pp.begin(); itp != tics_pp.end(); ++itp)
            {
              ConsensusFeature f = *cit;
              f.setRT(itp->getMZ());
              cm.push_back(f);
            }

          }
          else
          { // default feature with no auto-rt
            LOG_INFO << "copying feature with RT " << cit->getRT() << std::endl;
            cm.push_back(*cit);
          }
        }

        // resize, since we have more positions now
        vec_single.resize(cm.size());
      }


      // search for each EIC and add up
      Int not_found(0);
      Map<Size, double> quant;

      String description;
      if (fi < in_header.size())
      {
        HeaderInfo info(in_header[fi]);
        description = info.header_description;
      }

      if (fi == 0)
      { // two additional columns for first file (theoretical RT and m/z)
        tf_single_header0 << "" << "";
        tf_single_header1 << "" << "";
        tf_single_header2 << "RT" << "mz";
      }

      // 5 entries for each input file
      tf_single_header0 << File::basename(in[fi]) << "" << "" << "" << "";
      tf_single_header1 << description << "" << "" << "" << "";
      tf_single_header2 << "RTobs" << "dRT" << "mzobs" << "dppm" << "intensity";

      for (Size i = 0; i < cm.size(); ++i)
      {
        //std::cerr << "Rt" << cm[i].getRT() << "  mz: " << cm[i].getMZ() << " R " <<  cm[i].getMetaValue("rank") << "\n";

        double mz_da = mztol * cm[i].getMZ() / 1e6; // mz tolerance in Dalton
        MSExperiment<>::ConstAreaIterator it = exp.areaBeginConst(cm[i].getRT() - rttol / 2,
                                                                  cm[i].getRT() + rttol / 2,
                                                                  cm[i].getMZ() - mz_da,
                                                                  cm[i].getMZ() + mz_da);
        Peak2D max_peak;
        max_peak.setIntensity(0);
        max_peak.setRT(cm[i].getRT());
        max_peak.setMZ(cm[i].getMZ());
        for (; it != exp.areaEndConst(); ++it)
        {
          if (max_peak.getIntensity() < it->getIntensity())
          {
            max_peak.setIntensity(it->getIntensity());
            max_peak.setRT(it.getRT());
            max_peak.setMZ(it->getMZ());
          }
        }
        double ppm = 0; // observed m/z offset

        if (max_peak.getIntensity() == 0)
        {
          ++not_found;
        }
        else
        {
          // take median for m/z found
          std::vector<double> mz;
          MSExperiment<>::Iterator itm = exp.RTBegin(max_peak.getRT());
          SignedSize low = std::min<SignedSize>(std::distance(exp.begin(), itm), rt_collect);
          SignedSize high = std::min<SignedSize>(std::distance(itm, exp.end()) - 1, rt_collect);
          MSExperiment<>::AreaIterator itt = exp.areaBegin((itm - low)->getRT() - 0.01, (itm + high)->getRT() + 0.01, cm[i].getMZ() - mz_da, cm[i].getMZ() + mz_da);
          for (; itt != exp.areaEnd(); ++itt)
          {
            mz.push_back(itt->getMZ());
            //std::cerr << "ppm: " << itt.getRT() << " " <<  itt->getMZ() << " " << itt->getIntensity() << std::endl;
          }

          if ((SignedSize)mz.size() > (low + high + 1)) LOG_WARN << "Compound " << i << " has overlapping peaks [" << mz.size() << "/" << low + high + 1 << "]" << std::endl;

          if (!mz.empty())
          {
            double avg_mz = std::accumulate(mz.begin(), mz.end(), 0.0) / double(mz.size());
            //std::cerr << "avg: " << avg_mz << "\n";
            ppm = (avg_mz - cm[i].getMZ()) / cm[i].getMZ() * 1e6;
          }

        }

        // appending the second column set requires separator
        String append_sep = (fi == 0 ? "" : out_sep);

        vec_single[i] += append_sep; // new line
        if (fi == 0)
        {
          vec_single[i] += String(cm[i].getRT()) + out_sep +
                           String(cm[i].getMZ()) + out_sep;
        }
        vec_single[i] += String(max_peak.getRT()) + out_sep +
                         String(max_peak.getRT() - cm[i].getRT()) + out_sep +
                         String(max_peak.getMZ()) + out_sep +
                         String(ppm)  + out_sep +
                         String(max_peak.getIntensity());
      }

      if (not_found) LOG_INFO << "Missing peaks for " << not_found << " compounds in file '" << in[fi] << "'.\n";
    }

    //-------------------------------------------------------------
    // create header
    //-------------------------------------------------------------
    vec_single.insert(vec_single.begin(), ListUtils::concatenate(tf_single_header2, out_sep));
    vec_single.insert(vec_single.begin(), ListUtils::concatenate(tf_single_header1, out_sep));
    vec_single.insert(vec_single.begin(), ListUtils::concatenate(tf_single_header0, out_sep));

    //-------------------------------------------------------------
    // writing output
    //-------------------------------------------------------------
    TextFile tf;
    for (std::vector<String>::iterator v_it = vec_single.begin(); v_it != vec_single.end(); ++v_it)
    {
      tf.addLine(*v_it);
    }
    tf.store(out);

    return EXECUTION_OK;
  }
  ExitCodes main_(int, const char**) override
  {
    //-------------------------------------------------------------
    // parameter handling
    //-------------------------------------------------------------

    String in = getStringOption_("in");
    String out = getStringOption_("out");

    IndexedMzMLFileLoader loader; 

    //-------------------------------------------------------------
    // loading input
    //-------------------------------------------------------------
    OnDiscPeakMap exp;
    loader.load(in, exp);

    // We could write out this warning in the constructor if no spectra have come our way ...
    /*
    if (ms_exp_raw.empty() && ms_exp_raw.getChromatograms().size() == 0)
    {
      LOG_WARN << "The given file does not contain any conventional peak data, but might"
                  " contain chromatograms. This tool currently cannot handle them, sorry.";
      return INCOMPATIBLE_INPUT_DATA;
    }
    */

    // We could check with the first spectrum that we process whether it fulfills the requirements
    // check for peak type (profile data required)
    /*
    if (!ms_exp_raw.empty() && PeakTypeEstimator().estimateType(ms_exp_raw[0].begin(), ms_exp_raw[0].end()) == SpectrumSettings::CENTROID)
    {
      writeLog_("Warning: OpenMS peak type estimation indicates that this is not profile data!");
    }
    */

    // We can check each spectrum and throw an error if it is not sorted
    // check if spectra are sorted
    /*
    for (Size i = 0; i < ms_exp_raw.size(); ++i)
    {
      if (!ms_exp_raw[i].isSorted())
      {
        writeLog_("Error: Not all spectra are sorted according to peak m/z positions. Use FileFilter to sort the input!");
        return INCOMPATIBLE_INPUT_DATA;
      }
    }
    */

    // We can check each chromatogram and throw an error if it is not sorted
    //check if chromatograms are sorted
    /*
    for (Size i = 0; i < ms_exp_raw.getChromatograms().size(); ++i)
    {
      if (!ms_exp_raw.getChromatogram(i).isSorted())
      {
        writeLog_("Error: Not all chromatograms are sorted according to peak m/z positions. Use FileFilter to sort the input!");
        return INCOMPATIBLE_INPUT_DATA;
      }
    }
    */

    //-------------------------------------------------------------
    // pick
    //-------------------------------------------------------------
    PeakMap ms_exp_peaks;

    ///////////////////////////////////
    // Create PeakPickerHiRes and hand it to the PPHiResMzMLConsumer
    ///////////////////////////////////
    Param pepi_param = getParam_().copy("algorithm:", true);
    writeDebug_("Parameters passed to LowMemPeakPickerHiRes", pepi_param, 3);

    PeakPickerHiRes pp;
    pp.setLogType(log_type_);
    pp.setParameters(pepi_param);
    pp.pickExperiment(exp, ms_exp_peaks);

    //-------------------------------------------------------------
    // writing output
    //-------------------------------------------------------------
    //annotate output with data processing info
    addDataProcessing_(ms_exp_peaks, getProcessingInfo_(DataProcessing::PEAK_PICKING));

    MzMLFile mz_data_file;
    mz_data_file.setLogType(log_type_);
    mz_data_file.store(out, ms_exp_peaks);

    return EXECUTION_OK;
  }
예제 #3
0
  ExitCodes main_(int, const char **) override
  {
    //-------------------------------------------------------------
    // parameter handling
    //-------------------------------------------------------------

    in = getStringOption_("in");
    out = getStringOption_("out");
    String process_option = getStringOption_("processOption");

    Param pepi_param = getParam_().copy("algorithm:", true);
    writeDebug_("Parameters passed to PeakPickerHiRes", pepi_param, 3);

    PeakPickerHiRes pp;
    pp.setLogType(log_type_);
    pp.setParameters(pepi_param);

    if (process_option == "lowmemory")
    {
      return doLowMemAlgorithm(pp);
    }

    //-------------------------------------------------------------
    // loading input
    //-------------------------------------------------------------
    MzMLFile mz_data_file;
    mz_data_file.setLogType(log_type_);
    PeakMap ms_exp_raw;
    mz_data_file.load(in, ms_exp_raw);

    if (ms_exp_raw.empty() && ms_exp_raw.getChromatograms().size() == 0)
    {
      LOG_WARN << "The given file does not contain any conventional peak data, but might"
                  " contain chromatograms. This tool currently cannot handle them, sorry.";
      return INCOMPATIBLE_INPUT_DATA;
    }

    //check if spectra are sorted
    for (Size i = 0; i < ms_exp_raw.size(); ++i)
    {
      if (!ms_exp_raw[i].isSorted())
      {
        writeLog_("Error: Not all spectra are sorted according to peak m/z positions. Use FileFilter to sort the input!");
        return INCOMPATIBLE_INPUT_DATA;
      }
    }

    //check if chromatograms are sorted
    for (Size i = 0; i < ms_exp_raw.getChromatograms().size(); ++i)
    {
      if (!ms_exp_raw.getChromatogram(i).isSorted())
      {
        writeLog_("Error: Not all chromatograms are sorted according to peak m/z positions. Use FileFilter to sort the input!");
        return INCOMPATIBLE_INPUT_DATA;
      }
    }

    //-------------------------------------------------------------
    // pick
    //-------------------------------------------------------------
    PeakMap ms_exp_peaks;
    bool check_spectrum_type = !getFlag_("force");
    pp.pickExperiment(ms_exp_raw, ms_exp_peaks, check_spectrum_type);

    //-------------------------------------------------------------
    // writing output
    //-------------------------------------------------------------
    //annotate output with data processing info
    addDataProcessing_(ms_exp_peaks, getProcessingInfo_(DataProcessing::PEAK_PICKING));
    mz_data_file.store(out, ms_exp_peaks);

    return EXECUTION_OK;
  }