/**
* Preprocess the spectra in a given folder and merge those with a different collision energy
* (from the same compound). All results are saved!!!!
*
* @param folder the folder
*/
private void preprocessUnsorted(String folder, double mzabs, double mzppm) {
// loop over all files in folder
File f = new File(folder);
File files[] = f.listFiles();
Arrays.sort(files);
// create new folder
String path = folder + "Merged";
new File(path).mkdir();
int temp = 0;
Map<String, List<File>> pubchemToFiles = new HashMap<String, List<File>>();
for (int i = 0; i < files.length; i++) {
if (files[i].isFile()) {
WrapperSpectrum spectrum = new WrapperSpectrum(files[i].toString());
if (pubchemToFiles.containsKey(Integer.toString(spectrum.getCID())))
pubchemToFiles.get(Integer.toString(spectrum.getCID())).add(files[i]);
else {
List<File> fileList = new ArrayList<File>();
fileList.add(files[i]);
pubchemToFiles.put(Integer.toString(spectrum.getCID()), fileList);
}
System.out.println(files[i].toString() + spectrum.getCID());
}
}
for (String pubchemID : pubchemToFiles.keySet()) {
String mergedNames = "";
Vector<WrapperSpectrum> spectra = new Vector<WrapperSpectrum>();
String lastFile = "";
for (File file : pubchemToFiles.get(pubchemID)) {
WrapperSpectrum spectrum = new WrapperSpectrum(file.toString());
// no pubchem id given ... skip this spectrum!
if (spectrum.getCID() == 0) continue;
spectra.add(spectrum);
mergedNames += file.getName().split("\\.")[0];
lastFile = file.toString();
}
// now merge the peaks from the different collision energies
Vector<Double> mergedPeaks = mergePeaks(spectra, mzabs, mzppm);
// now write back those peaks into a new file
String line = "";
try {
BufferedReader reader = new BufferedReader(new FileReader(lastFile));
line += reader.readLine() + "\n";
while (line != null && !line.contains("PK$NUM_PEAK:")) {
String currentLine = reader.readLine();
if (currentLine.contains("PK$NUM_PEAK:")) break;
else line += currentLine + "\n";
}
line += "PK$NUM_PEAK: " + mergedPeaks.size() + "\n";
line += "PK$PEAK: m/z int. rel.int.\n";
for (int k = 0; k < mergedPeaks.size(); k++) {
// those intensities should be in sync with the merged peaks....quick and dirty...
line +=
" "
+ mergedPeaks.get(k)
+ " "
+ this.newIntensity.get(k)
+ " "
+ this.newRelIntensity.get(k)
+ "\n";
}
line += "//";
// write to file
File outFile = new File(path + "/" + mergedNames + ".txt");
System.out.println(mergedNames);
FileWriter out = new FileWriter(outFile);
out.write(line);
out.close();
} catch (FileNotFoundException e) {
System.err.println("File not found: " + e.getMessage());
} catch (IOException e) {
System.err.println("Error: " + e.getMessage());
}
}
}
/**
* Preprocess the spectra in a given folder and merge those with a different collision energy
* (from the same compound). All results are saved!!!!
*
* @param folder the folder
*/
private void preprocess(String folder, double threshold) {
// loop over all files in folder
File f = new File(folder);
File files[] = f.listFiles();
Arrays.sort(files);
// create new folder
String path = folder + "Merged";
new File(path).mkdir();
int temp = 0;
for (int i = 0; i < (files.length - 1); i++) {
if (files[i].isFile()) {
WrapperSpectrum spectrum = new WrapperSpectrum(files[i].toString());
// no pubchem id given ... skip this spectrum!
if (spectrum.getCID() == 0) continue;
temp = i;
Vector<WrapperSpectrum> spectra = new Vector<WrapperSpectrum>();
spectra.add(spectrum);
// next file
WrapperSpectrum spectrumTemp = new WrapperSpectrum(files[i + 1].toString());
// same InchI
int j = 2;
String mergedNames = files[i].getName().split("\\.")[0];
while (spectrum.getInchI().compareTo(spectrumTemp.getInchI()) == 0
&& (i + j) <= files.length)
// while(spectrum.getTrivialName().equals(spectrumTemp.getTrivialName()) && (i+j) <=
// files.length)
{
// same InchI --> add to list
spectra.add(spectrumTemp);
mergedNames += files[i + (j - 1)].getName().split("\\.")[0];
// fix for last file
if ((i + j) < files.length) spectrumTemp = new WrapperSpectrum(files[i + j].toString());
j++;
}
i = i + (j - 2);
// now merge the peaks from the different collision energies
Vector<Double> mergedPeaks = mergePeaks(spectra, threshold, 0.0);
// now write back those peaks into a new file
String line = "";
try {
BufferedReader reader = new BufferedReader(new FileReader(files[temp].toString()));
line += reader.readLine() + "\n";
while (line != null && !line.contains("AC$ANALYTICAL_CONDITION: COLLISION_ENERGY")) {
line += reader.readLine() + "\n";
}
line += "PK$NUM_PEAK: " + mergedPeaks.size() + "\n";
line += "PK$PEAK: m/z int. rel.int.\n";
for (int k = 0; k < mergedPeaks.size(); k++) {
// those intensities should be in sync with the merged peaks....quick and dirty...
line +=
" "
+ mergedPeaks.get(k)
+ " "
+ this.newIntensity.get(k)
+ " "
+ this.newRelIntensity.get(k)
+ "\n";
}
// write to file
File outFile = new File(path + "/" + mergedNames + ".txt");
System.out.println(mergedNames);
FileWriter out = new FileWriter(outFile);
out.write(line);
out.close();
} catch (FileNotFoundException e) {
System.err.println("File not found: " + e.getMessage());
} catch (IOException e) {
System.err.println("Error: " + e.getMessage());
}
}
}
}
