public @Nonnull Range getDataRTRange(int msLevel) {
// check if we have this value already cached
Range rtRange = dataRTRange.get(msLevel);
if (rtRange != null) return rtRange;
// find the value
for (Scan scan : scans.values()) {
// ignore scans of other ms levels
if ((msLevel != 0) && (scan.getMSLevel() != msLevel)) continue;
if (rtRange == null) rtRange = new Range(scan.getRetentionTime());
else rtRange.extendRange(scan.getRetentionTime());
}
// cache the value
if (rtRange != null) dataRTRange.put(msLevel, rtRange);
else rtRange = new Range(0);
return rtRange;
}
public @Nonnull Range getDataMZRange(int msLevel) {
// check if we have this value already cached
Range mzRange = dataMZRange.get(msLevel);
if (mzRange != null) return mzRange;
// find the value
for (Scan scan : scans.values()) {
// ignore scans of other ms levels
if ((msLevel != 0) && (scan.getMSLevel() != msLevel)) continue;
if (mzRange == null) mzRange = scan.getMZRange();
else mzRange.extendRange(scan.getMZRange());
}
// cache the value, if we found any
if (mzRange != null) dataMZRange.put(msLevel, mzRange);
else mzRange = new Range(0);
return mzRange;
}
/** @see net.sf.mzmine.datamodel.RawDataFile#getScanNumbers(int, double, double) */
public @Nonnull int[] getScanNumbers(int msLevel, @Nonnull Range rtRange) {
assert rtRange != null;
ArrayList<Integer> eligibleScanNumbers = new ArrayList<Integer>();
Enumeration<StorableScan> scansEnum = scans.elements();
while (scansEnum.hasMoreElements()) {
Scan scan = scansEnum.nextElement();
if ((scan.getMSLevel() == msLevel) && (rtRange.contains(scan.getRetentionTime())))
eligibleScanNumbers.add(scan.getScanNumber());
}
int[] numbersArray = CollectionUtils.toIntArray(eligibleScanNumbers);
Arrays.sort(numbersArray);
return numbersArray;
}
/** @see Runnable#run() */
public void run() {
if ((mzWeight == 0) && (rtWeight == 0)) {
setStatus(TaskStatus.ERROR);
errorMessage = "Cannot run alignment, all the weight parameters are zero";
return;
}
setStatus(TaskStatus.PROCESSING);
logger.info("Running join aligner");
// Remember how many rows we need to process. Each row will be processed
// twice, first for score calculation, second for actual alignment.
for (int i = 0; i < peakLists.length; i++) {
totalRows += peakLists[i].getNumberOfRows() * 2;
}
// Collect all data files
Vector<RawDataFile> allDataFiles = new Vector<RawDataFile>();
for (PeakList peakList : peakLists) {
for (RawDataFile dataFile : peakList.getRawDataFiles()) {
// Each data file can only have one column in aligned peak list
if (allDataFiles.contains(dataFile)) {
setStatus(TaskStatus.ERROR);
errorMessage =
"Cannot run alignment, because file "
+ dataFile
+ " is present in multiple peak lists";
return;
}
allDataFiles.add(dataFile);
}
}
// Create a new aligned peak list
alignedPeakList = new SimplePeakList(peakListName, allDataFiles.toArray(new RawDataFile[0]));
// Iterate source peak lists
for (PeakList peakList : peakLists) {
// Create a sorted set of scores matching
TreeSet<RowVsRowScore> scoreSet = new TreeSet<RowVsRowScore>();
PeakListRow allRows[] = peakList.getRows();
// Calculate scores for all possible alignments of this row
for (PeakListRow row : allRows) {
if (isCanceled()) return;
// Calculate limits for a row with which the row can be aligned
Range mzRange = mzTolerance.getToleranceRange(row.getAverageMZ());
Range rtRange = rtTolerance.getToleranceRange(row.getAverageRT());
// Get all rows of the aligned peaklist within parameter limits
PeakListRow candidateRows[] = alignedPeakList.getRowsInsideScanAndMZRange(rtRange, mzRange);
// Calculate scores and store them
for (PeakListRow candidate : candidateRows) {
if (sameChargeRequired) {
if (!PeakUtils.compareChargeState(row, candidate)) continue;
}
if (sameIDRequired) {
if (!PeakUtils.compareIdentities(row, candidate)) continue;
}
if (compareIsotopePattern) {
IsotopePattern ip1 = row.getBestIsotopePattern();
IsotopePattern ip2 = candidate.getBestIsotopePattern();
if ((ip1 != null) && (ip2 != null)) {
ParameterSet isotopeParams =
parameters
.getParameter(JoinAlignerParameters.compareIsotopePattern)
.getEmbeddedParameters();
if (!IsotopePatternScoreCalculator.checkMatch(ip1, ip2, isotopeParams)) {
continue;
}
}
}
RowVsRowScore score =
new RowVsRowScore(
row, candidate, mzRange.getSize() / 2, mzWeight, rtRange.getSize() / 2, rtWeight);
scoreSet.add(score);
}
processedRows++;
}
// Create a table of mappings for best scores
Hashtable<PeakListRow, PeakListRow> alignmentMapping =
new Hashtable<PeakListRow, PeakListRow>();
// Iterate scores by descending order
Iterator<RowVsRowScore> scoreIterator = scoreSet.iterator();
while (scoreIterator.hasNext()) {
RowVsRowScore score = scoreIterator.next();
// Check if the row is already mapped
if (alignmentMapping.containsKey(score.getPeakListRow())) continue;
// Check if the aligned row is already filled
if (alignmentMapping.containsValue(score.getAlignedRow())) continue;
alignmentMapping.put(score.getPeakListRow(), score.getAlignedRow());
}
// Align all rows using mapping
for (PeakListRow row : allRows) {
PeakListRow targetRow = alignmentMapping.get(row);
// If we have no mapping for this row, add a new one
if (targetRow == null) {
targetRow = new SimplePeakListRow(newRowID);
newRowID++;
alignedPeakList.addRow(targetRow);
}
// Add all peaks from the original row to the aligned row
for (RawDataFile file : row.getRawDataFiles()) {
targetRow.addPeak(file, row.getPeak(file));
}
// Add all non-existing identities from the original row to the
// aligned row
PeakUtils.copyPeakListRowProperties(row, targetRow);
processedRows++;
}
} // Next peak list
// Add new aligned peak list to the project
MZmineProject currentProject = MZmineCore.getCurrentProject();
currentProject.addPeakList(alignedPeakList);
// Add task description to peakList
alignedPeakList.addDescriptionOfAppliedTask(
new SimplePeakListAppliedMethod("Join aligner", parameters));
logger.info("Finished join aligner");
setStatus(TaskStatus.FINISHED);
}
public void setValue(Range value) {
minTxtField.setText(format.format(value.getMin()));
maxTxtField.setText(format.format(value.getMax()));
}
/**
* @see org.xml.sax.helpers.DefaultHandler#endElement(java.lang.String, java.lang.String,
* java.lang.String)
*/
public void endElement(String namespaceURI, String sName, String qName) throws SAXException {
if (canceled) throw new SAXException("Parsing canceled");
// <NAME>
if (qName.equals(PeakListElementName_2_3.PEAKLIST_NAME.getElementName())) {
name = getTextOfElement();
logger.info("Loading peak list: " + name);
peakListName = name;
}
// <PEAKLIST_DATE>
if (qName.equals(PeakListElementName_2_3.PEAKLIST_DATE.getElementName())) {
dateCreated = getTextOfElement();
}
// <QUANTITY>
if (qName.equals(PeakListElementName_2_3.QUANTITY.getElementName())) {
String text = getTextOfElement();
totalRows = Integer.parseInt(text);
}
// <RAW_FILE>
if (qName.equals(PeakListElementName_2_3.RAWFILE.getElementName())) {
rawDataFileID = getTextOfElement();
RawDataFile dataFile = dataFilesIDMap.get(rawDataFileID);
if (dataFile == null) {
throw new SAXException(
"Cannot open peak list, because raw data file " + rawDataFileID + " is missing.");
}
currentPeakListDataFiles.add(dataFile);
}
// <SCAN_ID>
if (qName.equals(PeakListElementName_2_3.SCAN_ID.getElementName())) {
byte[] bytes = Base64.decodeToBytes(getTextOfElement());
// make a data input stream
DataInputStream dataInputStream = new DataInputStream(new ByteArrayInputStream(bytes));
scanNumbers = new int[numOfMZpeaks];
for (int i = 0; i < numOfMZpeaks; i++) {
try {
scanNumbers[i] = dataInputStream.readInt();
} catch (IOException ex) {
throw new SAXException(ex);
}
}
}
// <REPRESENTATIVE_SCAN>
if (qName.equals(PeakListElementName_2_3.REPRESENTATIVE_SCAN.getElementName())) {
representativeScan = Integer.valueOf(getTextOfElement());
}
// <FRAGMENT_SCAN>
if (qName.equals(PeakListElementName_2_3.FRAGMENT_SCAN.getElementName())) {
fragmentScan = Integer.valueOf(getTextOfElement());
}
// <MASS>
if (qName.equals(PeakListElementName_2_3.MZ.getElementName())) {
byte[] bytes = Base64.decodeToBytes(getTextOfElement());
// make a data input stream
DataInputStream dataInputStream = new DataInputStream(new ByteArrayInputStream(bytes));
masses = new double[numOfMZpeaks];
for (int i = 0; i < numOfMZpeaks; i++) {
try {
masses[i] = (double) dataInputStream.readFloat();
} catch (IOException ex) {
throw new SAXException(ex);
}
}
}
// <HEIGHT>
if (qName.equals(PeakListElementName_2_3.HEIGHT.getElementName())) {
byte[] bytes = Base64.decodeToBytes(getTextOfElement());
// make a data input stream
DataInputStream dataInputStream = new DataInputStream(new ByteArrayInputStream(bytes));
intensities = new double[numOfMZpeaks];
for (int i = 0; i < numOfMZpeaks; i++) {
try {
intensities[i] = (double) dataInputStream.readFloat();
} catch (IOException ex) {
throw new SAXException(ex);
}
}
}
// <PEAK>
if (qName.equals(PeakListElementName_2_3.PEAK.getElementName())) {
DataPoint[] mzPeaks = new DataPoint[numOfMZpeaks];
Range peakRTRange = null, peakMZRange = null, peakIntensityRange = null;
RawDataFile dataFile = dataFilesIDMap.get(peakColumnID);
if (dataFile == null)
throw new SAXException("Error in project: data file " + peakColumnID + " not found");
for (int i = 0; i < numOfMZpeaks; i++) {
Scan sc = dataFile.getScan(scanNumbers[i]);
double retentionTime = sc.getRetentionTime();
double mz = masses[i];
double intensity = intensities[i];
if ((peakRTRange == null) || (peakIntensityRange == null)) {
peakRTRange = new Range(retentionTime);
peakIntensityRange = new Range(intensity);
} else {
peakRTRange.extendRange(retentionTime);
peakIntensityRange.extendRange(intensity);
}
if (mz > 0.0) {
mzPeaks[i] = new SimpleDataPoint(mz, intensity);
if (peakMZRange == null) peakMZRange = new Range(mz);
else peakMZRange.extendRange(mz);
}
}
FeatureStatus status = FeatureStatus.valueOf(peakStatus);
SimpleFeature peak =
new SimpleFeature(
dataFile,
mass,
rt,
height,
area,
scanNumbers,
mzPeaks,
status,
representativeScan,
fragmentScan,
peakRTRange,
peakMZRange,
peakIntensityRange);
peak.setCharge(currentPeakCharge);
if (currentIsotopes.size() > 0) {
SimpleIsotopePattern newPattern =
new SimpleIsotopePattern(
currentIsotopes.toArray(new DataPoint[0]),
currentIsotopePatternStatus,
currentIsotopePatternDescription);
peak.setIsotopePattern(newPattern);
currentIsotopes.clear();
}
buildingRow.addPeak(dataFile, peak);
}
// <IDENTITY_PROPERTY>
if (qName.equals(PeakListElementName_2_3.IDPROPERTY.getElementName())) {
identityProperties.put(identityPropertyName, getTextOfElement());
}
// <PEAK_IDENTITY>
if (qName.equals(PeakListElementName_2_3.PEAK_IDENTITY.getElementName())) {
SimplePeakIdentity identity = new SimplePeakIdentity(identityProperties);
buildingRow.addPeakIdentity(identity, preferred);
}
// <ROW>
if (qName.equals(PeakListElementName_2_3.ROW.getElementName())) {
buildingPeakList.addRow(buildingRow);
buildingRow = null;
parsedRows++;
}
// <ISOTOPE>
if (qName.equals(PeakListElementName_2_3.ISOTOPE.getElementName())) {
String text = getTextOfElement();
String items[] = text.split(":");
double mz = Double.valueOf(items[0]);
double intensity = Double.valueOf(items[1]);
DataPoint isotope = new SimpleDataPoint(mz, intensity);
currentIsotopes.add(isotope);
}
if (qName.equals(PeakListElementName_2_3.METHOD_NAME.getElementName())) {
String appliedMethod = getTextOfElement();
appliedMethods.add(appliedMethod);
}
if (qName.equals(PeakListElementName_2_3.METHOD_PARAMETERS.getElementName())) {
String appliedMethodParam = getTextOfElement();
appliedMethodParameters.add(appliedMethodParam);
}
}