public RawDataCollection(IRawDataPlus rawFile)
{
rawFileName = rawFile.FileName;
instrument = rawFile.GetInstrumentData().Name;
isExactive = instrument.ToLower().Contains("exactive");
isBoxCar = rawFile.GetScanEventForScanNumber(1).MassRangeCount > 1;
centroidStreams = new Dictionary <int, CentroidStreamData>();
segmentedScans = new Dictionary <int, SegmentedScanData>();
trailerExtras = new Dictionary <int, TrailerExtraData>();
precursorScans = new Dictionary <int, PrecursorScanData>();
precursorMasses = new Dictionary <int, PrecursorMassData>();
retentionTimes = new Dictionary <int, double>();
methodData = new Containers.MethodData();
quantData = new QuantDataCollection();
metaData = new ScanMetaDataCollection();
peakData = new PrecursorPeakDataCollection();
this.ExtractScanIndex(rawFile);
this.ExtractMethodData(rawFile);
}
public static void GetMetricsData(this MetricsData metricsData, ScanMetaDataCollection metaData, RawDataCollection rawData, IRawDataPlus rawFile, QuantDataCollection quantData = null)
{
List <Operations> operations = new List <Operations> {
Operations.ScanIndex, Operations.RetentionTimes, Operations.MethodData, Operations.MetaData
};
if (!rawData.isBoxCar)
{
operations.Add(Operations.PeakRetAndInt);
operations.Add(Operations.PeakShape);
}
rawData.Check(rawFile, operations);
metricsData.RawFileName = rawData.rawFileName;
metricsData.Instrument = rawData.instrument;
metricsData.MS1Analyzer = rawData.methodData.MassAnalyzers[MSOrderType.Ms];
metricsData.MS2Analyzer = rawData.methodData.MassAnalyzers[MSOrderType.Ms2];
metricsData.TotalAnalysisTime = rawData.retentionTimes[rawData.scanIndex.ScanEnumerators[MSOrderType.Any].Last()] -
rawData.retentionTimes[rawData.scanIndex.ScanEnumerators[MSOrderType.Any].First()];
metricsData.TotalScans = rawData.scanIndex.allScans.Count();
metricsData.MS1Scans = rawData.scanIndex.ScanEnumerators[MSOrderType.Ms].Length;
metricsData.MS2Scans = rawData.scanIndex.ScanEnumerators[MSOrderType.Ms2].Length;
if (rawData.methodData.AnalysisOrder == MSOrderType.Ms3)
{
metricsData.MS3Analyzer = rawData.methodData.MassAnalyzers[MSOrderType.Ms3];
metricsData.MS3Scans = rawData.scanIndex.ScanEnumerators[MSOrderType.Ms3].Length;
}
else
{
metricsData.MS3Analyzer = MassAnalyzerType.Any;
metricsData.MS3Scans = 0;
}
metricsData.MSOrder = rawData.methodData.AnalysisOrder;
List <double> ms2intensities = new List <double>();
List <double> precursorIntensities = new List <double>();
List <double> ms1fillTimes = new List <double>();
List <double> ms2fillTimes = new List <double>();
List <double> ms3fillTimes = new List <double>();
List <double> ms2scansPerCycle = new List <double>();
List <double> dutyCycles = new List <double>();
List <double> fractionConsuming80 = new List <double>();
foreach (int scan in rawData.scanIndex.ScanEnumerators[MSOrderType.Ms])
{
ms1fillTimes.Add(metaData[scan].FillTime);
ms2scansPerCycle.Add(metaData[scan].MS2ScansPerCycle);
dutyCycles.Add(metaData[scan].DutyCycle);
}
foreach (int scan in rawData.scanIndex.ScanEnumerators[MSOrderType.Ms2])
{
precursorIntensities.Add(rawData.peakData[scan].ParentIntensity);
ms2intensities.Add(metaData[scan].SummedIntensity);
ms2fillTimes.Add(metaData[scan].FillTime);
fractionConsuming80.Add(metaData[scan].FractionConsumingTop80PercentTotalIntensity);
}
if (rawData.methodData.AnalysisOrder == MSOrderType.Ms3)
{
foreach (int scan in rawData.scanIndex.ScanEnumerators[MSOrderType.Ms3])
{
ms3fillTimes.Add(metaData[scan].FillTime);
}
}
metricsData.MedianPrecursorIntensity = precursorIntensities.ToArray().Percentile(50);
metricsData.MedianMs2FractionConsumingTop80PercentTotalIntensity = fractionConsuming80.ToArray().Percentile(50);
metricsData.MedianSummedMS2Intensity = ms2intensities.ToArray().Percentile(50);
metricsData.MedianMS1FillTime = ms1fillTimes.ToArray().Percentile(50);
metricsData.MedianMS2FillTime = ms2fillTimes.ToArray().Percentile(50);
if (rawData.methodData.AnalysisOrder == MSOrderType.Ms3)
{
metricsData.MedianMS3FillTime = ms3fillTimes.ToArray().Percentile(50);
}
else
{
metricsData.MedianMS3FillTime = -1;
}
metricsData.MeanTopN = ms2scansPerCycle.Average();
metricsData.MeanDutyCycle = dutyCycles.Average();
metricsData.MS1ScanRate = metricsData.MS1Scans / metricsData.TotalAnalysisTime;
metricsData.MS2ScanRate = metricsData.MS2Scans / metricsData.TotalAnalysisTime;
metricsData.MS3ScanRate = metricsData.MS3Scans / metricsData.TotalAnalysisTime;
// only do the following if it isn't a boxcar experiment
if (!rawData.isBoxCar)
{
metricsData.MedianBaselinePeakWidth = rawData.peakData.PeakShapeMedians.Width.P10;
metricsData.MedianHalfHeightPeakWidth = rawData.peakData.PeakShapeMedians.Width.P50;
// we can't access the instrument method in Linux, so we will assume the gradient length is the length of the MS acquisition
metricsData.Gradient = rawData.retentionTimes[rawData.scanIndex.allScans.Keys.Max()];
metricsData.PeakCapacity = metricsData.Gradient / metricsData.MedianHalfHeightPeakWidth;
metricsData.MedianAsymmetryFactor = rawData.peakData.PeakShapeMedians.Asymmetry.P10;
}
// add isolation interference
metricsData.MedianMs1IsolationInterference = (from scan in rawData.scanIndex.ScanEnumerators[rawData.methodData.AnalysisOrder]
select rawData.metaData[scan].Ms1IsolationInterference).ToArray().Percentile(50);
// now add the quant meta data, if quant was performed
double medianReporterIntensity = 0;
QuantMetaData quantMetaData = new QuantMetaData();
SerializableDictionary <string, double> medianReporterIntensityByChannel = new SerializableDictionary <string, double>();
if (quantData != null & rawData.Performed.Contains(Operations.Quantification))
{
string reagent = quantData.LabelingReagents;
string[] allTags = new LabelingReagents().Reagents[reagent].Labels;
List <double> allChannels = new List <double>();
Dictionary <string, List <double> > byChannel = new Dictionary <string, List <double> >();
foreach (string tag in allTags)
{
byChannel.Add(tag, new List <double>());
}
foreach (int scan in rawData.scanIndex.ScanEnumerators[rawData.methodData.AnalysisOrder])
{
foreach (string tag in allTags)
{
byChannel[tag].Add(quantData[scan][tag].Intensity);
allChannels.Add(quantData[scan][tag].Intensity);
}
}
medianReporterIntensity = allChannels.ToArray().Percentile(50);
foreach (string tag in allTags)
{
medianReporterIntensityByChannel[tag] = byChannel[tag].ToArray().Percentile(50);
}
quantMetaData.medianReporterIntensity = medianReporterIntensity;
quantMetaData.medianReporterIntensityByChannel = medianReporterIntensityByChannel;
quantMetaData.quantTags = allTags;
metricsData.QuantMeta = quantMetaData;
metricsData.IncludesQuant = true;
}
}
public static void AggregateMetaData(this ScanMetaDataCollection metaData, RawDataCollection rawData, IRawDataPlus rawFile)
{
List <Operations> operations = new List <Operations>()
{
Operations.ScanIndex, Operations.MethodData,
Operations.RetentionTimes, Operations.TrailerExtras, Operations.PrecursorScans
};
if (rawData.methodData.MassAnalyzers[MSOrderType.Ms] == MassAnalyzerType.MassAnalyzerFTMS)
{
operations.Add(Operations.Ms1CentroidStreams);
}
else
{
operations.Add(Operations.Ms1SegmentedScans);
}
if (rawData.methodData.MassAnalyzers[MSOrderType.Ms2] == MassAnalyzerType.MassAnalyzerFTMS)
{
operations.Add(Operations.Ms2CentroidStreams);
}
else
{
operations.Add(Operations.Ms2SegmentedScans);
}
if (rawData.methodData.AnalysisOrder == MSOrderType.Ms3)
{
if (rawData.methodData.MassAnalyzers[MSOrderType.Ms3] == MassAnalyzerType.MassAnalyzerFTMS)
{
operations.Add(Operations.Ms3CentroidStreams);
}
else
{
operations.Add(Operations.Ms3SegmentedScans);
}
}
rawData.Check(rawFile, operations);
ProgressIndicator progress = new ProgressIndicator(rawData.scanIndex.ScanEnumerators[MSOrderType.Any].Count(),
"Formatting scan meta data");
// add a new ScanMetaData class for each scan
foreach (int scan in rawData.scanIndex.ScanEnumerators[MSOrderType.Any])
{
metaData.Add(scan, new ScanMetaData());
}
// get isolation window
double isoWindow;
if (rawData.methodData.AnalysisOrder == MSOrderType.Ms2)
{
isoWindow = rawData.methodData.IsolationWindow.MS2;
}
else
{
isoWindow = rawData.methodData.IsolationWindow.MS3.MS1Window;
}
// get topN
foreach (int scan in rawData.scanIndex.ScanEnumerators[MSOrderType.Ms])
{
// if the ms1 scan has no scan dependents then topN = 0
if (rawFile.GetScanDependents(scan, 0) == null)
{
metaData[scan].MS2ScansPerCycle = 0;
}
else
{
metaData[scan].MS2ScansPerCycle = rawFile.GetScanDependents(scan, 0).ScanDependentDetailArray.Length;
}
}
// go through scans for each ms order sequentially
foreach (MSOrderType MSOrder in new List <MSOrderType> {
MSOrderType.Ms, MSOrderType.Ms2, MSOrderType.Ms3
})
{
int[] scans = rawData.scanIndex.ScanEnumerators[MSOrder];
for (int i = 0; i < scans.Length; i++)
{
metaData[scans[i]].FillTime = rawData.trailerExtras[scans[i]].InjectionTime;
// populate duty cycle
if (i < scans.Length - 1)
{
metaData[scans[i]].DutyCycle = (rawData.retentionTimes[scans[i + 1]] - rawData.retentionTimes[scans[i]]) * 60;
}
else
{
metaData[scans[i]].DutyCycle = 0;
}
// populate scan rate
if (MSOrder == MSOrderType.Ms2 | MSOrder == MSOrderType.Ms3)
{
metaData[scans[i]].MS2ScansPerCycle = metaData[rawData.precursorScans[scans[i]].MasterScan].MS2ScansPerCycle;
}
// populate intensity distributions
if (rawData.methodData.MassAnalyzers[MSOrder] == MassAnalyzerType.MassAnalyzerFTMS)
{
metaData[scans[i]].IntensityDistribution = new Distribution(rawData.centroidStreams[scans[i]].Intensities);
metaData[scans[i]].SummedIntensity = rawData.centroidStreams[scans[i]].Intensities.Sum();
}
else
{
metaData[scans[i]].IntensityDistribution = new Distribution(rawData.segmentedScans[scans[i]].Intensities);
metaData[scans[i]].SummedIntensity = rawData.segmentedScans[scans[i]].Intensities.Sum();
}
// populate fraction of scans consuming 80% of total intensity
if (rawData.methodData.MassAnalyzers[MSOrder] == MassAnalyzerType.MassAnalyzerFTMS)
{
metaData[scans[i]].FractionConsumingTop80PercentTotalIntensity = rawData.centroidStreams[scans[i]].Intensities.FractionOfScansConsumingTotalIntensity(percent: 80);
}
else
{
metaData[scans[i]].FractionConsumingTop80PercentTotalIntensity = rawData.segmentedScans[scans[i]].Intensities.FractionOfScansConsumingTotalIntensity(percent: 80);
}
// calculate ms1 isolation interference
if (rawData.methodData.AnalysisOrder == MSOrder)
{
int preScan = rawData.precursorScans[scans[i]].MasterScan;
metaData[scans[i]].Ms1IsolationInterference = Ms1Interference.CalculateForOneScan(rawData.centroidStreams[preScan],
rawData.precursorMasses[scans[i]].MonoisotopicMZ, isoWindow, rawData.trailerExtras[scans[i]].ChargeState);
}
progress.Update();
}
}
progress.Done();
rawData.Performed.Add(Operations.MetaData);
}
public static void WriteMatrix(RawDataCollection rawData, ScanMetaDataCollection metaData, IRawDataPlus rawFile, QuantDataCollection quantData = null, string outputDirectory = null)
{
string fileName = ReadWrite.GetPathToFile(outputDirectory, rawData.rawFileName, "_Matrix.txt");
CheckIfDone.Check(rawData, rawFile, new List <Operations> {
Operations.ScanIndex, Operations.MethodData, Operations.PrecursorScans,
Operations.RetentionTimes, Operations.PrecursorMasses, Operations.TrailerExtras, Operations.MetaData
});
using (StreamWriter f = new StreamWriter(fileName)) //Open a new file
{
List <int> scans;
if (!rawData.isBoxCar)
{
scans = rawData.scanIndex.ScanEnumerators[rawData.scanIndex.AnalysisOrder].ToList();
}
else
{
scans = rawData.precursorScans.Keys.ToList();
scans.Sort();
}
ProgressIndicator progress = new ProgressIndicator(scans.Count(), "Writing matrix to disk");
f.Write("MS3ScanNumber\tMS2ScanNumber\tMS1ScanNumber\tQuantScanRetTime\tParentScanRetTime\tDutyCycle" +
"\tMS2ScansPerCycle\tParentIonMass\tMonoisotopicMass\tPrecursorCharge\tMS1IsolationInterference");
if (!rawData.isBoxCar)
{
f.Write("\tParentPeakFound");
}
if (rawData.Performed.Contains(Operations.PeakArea) & !rawData.isBoxCar)
{
f.Write("\tParentPeakArea");
}
if (!rawData.isBoxCar)
{
f.Write("\tPeakFirstScan\tPeakMaxScan\tPeakLastScan\tBaseLinePeakWidth(s)\tPeakParentScanIntensity\tPeakMaxIntensity");
}
f.Write("\tMS1IonInjectionTime\tMS2IonInjectionTime" +
"\tMS3IonInjectionTime\tHCDEnergy\tMS1MedianIntensity\tMS2MedianIntensity\t");
if (quantData != null)
{
string reagents = quantData.LabelingReagents;
foreach (string label in new LabelingReagents().Reagents[reagents].Labels)
{
f.Write(label + "Intensity\t");
}
foreach (string label in new LabelingReagents().Reagents[reagents].Labels)
{
f.Write(label + "Mass\t");
}
foreach (string label in new LabelingReagents().Reagents[reagents].Labels)
{
f.Write(label + "Noise\t");
}
foreach (string label in new LabelingReagents().Reagents[reagents].Labels)
{
f.Write(label + "Resolution\t");
}
foreach (string label in new LabelingReagents().Reagents[reagents].Labels)
{
f.Write(label + "Baseline\t");
}
}
f.Write("\n");
foreach (int scan in scans)
{
int ms3scan, ms2scan, masterScan;
if (rawData.scanIndex.AnalysisOrder == MSOrderType.Ms3)
{
ms3scan = rawData.precursorScans[scan].MS3Scan;
ms2scan = rawData.precursorScans[scan].MS2Scan;
masterScan = rawData.precursorScans[scan].MasterScan;
}
else
{
ms3scan = -1;
ms2scan = rawData.precursorScans[scan].MS2Scan;
masterScan = rawData.precursorScans[scan].MasterScan;
}
f.Write(ms3scan.ToString() + "\t" + ms2scan.ToString() + "\t" + masterScan.ToString() + "\t");
f.Write(rawData.retentionTimes[scan].ToString() + "\t" + rawData.retentionTimes[masterScan].ToString() + "\t");
f.Write(metaData[masterScan].DutyCycle.ToString() + "\t" + metaData[masterScan].MS2ScansPerCycle.ToString() + "\t");
f.Write(rawData.precursorMasses[ms2scan].ParentMZ.ToString() + "\t");
f.Write(rawData.precursorMasses[ms2scan].MonoisotopicMZ.ToString() + "\t");
f.Write(rawData.trailerExtras[ms2scan].ChargeState.ToString() + "\t");
f.Write(rawData.metaData[scan].Ms1IsolationInterference.ToString() + "\t");
if (!rawData.isBoxCar)
{
f.Write(rawData.peakData[ms2scan].PeakFound.ToString() + "\t");
}
if (rawData.Performed.Contains(Operations.PeakArea) & !rawData.isBoxCar)
{
f.Write(rawData.peakData[ms2scan].Area.ToString() + "\t");
}
if (!rawData.isBoxCar)
{
f.Write(rawData.peakData[ms2scan].FirstScan.ToString() + "\t");
f.Write(rawData.peakData[ms2scan].MaxScan.ToString() + "\t");
f.Write(rawData.peakData[ms2scan].LastScan.ToString() + "\t");
f.Write((rawData.peakData[ms2scan].BaselineWidth * 60).ToString() + "\t");
f.Write(rawData.peakData[ms2scan].ParentIntensity.ToString() + "\t");
f.Write(rawData.peakData[ms2scan].MaximumIntensity.ToString() + "\t");
}
f.Write(rawData.trailerExtras[masterScan].InjectionTime.ToString() + "\t");
if (rawData.scanIndex.AnalysisOrder == MSOrderType.Ms3)
{
f.Write(rawData.trailerExtras[ms2scan].InjectionTime.ToString() + "\t");
f.Write(rawData.trailerExtras[ms3scan].InjectionTime.ToString() + "\t");
}
else
{
f.Write(rawData.trailerExtras[ms2scan].InjectionTime.ToString() + "\t");
f.Write("-1\t");
}
f.Write(rawData.trailerExtras[scan].HCDEnergy + "\t");
f.Write(metaData[masterScan].IntensityDistribution.P50 + "\t");
f.Write(metaData[ms2scan].IntensityDistribution.P50 + "\t");
if (quantData != null)
{
foreach (string label in quantData[scan].Keys)
{
f.Write(quantData[scan][label].Intensity + "\t");
}
foreach (string label in quantData[scan].Keys)
{
f.Write(quantData[scan][label].Mass + "\t");
}
foreach (string label in quantData[scan].Keys)
{
f.Write(quantData[scan][label].Noise + "\t");
}
foreach (string label in quantData[scan].Keys)
{
f.Write(quantData[scan][label].Resolution + "\t");
}
foreach (string label in quantData[scan].Keys)
{
f.Write(quantData[scan][label].Baseline + "\t");
}
}
f.Write("\n");
progress.Update();
}
progress.Done();
}
}