private IEnumerable<QuantFile> LoadFiles(IEnumerable<string> filePaths, bool ms3Quant = false)
{
MSDataFile.CacheScans = false;
//int largestQuantPeak = 0;
int i = 0;
foreach (TagInformation tag in UsedTags.Values)
{
tag.UniqueTagNumber = i++;
tag.TotalSignal = 0;
tag.NormalizedTotalSignal = 0;
}
int largestQuantPeak = i-1;
//int largestQuantPeak = UsedTags.Values.Select(tag => tag.UniqueTagNumber).Concat(new[] {0}).Max();
foreach (string filePath in filePaths)
{
Log("Processing file:\t" + filePath);
OnUpdateLog("Processing File "+filePath+"...");
QuantFile quantFile = new QuantFile(filePath);
StreamReader basestreamReader = new StreamReader(filePath);
int oldProgress = -1;
using (CsvReader reader = new CsvReader(basestreamReader, true))
{
while (reader.ReadNextRecord()) // go through csv and raw file to extract the info we want
{
int scanNumber = int.Parse(reader["Spectrum number"]);
string filenameID = reader["Filename/id"];
string rawFileName = filenameID.Split('.')[0];
bool isDecoy = reader["DEFLINE"].StartsWith("DECOY_");
ThermoRawFile rawFile;
if (!RawFiles.TryGetValue(rawFileName, out rawFile))
{
throw new ArgumentException("Cannot find this raw file: " + rawFileName + ".raw");
}
if (!rawFile.IsOpen)
{
rawFile.Open();
}
int progress = (int) (100*(double)basestreamReader.BaseStream.Position/basestreamReader.BaseStream.Length);
if (progress != oldProgress)
{
OnProgressUpdate(progress);
oldProgress = progress;
}
//// Set default fragmentation to CAD / HCD
//FragmentationMethod ScanFragMethod = filenameID.Contains(".ETD.")
// ? FragmentationMethod.ETD
// : FragmentationMethod.CAD;
//if (ScanFragMethod == FragmentationMethod.ETD)
//{
// ScanFragMethod = FragmentationMethod.CAD;
// scanNumber += ETDQuantPosition;
//}
// Get the scan object for the sequence ms2 scan
MsnDataScan quantitationMsnScan = rawFile[scanNumber] as MsnDataScan;
double purity = 1;
if (CalculatePurity)
{
double mz = quantitationMsnScan.PrecursorMz;
int charge = quantitationMsnScan.PrecursorCharge;
DoubleRange isolationRange = MzRange.FromDa(mz, PurityWindowInTh);
MSDataScan parentScan = rawFile[quantitationMsnScan.ParentScanNumber];
purity = DeterminePurity(parentScan, mz, charge, isolationRange);
}
if (quantitationMsnScan == null)
{
throw new ArgumentException("Spectrum Number " + scanNumber + " is not a valid MS2 scan from: "+rawFile.FilePath);
}
if (MS3Quant)
{
quantitationMsnScan = null;
// Look forward to find associated MS3 quant scan (based on parent scan number)
int ms3ScanNumber = scanNumber + 1;
while (ms3ScanNumber < rawFile.LastSpectrumNumber)
{
if (rawFile.GetParentSpectrumNumber(ms3ScanNumber) == scanNumber)
{
quantitationMsnScan = rawFile[ms3ScanNumber] as MsnDataScan;
break;
}
ms3ScanNumber++;
}
if (quantitationMsnScan == null)
{
throw new ArgumentException("Cannot find a MS3 spectrum associated with spectrum number " + scanNumber);
}
}
Tolerance Tolerance = quantitationMsnScan.MzAnalyzer == MZAnalyzerType.IonTrap2D ? ItMassTolerance : FtMassTolerance;
bool isETD = quantitationMsnScan.DissociationType == DissociationType.ETD;
double injectionTime = quantitationMsnScan.InjectionTime;
var massSpectrum = quantitationMsnScan.MassSpectrum;
double noise = 0;
if (NoisebandCap)
{
// Noise is pretty constant over a small region, find the noise of the center of all isobaric tags
MassRange range = new MassRange(UsedTags.Keys[0], UsedTags.Keys[UsedTags.Count - 1]);
ThermoSpectrum thermoSpectrum = massSpectrum as ThermoSpectrum;
if (thermoSpectrum != null)
{
var peak = thermoSpectrum.GetClosestPeak(range);
if (peak != null)
{
noise = peak.Noise;
}
else
{
throw new ArgumentException("Spectrum (#" + quantitationMsnScan.SpectrumNumber + ") has no m/z peaks");
}
}
else
{
throw new ArgumentException("Spectrum (#" + quantitationMsnScan.SpectrumNumber+") , or they are low-resolution data without noise information");
}
//ThermoLabeledPeak peak = massSpectrum.GetClosestPeak(range) as ThermoLabeledPeak;
//if (peak != null)
//{
// noise = peak.Noise;
//}
//else
//{
// peak = massSpectrum as ThermoLabeledPeak;
// if (peak == null)
// {
// throw new ArgumentException("Either the spectrum (#" + quantitationMsnScan.SpectrumNumber+") has no m/z peaks, or they are low-resolution data without noise information");
// }
// noise = peak.Noise;
//}
}
//Dictionary<TagInformation, QuantPeak> peaks = new Dictionary<TagInformation, QuantPeak>();
QuantPeak[] peaks = new QuantPeak[largestQuantPeak+1];
// Read in the peak data
foreach (TagInformation tag in UsedTags.Values)
{
double tagMz = isETD
? tag.MassEtd
: tag.MassCAD;
var peak = massSpectrum.GetClosestPeak(Tolerance.GetRange(tagMz));
QuantPeak qPeak = new QuantPeak(tag, peak, injectionTime, quantitationMsnScan, noise, peak == null && NoisebandCap);
peaks[tag.UniqueTagNumber] = qPeak;
}
PurityCorrect(peaks, isDecoy);
PSM psm = new PSM(filenameID, scanNumber, peaks, purity);
quantFile.AddPSM(psm);
}
}
// Dispose of all raw files
foreach (ThermoRawFile rawFile in RawFiles.Values)
{
rawFile.Dispose();
}
OnUpdateLog("PSMs loaded " + quantFile.Psms.Count);
Log("PSMs Loaded:\t" + quantFile.Psms.Count );
yield return quantFile;
}
}
private IEnumerable <QuantFile> LoadFiles(IEnumerable <string> filePaths, bool ms3Quant = false)
{
MSDataFile.CacheScans = false;
//int largestQuantPeak = 0;
int i = 0;
foreach (TagInformation tag in UsedTags.Values)
{
tag.UniqueTagNumber = i++;
tag.TotalSignal = 0;
tag.NormalizedTotalSignal = 0;
}
int largestQuantPeak = i - 1;
//int largestQuantPeak = UsedTags.Values.Select(tag => tag.UniqueTagNumber).Concat(new[] {0}).Max();
foreach (string filePath in filePaths)
{
Log("Processing file:\t" + filePath);
OnUpdateLog("Processing File " + filePath + "...");
QuantFile quantFile = new QuantFile(filePath);
StreamReader basestreamReader = new StreamReader(filePath);
int oldProgress = -1;
using (CsvReader reader = new CsvReader(basestreamReader, true))
{
while (reader.ReadNextRecord()) // go through csv and raw file to extract the info we want
{
int scanNumber = int.Parse(reader["Spectrum number"]);
string filenameID = reader["Filename/id"];
string rawFileName = filenameID.Split('.')[0];
bool isDecoy = reader["DEFLINE"].StartsWith("DECOY_");
ThermoRawFile rawFile;
if (!RawFiles.TryGetValue(rawFileName, out rawFile))
{
throw new ArgumentException("Cannot find this raw file: " + rawFileName + ".raw");
}
if (!rawFile.IsOpen)
{
rawFile.Open();
}
int progress = (int)(100 * (double)basestreamReader.BaseStream.Position / basestreamReader.BaseStream.Length);
if (progress != oldProgress)
{
OnProgressUpdate(progress);
oldProgress = progress;
}
//// Set default fragmentation to CAD / HCD
//FragmentationMethod ScanFragMethod = filenameID.Contains(".ETD.")
// ? FragmentationMethod.ETD
// : FragmentationMethod.CAD;
//if (ScanFragMethod == FragmentationMethod.ETD)
//{
// ScanFragMethod = FragmentationMethod.CAD;
// scanNumber += ETDQuantPosition;
//}
// Get the scan object for the sequence ms2 scan
MsnDataScan quantitationMsnScan = rawFile[scanNumber] as MsnDataScan;
double purity = 1;
if (CalculatePurity)
{
double mz = quantitationMsnScan.PrecursorMz;
int charge = quantitationMsnScan.PrecursorCharge;
DoubleRange isolationRange = MzRange.FromDa(mz, PurityWindowInTh);
MSDataScan parentScan = rawFile[quantitationMsnScan.ParentScanNumber];
purity = DeterminePurity(parentScan, mz, charge, isolationRange);
}
if (quantitationMsnScan == null)
{
OnUpdateLog("Spectrum Number " + scanNumber + " is not a valid MS2 scan from: " + rawFile.FilePath + ". Skipping PSM...");
continue;
}
if (MS3Quant)
{
quantitationMsnScan = null;
// Look forward to find associated MS3 quant scan (based on parent scan number)
int ms3ScanNumber = scanNumber + 1;
while (ms3ScanNumber < rawFile.LastSpectrumNumber)
{
if (rawFile.GetParentSpectrumNumber(ms3ScanNumber) == scanNumber)
{
quantitationMsnScan = rawFile[ms3ScanNumber] as MsnDataScan;
break;
}
ms3ScanNumber++;
}
if (quantitationMsnScan == null)
{
OnUpdateLog("Cannot find a MS3 spectrum associated with spectrum number " + scanNumber + ". Skipping PSM...");
continue;
}
}
Tolerance Tolerance = quantitationMsnScan.MzAnalyzer == MZAnalyzerType.IonTrap2D ? ItMassTolerance : FtMassTolerance;
bool isETD = quantitationMsnScan.DissociationType == DissociationType.ETD;
double injectionTime = quantitationMsnScan.InjectionTime;
//var massSpectrum = quantitationMsnScan.MassSpectrum;
var thermoSpectrum = rawFile.GetLabeledSpectrum(quantitationMsnScan.SpectrumNumber);
double noise = 0;
if (NoisebandCap)
{
// Noise is pretty constant over a small region, find the noise of the center of all isobaric tags
MassRange range = new MassRange(UsedTags.Keys[0], UsedTags.Keys[UsedTags.Count - 1]);
if (thermoSpectrum != null)
{
var peak = thermoSpectrum.GetClosestPeak(range.Mean, 500);
if (peak != null)
{
noise = peak.Noise;
}
else
{
OnUpdateLog("Spectrum (#" + quantitationMsnScan.SpectrumNumber + ") has no m/z peaks. Skipping PSM...");
continue;
}
}
else
{
OnUpdateLog("Spectrum (#" + quantitationMsnScan.SpectrumNumber + ") is low-resolution data without noise information. Skipping PSM...");
continue;
}
}
//Dictionary<TagInformation, QuantPeak> peaks = new Dictionary<TagInformation, QuantPeak>();
QuantPeak[] peaks = new QuantPeak[largestQuantPeak + 1];
// Read in the peak data
foreach (TagInformation tag in UsedTags.Values)
{
double tagMz = isETD
? tag.MassEtd
: tag.MassCAD;
var peak = thermoSpectrum.GetClosestPeak(Tolerance.GetRange(tagMz));
QuantPeak qPeak = new QuantPeak(tag, peak, injectionTime, quantitationMsnScan, noise, peak == null && NoisebandCap);
peaks[tag.UniqueTagNumber] = qPeak;
}
PurityCorrect(peaks, isDecoy);
PSM psm = new PSM(filenameID, scanNumber, peaks, purity);
quantFile.AddPSM(psm);
}
}
// Dispose of all raw files
foreach (ThermoRawFile rawFile in RawFiles.Values)
{
rawFile.Dispose();
}
OnUpdateLog("PSMs loaded " + quantFile.Psms.Count);
Log("PSMs Loaded:\t" + quantFile.Psms.Count);
yield return(quantFile);
}
}