private int numDistrib = 10; // number of isotopic distribution to consider
public MonoisotopicSearcher(Averagine averagine, BrainCSharp brain)
{
this.averagine = averagine;
this.brain = brain;
}
public void ParallelRun(string path, string outputDir, AveragineType type, ChargerType chargerType)
{
string file = Path.GetFileNameWithoutExtension(path) + ".mgf";
string output = Path.Combine(outputDir, file);
ThermoRawSpectrumReader reader = new ThermoRawSpectrumReader();
LocalMaximaPicking picking = new LocalMaximaPicking(ms1PrcisionPPM);
reader.Init(path);
Dictionary <int, List <int> > scanGroup = new Dictionary <int, List <int> >();
int current = -1;
int start = reader.GetFirstScan();
int end = reader.GetLastScan();
for (int i = start; i < end; i++)
{
if (reader.GetMSnOrder(i) == 1)
{
current = i;
scanGroup[i] = new List <int>();
}
else if (reader.GetMSnOrder(i) == 2)
{
scanGroup[current].Add(i);
}
}
List <MS2Info> ms2Infos = new List <MS2Info>();
Parallel.ForEach(scanGroup, (scanPair) =>
{
if (scanPair.Value.Count > 0)
{
ISpectrum ms1 = reader.GetSpectrum(scanPair.Key);
foreach (int i in scanPair.Value)
{
double mz = reader.GetPrecursorMass(i, reader.GetMSnOrder(i));
List <IPeak> ms1Peaks = FilterPeaks(ms1.GetPeaks(), mz, searchRange);
if (ms1Peaks.Count() == 0)
{
continue;
}
// insert pseudo peaks for large gap
List <IPeak> peaks = new List <IPeak>();
double precision = 0.02;
double last = ms1Peaks.First().GetMZ();
foreach (IPeak peak in ms1Peaks)
{
if (peak.GetMZ() - last > precision)
{
peaks.Add(new GeneralPeak(last + precision / 2, 0));
peaks.Add(new GeneralPeak(peak.GetMZ() - precision / 2, 0));
}
peaks.Add(peak);
last = peak.GetMZ();
}
List <IPeak> majorPeaks = picking.Process(peaks);
ICharger charger = new Patterson();
if (chargerType == ChargerType.Fourier)
{
charger = new Fourier();
}
else if (chargerType == ChargerType.Combined)
{
charger = new PattersonFourierCombine();
}
int charge = charger.Charge(peaks, mz - searchRange, mz + searchRange);
// find evelope cluster
EnvelopeProcess envelope = new EnvelopeProcess();
var cluster = envelope.Cluster(majorPeaks, mz, charge);
if (cluster.Count == 0)
{
continue;
}
// find monopeak
Averagine averagine = new Averagine(type);
BrainCSharp braincs = new BrainCSharp();
MonoisotopicSearcher searcher = new MonoisotopicSearcher(averagine, braincs);
MonoisotopicScore result = searcher.Search(mz, charge, cluster);
double precursorMZ = result.GetMZ();
// write mgf
ISpectrum ms2 = reader.GetSpectrum(i);
IProcess processer = new WeightedAveraging(new LocalNeighborPicking());
ms2 = processer.Process(ms2);
MS2Info ms2Info = new MS2Info
{
PrecursorMZ = result.GetMZ(),
PrecursorCharge = charge,
Scan = ms2.GetScanNum(),
Retention = ms2.GetRetention(),
Peaks = ms2.GetPeaks()
};
lock (resultLock)
{
ms2Infos.Add(ms2Info);
}
}
}
readingProgress.Add(scanGroup.Count);
});
ms2Infos = ms2Infos.OrderBy(m => m.Scan).ToList();
using (FileStream ostrm = new FileStream(output, FileMode.OpenOrCreate, FileAccess.Write))
{
using (StreamWriter writer = new StreamWriter(ostrm))
{
foreach (MS2Info ms2 in ms2Infos)
{
WriteMGF(writer, path + ",SCANS=" + ms2.Scan.ToString() + ",PRECURSOR=" + ms2.PrecursorMZ, ms2.PrecursorMZ, ms2.PrecursorCharge,
ms2.Scan, ms2.Retention * 60, reader.GetActivation(ms2.Scan), ms2.Peaks);
writer.Flush();
}
}
}
// update progress
progress.Add();
}
void GenerateTasks()
{
if (Path.GetExtension(msPath) == ".mgf")
{
MGFSpectrumReader reader = new MGFSpectrumReader();
reader.Init(msPath);
Dictionary <int, MS2Spectrum> spectraData = reader.GetSpectrum();
foreach (int scan in spectraData.Keys)
{
MS2Spectrum spectrum = spectraData[scan];
SearchTask searchTask = new SearchTask(spectrum,
spectrum.PrecursorMZ(), spectrum.PrecursorCharge());
tasks.Enqueue(searchTask);
readingCounter.Add(spectraData.Count);
}
}
else
{
ISpectrumReader reader = new ThermoRawSpectrumReader();
LocalMaximaPicking picking = new LocalMaximaPicking();
IProcess process = new WeightedAveraging(new LocalNeighborPicking());
reader.Init(msPath);
int start = reader.GetFirstScan();
int end = reader.GetLastScan();
Dictionary <int, List <int> > scanGroup = new Dictionary <int, List <int> >();
int current = -1;
for (int i = start; i < end; i++)
{
if (reader.GetMSnOrder(i) == 1)
{
current = i;
scanGroup[i] = new List <int>();
}
else if (reader.GetMSnOrder(i) == 2)
{
scanGroup[current].Add(i);
}
}
Parallel.ForEach(scanGroup,
new ParallelOptions {
MaxDegreeOfParallelism = SearchingParameters.Access.ThreadNums
},
(scanPair) =>
{
if (scanPair.Value.Count > 0)
{
ISpectrum ms1 = reader.GetSpectrum(scanPair.Key);
foreach (int i in scanPair.Value)
{
double mz = reader.GetPrecursorMass(i, reader.GetMSnOrder(i));
List <IPeak> ms1Peaks = FilterPeaks(ms1.GetPeaks(), mz, searchRange);
if (ms1Peaks.Count() == 0)
{
continue;
}
ICharger charger = new Patterson();
int charge = charger.Charge(ms1Peaks, mz - searchRange, mz + searchRange);
// insert pseudo peaks for large gap
List <IPeak> peaks = new List <IPeak>();
double precision = 0.02;
double last = ms1Peaks.First().GetMZ();
foreach (IPeak peak in ms1Peaks)
{
if (peak.GetMZ() - last > precision)
{
peaks.Add(new GeneralPeak(last + precision / 2, 0));
peaks.Add(new GeneralPeak(peak.GetMZ() - precision / 2, 0));
}
peaks.Add(peak);
last = peak.GetMZ();
}
List <IPeak> majorPeaks = picking.Process(peaks);
// find evelope cluster
EnvelopeProcess envelope = new EnvelopeProcess();
var cluster = envelope.Cluster(majorPeaks, mz, charge);
if (cluster.Count == 0)
{
continue;
}
// find monopeak
Averagine averagine = new Averagine(AveragineType.GlycoPeptide);
BrainCSharp braincs = new BrainCSharp();
MonoisotopicSearcher searcher = new MonoisotopicSearcher(averagine, braincs);
MonoisotopicScore result = searcher.Search(mz, charge, cluster);
double precursorMZ = result.GetMZ();
// search
ISpectrum ms2 = reader.GetSpectrum(i);
ms2 = process.Process(ms2);
SearchTask searchTask = new SearchTask(ms2, precursorMZ, charge);
tasks.Enqueue(searchTask);
}
}
readingCounter.Add(scanGroup.Count);
});
}
}
public void Test1()
{
string path = @"C:\Users\Rui Zhang\Downloads\ZC_20171218_C16_R1.raw";
string fasta = @"C:\Users\Rui Zhang\Downloads\haptoglobin.fasta";
// peptides
IProteinReader proteinReader = new FastaReader();
List <IProtein> proteins = proteinReader.Read(fasta);
List <IProtein> decoyProteins = new List <IProtein>();
foreach (IProtein protein in proteins)
{
IProtein p = new BaseProtein();
p.SetSequence(Reverse(protein.Sequence()));
decoyProteins.Add(p);
}
List <Proteases> proteases = new List <Proteases>()
{
Proteases.Trypsin, Proteases.GluC
};
HashSet <string> peptides = new HashSet <string>();
ProteinDigest proteinDigest = new ProteinDigest(2, 5, proteases[0]);
foreach (IProtein protein in decoyProteins)
{
peptides.UnionWith(proteinDigest.Sequences(protein.Sequence(),
ProteinPTM.ContainsNGlycanSite));
}
for (int i = 1; i < proteases.Count; i++)
{
proteinDigest.SetProtease(proteases[i]);
List <string> peptidesList = peptides.ToList();
foreach (string seq in peptidesList)
{
peptides.UnionWith(proteinDigest.Sequences(seq,
ProteinPTM.ContainsNGlycanSite));
}
}
Assert.True(peptides.Contains("KDNLTYVGDGETR"));
// build glycan
GlycanBuilder glycanBuilder = new GlycanBuilder();
glycanBuilder.Build();
// search
List <SearchResult> searchResults = new List <SearchResult>();
ThermoRawSpectrumReader reader = new ThermoRawSpectrumReader();
LocalMaximaPicking picking = new LocalMaximaPicking();
IProcess process = new LocalNeighborPicking();
reader.Init(path);
double searchRange = 2;
ISpectrum ms1 = null;
List <IPeak> majorPeaks = new List <IPeak>();
ISearch <string> oneSearcher = new BucketSearch <string>(ToleranceBy.PPM, 10);
PrecursorMatch precursorMatcher = new PrecursorMatch(oneSearcher);
precursorMatcher.Init(peptides.ToList(), glycanBuilder.GlycanMaps());
ISearch <string> moreSearcher = new BucketSearch <string>(ToleranceBy.Dalton, 0.01);
SequenceSearch sequenceSearcher = new SequenceSearch(moreSearcher);
ISearch <int> extraSearcher = new BucketSearch <int>(ToleranceBy.Dalton, 0.01);
GlycanSearch glycanSearcher = new GlycanSearch(extraSearcher, glycanBuilder.GlycanMaps());
SearchAnalyzer searchAnalyzer = new SearchAnalyzer();
for (int i = reader.GetFirstScan(); i < reader.GetLastScan(); i++)
{
if (reader.GetMSnOrder(i) < 2)
{
ms1 = reader.GetSpectrum(i);
majorPeaks = picking.Process(ms1.GetPeaks());
}
else
{
double mz = reader.GetPrecursorMass(i, reader.GetMSnOrder(i));
if (ms1.GetPeaks()
.Where(p => p.GetMZ() > mz - searchRange && p.GetMZ() < mz + searchRange)
.Count() == 0)
{
continue;
}
Patterson charger = new Patterson();
int charge = charger.Charge(ms1.GetPeaks(), mz - searchRange, mz + searchRange);
// find evelope cluster
EnvelopeProcess envelope = new EnvelopeProcess();
var cluster = envelope.Cluster(majorPeaks, mz, charge);
if (cluster.Count == 0)
{
continue;
}
// find monopeak
Averagine averagine = new Averagine(AveragineType.GlycoPeptide);
BrainCSharp braincs = new BrainCSharp();
MonoisotopicSearcher searcher = new MonoisotopicSearcher(averagine, braincs);
MonoisotopicScore result = searcher.Search(mz, charge, cluster);
double precursorMZ = result.GetMZ();
// search
ISpectrum ms2 = reader.GetSpectrum(i);
ms2 = process.Process(ms2);
//precursor match
var pre_results = precursorMatcher.Match(precursorMZ, charge);
if (pre_results.Count == 0)
{
continue;
}
// spectrum search
var peptide_results = sequenceSearcher.Search(ms2.GetPeaks(), charge, pre_results);
if (peptide_results.Count == 0)
{
continue;
}
var glycan_results = glycanSearcher.Search(ms2.GetPeaks(), charge, pre_results);
if (glycan_results.Count == 0)
{
continue;
}
var temp_results = searchAnalyzer.Analyze(i, ms2.GetPeaks(), peptide_results, glycan_results);
break;
}
}
}
public static Spectrum GetMS2Spectrum(ref ThermoRawSpectrumReader reader,
int scan, AveragineType type, ChargerType chargerType, LocalMaximaPicking picking, IProcess process,
ISpectrum ms1)
{
// scan header
Spectrum spectrum = new Spectrum
{
id = "scan=" + scan.ToString()
};
double dLowMass = 0;
double dHighMass = 0;
double dTIC = 0;
double dBasePeakMass = 0;
double dBasePeakIntensity = 0;
reader.GetScanHeaderInfoForScanNum(scan, ref dLowMass, ref dHighMass,
ref dTIC, ref dBasePeakMass, ref dBasePeakIntensity);
SetScanHeader(spectrum, dLowMass, dHighMass, dTIC,
dBasePeakMass, dBasePeakIntensity);
// binary data
spectrum.binaryDataArrayList = new BinaryDataArrayList();
SetBinaryDataArrayHeader(spectrum.binaryDataArrayList);
spectrum.cvParam[0] = new Component.CVParam()
{
cvRef = "MS",
accession = "MS:1000511",
name = "ms level",
value = "2",
};
double mz = reader.GetPrecursorMass(scan, reader.GetMSnOrder(scan));
List <IPeak> ms1Peaks = FilterPeaks(ms1.GetPeaks(), mz, searchRange);
if (ms1Peaks.Count() == 0)
{
return(null);
}
// insert pseudo peaks for large gaps
List <IPeak> peaks = new List <IPeak>();
double precision = 0.02;
double last = ms1Peaks.First().GetMZ();
foreach (IPeak peak in ms1Peaks)
{
if (peak.GetMZ() - last > precision)
{
peaks.Add(new GeneralPeak(last + precision / 2, 0));
peaks.Add(new GeneralPeak(peak.GetMZ() - precision / 2, 0));
}
peaks.Add(peak);
last = peak.GetMZ();
}
List <IPeak> majorPeaks = picking.Process(peaks);
ICharger charger = new Patterson();
if (chargerType == ChargerType.Fourier)
{
charger = new Fourier();
}
else if (chargerType == ChargerType.Combined)
{
charger = new PattersonFourierCombine();
}
int charge = charger.Charge(peaks, mz - searchRange, mz + searchRange);
// find evelope cluster
EnvelopeProcess envelope = new EnvelopeProcess();
var cluster = envelope.Cluster(majorPeaks, mz, charge);
if (cluster.Count == 0)
{
return(null);
}
// find monopeak
Averagine averagine = new Averagine(type);
BrainCSharp braincs = new BrainCSharp();
MonoisotopicSearcher searcher = new MonoisotopicSearcher(averagine, braincs);
MonoisotopicScore result = searcher.Search(mz, charge, cluster);
// process spectrum
ISpectrum ms2 = reader.GetSpectrum(scan);
List <IPeak> ms2Peaks = process.Process(ms2).GetPeaks();
spectrum.binaryDataArrayList.binaryDataArray[0].binary =
ms2Peaks.SelectMany(p => BitConverter.GetBytes(p.GetMZ())).ToArray();
spectrum.binaryDataArrayList.binaryDataArray[1].binary =
ms2Peaks.SelectMany(p => BitConverter.GetBytes(p.GetIntensity())).ToArray();
spectrum.defaultArrayLength = ms2Peaks.Count.ToString();
spectrum.precursorList = new PrecursorList
{
count = "1",
precursor = new Precursor[1]
};
for (int i = 0; i < spectrum.precursorList.precursor.Length; i++)
{
spectrum.precursorList.precursor[i] = new Precursor();
}
spectrum.precursorList.precursor[0].selectedIonList = new SelectedIonList
{
count = "1",
selectedIon = new SelectedIon[1]
};
for (int i = 0; i < spectrum.precursorList.precursor[0].selectedIonList.selectedIon.Length; i++)
{
spectrum.precursorList.precursor[0].selectedIonList.selectedIon[i] = new SelectedIon();
}
spectrum.precursorList.precursor[0].selectedIonList.selectedIon[0].cvParam = new Component.CVParam[2];
spectrum.precursorList.precursor[0].selectedIonList.selectedIon[0].cvParam[0] = new Component.CVParam()
{
cvRef = "MS",
accession = "MS:1000744",
name = "selected ion m/z",
value = result.GetMZ().ToString(),
unitCvRef = "MS",
unitAccession = "MS:1000040",
unitName = "m/z"
};
spectrum.precursorList.precursor[0].selectedIonList.selectedIon[0].cvParam[1] = new Component.CVParam()
{
cvRef = "MS",
accession = "MS:1000041",
name = "charge state",
value = charge.ToString()
};
spectrum.precursorList.precursor[0].activation = new Activation
{
cvParam = new Component.CVParam[1]
};
spectrum.precursorList.precursor[0].activation.cvParam[0] =
ActivationCVParam(reader.GetActivation(scan));
spectrum.binaryDataArrayList.binaryDataArray[0].encodedLength =
Convert.ToBase64String(spectrum.binaryDataArrayList.binaryDataArray[0].binary).Length.ToString();
spectrum.binaryDataArrayList.binaryDataArray[1].encodedLength =
Convert.ToBase64String(spectrum.binaryDataArrayList.binaryDataArray[1].binary).Length.ToString();
return(spectrum);
}
