public virtual void Analyze()
{
if (product == null)
{
return;
}
product.MaxScore();
ISpectrum spectrum = product.GetSpectrum();
HashSet <string> seen = new HashSet <string>();
foreach (IGlycoPeptideScore g in product.GetResults())
{
// remove redudant
string structure = string.Join("_", g.GetGlycoPeptide().GetGlycan().GetStructure());
string seq = g.GetGlycoPeptide().GetPeptide().GetSequence();
if (!seen.Contains(structure + seq))
{
seen.Add(structure + seq);
}
else
{
continue;
}
writer.Write(spectrum.GetScanNum().ToString() + ", ");
writer.Write(seq + ", ");
writer.Write(structure + ", ");
writer.Write(g.GetScore().ToString() + ", ");
writer.Write((spectrum as ISpectrumMSn).GetParentMZ().ToString() + ", ");
writer.WriteLine((spectrum as ISpectrumMSn).GetParentCharge().ToString());
}
}
public void Analyze()
{
if (product == null)
{
return;
}
ISpectrum spectrum = product.GetSpectrum();
foreach (IAnnotatedPeak peak in product.GetResults())
{
HashSet <string> seen = new HashSet <string>();
foreach (IGlycoPeptide g in peak.GetFragmentations())
{
// remove redudant
string structure = string.Join("_", g.GetGlycan().GetStructure());
string seq = g.GetPeptide().GetSequence();
if (!seen.Contains(structure + seq))
{
seen.Add(structure + seq);
}
else
{
continue;
}
writer.WriteLine(spectrum.GetScanNum().ToString() + ", ");
writer.WriteLine(peak.GetMZ().ToString() + ", ");
writer.WriteLine(peak.GetIntensity().ToString() + ", ");
writer.WriteLine(structure + ", ");
writer.WriteLine(seq + ", ");
}
}
}
public void Add(ISpectrum spectrum, List <IScore> score)
{
int scan = spectrum.GetScanNum();
spectrumTable[scan] = spectrum;
scoreTable[scan] = score;
}
public List <GUI> FindGlucoseUnits(ISpectrum spectrum)
{
List <GUI> units = new List <GUI>();
List <IPeak> peaks = spectrum.GetPeaks();
int scan = spectrum.GetScanNum();
for (int i = 0; i < Glucose.Length; i++)
{
double mass = Glucose[i];
double bestIntensity = 0;
IPeak bestPeak = null;
List <double> mzCandidates = Calculator.To.ComputeMZ(mass, maxCharge);
foreach (double mz in mzCandidates)
{
int idx = BinarySearchPoints(peaks, mz);
if (idx > 0 && peaks[idx].GetIntensity() > bestIntensity)
{
bestIntensity = peaks[idx].GetIntensity();
bestPeak = peaks[idx];
}
}
if (bestPeak is null)
{
continue;
}
units.Add(new GUI(i + 2, scan, bestPeak));
}
return(units);
}
public List <IResult> Search(ISpectrum spectrum)
{
List <IResult> results = new();
ISpectrum spec = spectrumProcessor.Process(spectrum);
envelopeProcessor.Init(spec);
foreach (IGlycanPeak glycan in glycans)
{
IResult result = null;
double bestIntensity = 0;
double bestScore = 0;
for (int charge = 1; charge <= maxCharge; charge++)
{
List <double> mzList = Calculator.To.ComputeMZ(glycan.HighestPeak(), charge);
foreach (double mz in mzList)
{
List <IPeak> targets = envelopeProcessor.Search(mz);
if (targets.Count == 0)
{
continue;
}
SortedDictionary <int, List <IPeak> > clusters =
envelopeProcessor.Cluster(mz, charge);
IResult temp = monoisotopicSearcher.Match(glycan, clusters);
double score = temp.Score();
double intensity = temp.Matches().Select(m => m.GetIntensity()).Sum();
if (score > cutoff)
{
if (intensity > bestIntensity)
{
bestIntensity = intensity;
}
else if (intensity == bestIntensity && score > bestScore)
{
bestScore = score;
}
else
{
continue;
}
result = temp;
result.SetMZ(mz);
result.SetRetention(spectrum.GetRetention());
result.SetCharge(charge);
result.SetScan(spectrum.GetScanNum());
}
}
}
if (result != null)
{
results.Add(result);
}
}
return(results);
}
public void Report(string location)
{
foreach (IGlycoPeptideScore g in glycoPeptideScores)
{
Console.Write(spectrum.GetScanNum().ToString() + ", ");
Console.Write(g.GetGlycoPeptide().GetPeptide().GetSequence() + ", ");
Console.Write(string.Join("_", g.GetGlycoPeptide().GetGlycan().GetStructure()) + ", ");
Console.Write(g.GetScore().ToString() + ", ");
Console.Write((spectrum as ISpectrumMSn).GetParentMZ().ToString() + ", ");
Console.WriteLine((spectrum as ISpectrumMSn).GetParentCharge().ToString());
}
}
public void Report(string location)
{
Console.WriteLine(spectrum.GetScanNum());
foreach (IAnnotatedPeak peak in peaks)
{
Console.WriteLine("mz: " + peak.GetMZ().ToString());
Console.WriteLine("Intensity: " + peak.GetIntensity().ToString());
//Console.WriteLine(peak.GetFragmentations());
foreach (IGlycoPeptide g in peak.GetFragmentations())
{
Console.WriteLine(g.GetGlycan().GetName());
Console.WriteLine(g.GetPeptide().GetSequence());
}
Console.WriteLine();
}
Console.WriteLine();
}
public void ReportLines(ISpectrum spectrum, List <IScore> scores)
{
HashSet <string> seen = new HashSet <string>();
Tuple <List <double[]>, double[]> predict = Predicting(scores);
int idx = 0;
foreach (IScore score in scores)
{
if (0 >= 0)
{
// get glycan
IGlycoPeptide glycoPeptide = score.GetGlycoPeptide();
// remove redudant
string structure = string.Join("_", glycoPeptide.GetGlycan().GetStructure());
string seq = glycoPeptide.GetPeptide().GetSequence();
if (!seen.Contains(structure + seq))
{
seen.Add(structure + seq);
}
else
{
continue;
}
writer.Write(spectrum.GetScanNum().ToString() + ", ");
writer.Write(seq + ", ");
writer.Write(structure + ", ");
writer.Write(score.GetScore(MassType.Core).ToString() + ", ");
writer.Write(score.GetScore(MassType.Branch).ToString() + ", ");
writer.Write(score.GetScore(MassType.Glycan).ToString() + ", ");
writer.Write(score.GetScore(MassType.Peptide).ToString() + ", ");
writer.Write((spectrum as ISpectrumMSn).GetParentMZ().ToString() + ", ");
writer.Write((spectrum as ISpectrumMSn).GetParentCharge().ToString() + ", ");
writer.Write(string.Join(".", predict.Item1[idx]) + ", ");
writer.Write(predict.Item2[idx].ToString() + ", ");
writer.WriteLine();
}
idx++;
}
}
protected void ReportLines(ISpectrum spectrum, List <IScore> scores, double cutoff)
{
HashSet <string> seen = new HashSet <string>();
foreach (IScore score in scores)
{
if (score.GetScore() > cutoff)
{
// get glycan
IGlycoPeptide glycoPeptide = score.GetGlycoPeptide();
// remove redudant
string structure = string.Join("_", glycoPeptide.GetGlycan().GetStructure());
string seq = glycoPeptide.GetPeptide().GetSequence();
if (!seen.Contains(structure + seq))
{
seen.Add(structure + seq);
}
else
{
continue;
}
writer.Write(spectrum.GetScanNum().ToString() + ", ");
writer.Write(seq + ", ");
writer.Write(structure + ", ");
writer.Write(score.GetScore(MassType.Branch).ToString() + ", ");
writer.Write(score.GetScore(MassType.Core).ToString() + ", ");
writer.Write(score.GetScore(MassType.Glycan).ToString() + ", ");
writer.Write(score.GetScore(MassType.Peptide).ToString() + ", ");
writer.Write((spectrum as ISpectrumMSn).GetParentMZ().ToString() + ", ");
writer.Write((spectrum as ISpectrumMSn).GetParentCharge().ToString() + ", ");
if (score is IFDRScoreProxy)
{
writer.Write((score as IFDRScoreProxy).IsDecoy());
writer.Write(", ");
}
writer.WriteLine();
}
}
}
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 Search()
{
List <SearchResult> tempResults = new List <SearchResult>();
List <SearchResult> tempDecoyResults = new List <SearchResult>();
ISearch <string> oneSearcher = new BucketSearch <string>(
SearchingParameters.Access.MS1ToleranceBy, SearchingParameters.Access.MS1Tolerance);
PrecursorMatch precursorMatcher = new PrecursorMatch(oneSearcher);
precursorMatcher.Init(peptides, glycanBuilder.GlycanMaps());
ISearch <string> twoSearcher = new BucketSearch <string>(
SearchingParameters.Access.MS1ToleranceBy, SearchingParameters.Access.MS1Tolerance);
PrecursorMatch decoyPrecursorMatcher = new PrecursorMatch(twoSearcher);
decoyPrecursorMatcher.Init(decoyPeptides, glycanBuilder.GlycanMaps());
ISearch <string> moreSearcher = new BucketSearch <string>(
SearchingParameters.Access.MS2ToleranceBy, SearchingParameters.Access.MSMSTolerance);
SequenceSearch sequenceSearcher = new SequenceSearch(moreSearcher);
ISearch <int> extraSearcher = new BucketSearch <int>(
SearchingParameters.Access.MS2ToleranceBy, SearchingParameters.Access.MSMSTolerance);
GlycanSearch glycanSearcher = new GlycanSearch(extraSearcher, glycanBuilder.GlycanMaps(),
SearchingParameters.Access.ComplexInclude,
SearchingParameters.Access.HybridInclude,
SearchingParameters.Access.MannoseInclude);
SearchAnalyzer searchAnalyzer = new SearchAnalyzer();
SearchTask task;
while ((task = TryGetTask()) != null)
{
ISpectrum spectrum = task.Spectrum;
//precursor match
var pre_results = precursorMatcher.Match(task.PrecursorMZ, task.Charge);
if (pre_results.Count > 0)
{
// spectrum search
var peptide_results = sequenceSearcher.Search(spectrum.GetPeaks(), task.Charge, pre_results);
if (peptide_results.Count > 0)
{
var glycan_results = glycanSearcher.Search(spectrum.GetPeaks(), task.Charge, pre_results);
if (glycan_results.Count > 0)
{
var targets = searchAnalyzer.Analyze(spectrum.GetScanNum(), spectrum.GetPeaks(), peptide_results, glycan_results);
targets = searchAnalyzer.Filter(targets, glycanBuilder.GlycanMaps(),
task.PrecursorMZ, task.Charge);
tempResults.AddRange(targets);
}
}
}
var decoy_results = decoyPrecursorMatcher.Match(task.PrecursorMZ, task.Charge);
if (decoy_results.Count > 0)
{
// spectrum search
var peptide_results = sequenceSearcher.Search(spectrum.GetPeaks(), task.Charge, decoy_results);
if (peptide_results.Count > 0)
{
var glycan_results = glycanSearcher.Search(spectrum.GetPeaks(), task.Charge, decoy_results);
if (glycan_results.Count > 0)
{
var decoys = searchAnalyzer.Analyze(spectrum.GetScanNum(), spectrum.GetPeaks(), peptide_results, glycan_results);
decoys = searchAnalyzer.Filter(decoys, glycanBuilder.GlycanMaps(),
task.PrecursorMZ, task.Charge);
tempDecoyResults.AddRange(decoys);
}
}
}
searchCounter.Add(taskSize);
}
UpdateTask(tempResults, tempDecoyResults);
}
