//Do not use in parallel as the gappedItems sorting will screw up
public List <ScoredSequence> DetailedDenovoSequencing(PatternTools.MSParser.MSFull theMS)
{
MSFull myMS = PatternTools.ObjectCopier.Clone(theMS);
PrepareMSForDeNovo(myMS);
List <GraphNode> allNodes = GetNodes(myMS);
List <List <GraphNode> > allPaths = new List <List <GraphNode> >();
foreach (GraphNode g in allNodes)
{
//foreach (List<GraphNode> dp in g.DownPaths) {
// dp.Insert(0, g);
//}
allPaths.AddRange(g.DownPaths);
}
allPaths = allPaths.Distinct().ToList();
allPaths.Sort((a, b) => EvaluatePathForIntensity(b).CompareTo(EvaluatePathForIntensity(a)));
List <ScoredSequence> theResults = new List <ScoredSequence>();
foreach (List <GraphNode> p in allPaths)
{
double avgPPM = EvaluatePathForPPMError(p);
double intensity = EvaluatePathForIntensity(p);
int sequentialScore = p.Count;
string sequence = GetSequence(p);
ScoredSequence s = new ScoredSequence();
s.AvgPPMError = Math.Round(avgPPM, 1);
s.Sequence = sequence;
s.TotalIntensity = Math.Round(intensity, 0);
s.SequentialScore = sequentialScore;
theResults.Add(s);
}
int removed = theResults.RemoveAll(a => a.Sequence.Equals(""));
Console.WriteLine("Removing bad results: " + removed);
Dictionary <string, List <ScoredSequence> > cluster = (from result in theResults
group result by result.Sequence into resultGroup
select new { theRegex = resultGroup.Key, results = resultGroup }).ToDictionary(a => a.theRegex, a => a.results.ToList());
theResults.Clear();
foreach (KeyValuePair <string, List <ScoredSequence> > kvp in cluster)
{
kvp.Value.Sort((a, b) => b.TotalIntensity.CompareTo(a.TotalIntensity));
theResults.Add(kvp.Value[0]);
}
return(theResults);
}
public static string PrintSpectrumMGF(MSFull ms)
{
StringBuilder sw = new StringBuilder();
sw.AppendLine("BEGIN IONS");
sw.AppendLine("TITLE=Spectrum " + ms.ScanNumber + "; " + string.Join("; ", ms.Ilines));
sw.AppendLine("RITINSeconds=" + ms.CromatographyRetentionTime);
sw.AppendLine("CHARGE=" + string.Join(",", ms.Charges));
sw.AppendLine("PEPMASS=" + ms.ChargedPrecursor);
foreach (Ion i in ms.MSData)
{
sw.AppendLine(i.MZ + " " + i.Intensity);
}
return(sw.ToString());
}
public static string PrintSpectrumMS2(MSFull ms)
{
StringBuilder sw = new StringBuilder();
sw.Append("S\t" + ms.ScanNumber + "\t" + ms.ScanNumber + "\t" + ms.ChargedPrecursor + "\n");
sw.Append("I\tRetTime\t" + ms.CromatographyRetentionTime + "\n");
sw.Append("I\tIonInjectionTime\t" + ms.IonInjectionTime + "\n");
sw.Append("I\tActivationType\t" + ms.ActivationType + "\n");
sw.Append("I\tInstrumentType\t" + ms.InstrumentType + "\n");
foreach (string iline in ms.Ilines)
{
sw.Append("I\tAdditionalInformation:\t" + iline + "\n");
}
if (ms.isMS2)
{
if (ms.ZLines.Count > 0)
{
foreach (string zLine in ms.ZLines)
{
//Write the Z Lines
sw.Append(zLine + "\n");
}
}
else if (ms.Charges.Count > 0)
{
foreach (int charge in ms.Charges)
{
sw.Append("Z\t" + charge + "\t" + PatternTools.pTools.DechargeMSPeakToPlus1(ms.ChargedPrecursor, charge) + "\n");
}
}
}
foreach (Ion i in ms.MSData)
{
sw.Append(i.MZ + " " + i.Intensity + "\n");
}
return(sw.ToString());
}
//Optimized for finding most intense sequence
public double FastMaxSequentialScoreCoverage(PatternTools.MSParser.MSFull theMS)
{
MSFull myMS = PatternTools.ObjectCopier.Clone(theMS);
PrepareMSForDeNovo(myMS);
List <GraphNode> allNodes = GetNodes(myMS);
List <List <GraphNode> > allPaths = new List <List <GraphNode> >();
foreach (GraphNode g in allNodes)
{
allPaths.AddRange(g.DownPaths);
}
allPaths = allPaths.Distinct().ToList();
allPaths.Sort((a, b) => EvaluatePathForIntensity(b).CompareTo(EvaluatePathForIntensity(a)));
double coverage = allPaths[0].Max(a => a.MZ) - allPaths[0].Min(a => a.MZ);
return(coverage);
}
public List <MSFull> GetNext(int noSpectra)
{
List <MSFull> myScans = new List <MSFull>(noSpectra);
string thisLine = "";
MSFull thisScan;
if (cycleCounter == 0)
{
thisScan = new MSParser.MSFull();
}
else
{
thisScan = bufferScan;
}
while (true)
{
thisLine = sr.ReadLine();
if (thisLine == null)
{
break;
}
if (thisLine.Length == 0)
{
continue;
}
if (isNumber.Match(thisLine, 0, 1).Success)
{
//We are dealing with MS ion data
string[] ionData = mzSeparator.Split(thisLine);
Ion ion = new Ion(double.Parse(ionData[0]), double.Parse(ionData[1]), thisScan.CromatographyRetentionTime, thisScan.ScanNumber);
//Save our data
thisScan.MSData.Add(ion);
}
else if (thisLine.StartsWith("Z"))
{
string[] theStrings3 = tabSeparator.Split(thisLine);
if (thisScan.Charges == null)
{
thisScan.Charges = new List <int> {
int.Parse(theStrings3[1])
};
thisScan.Precursors = new List <double> {
double.Parse(theStrings3[2])
};
}
else
{
thisScan.Charges.Add(int.Parse(theStrings3[1]));
if (thisScan.Precursors == null)
{
thisScan.Precursors = new List <double>();
}
thisScan.Precursors.Add(double.Parse(theStrings3[2]));
}
}
else if (thisLine.Contains("ActivationType"))
{
string[] theStrings2 = tabSeparator.Split(thisLine);
thisScan.ActivationType = theStrings2[2];
}
else if (thisLine.Contains("InstrumentType"))
{
string[] theStrings2 = tabSeparator.Split(thisLine);
thisScan.InstrumentType = theStrings2[2];
}
else if (thisLine.Contains("RetTime"))
{
string[] theStrings2 = tabSeparator.Split(thisLine);
if (theStrings2.Length == 2)
{
//The retention time is separated with a space so we need to further break it up
string[] s = Regex.Split(theStrings2[1], @" ");
thisScan.CromatographyRetentionTime = double.Parse(s[1]);
}
else if (theStrings2.Length == 3)
{
thisScan.CromatographyRetentionTime = double.Parse(theStrings2[2]);
}
else
{
throw new Exception("Problems parsing Retention time Line.\n" + thisLine);
}
}
else if (thisLine.StartsWith("S"))
{
//Step 1:Save the Old One
if (thisScan.MSData.Count > 0) //Make sure we dont have an empty scan!
{
myScans.Add(thisScan);
}
//Step 2:Get the new one ready
thisScan = new MSParser.MSFull();
string[] theStrings = tabSeparator.Split(thisLine);
thisScan.ScanNumber = int.Parse(theStrings[1]);
thisScan.MSData = new List <Ion>();
if (theStrings.Length != 3)
{
//we are dealing with MS1 - the precursor shall be 0
//We are dealing with MS and should save precursor information
thisScan.ChargedPrecursor = double.Parse(theStrings[3]);
}
if (myScans.Count == noSpectra)
{
bufferScan = thisScan;
break;
}
}
}
if (thisLine == null)
{
sr.Close();
}
cycleCounter++;
return(myScans);
}
public static List <MSFull> ParseMGFFile(string file, string activationType = "CID", string instrumentType = "ITMS")
{
System.IO.StreamReader sr = new System.IO.StreamReader(file);
List <MSFull> theMS2 = new List <MSFull>();
string line = "";
MSFull ms = new MSFull();
while ((line = sr.ReadLine()) != null)
{
//take care of the header
if (line.StartsWith("#") || line.StartsWith("_") || line.Equals(""))
{
//This is a header line
//msFile.Header += line;
}
else if (line.Equals("BEGIN IONS"))
{
//Prepare Z lines
theMS2.Add(ms);
ms = new MSFull();
ms.isMS2 = true;
ms.ActivationType = activationType;
ms.InstrumentType = instrumentType;
}
else if (line.StartsWith("CHARGE"))
{
string[] cols = Regex.Split(line, "=");
string c = cols[1].Replace(@"+", "");
if (c.Equals("2,3"))
{
ms.Charges.Add(2);
ms.Charges.Add(3);
}
else
{
int charge = int.Parse(c);
ms.Charges.Add(charge);
}
}
else if (line.StartsWith("PEPMASS"))
{
string[] cols = Regex.Split(line, "=");
string[] nums = Regex.Split(cols[1], " ");
double chargedPrecursor;
if (nums[0].Equals(""))
{
chargedPrecursor = double.Parse(nums[1]);
}
else
{
chargedPrecursor = double.Parse(nums[0]);
}
ms.ChargedPrecursor = (chargedPrecursor + ((double)ms.Charges[0] * 1.007276466)) / (double)ms.Charges[0];
ms.ZLines.Add("Z\t" + ms.Charges[0] + "\t" + PatternTools.pTools.DechargeMSPeakToPlus1(ms.ChargedPrecursor, ms.Charges[0]));
}
else if (line.StartsWith("RTINSECONDS"))
{
string[] cols = Regex.Split(line, "=");
double retTime;
if (cols[1].Contains("-"))
{
string[] nums = Regex.Split(cols[1], "-");
retTime = double.Parse(nums[0]);
}
else
{
retTime = double.Parse(cols[1]);
}
ms.CromatographyRetentionTime = retTime;
}
else if (line.StartsWith("SCANS"))
{
string[] cols = Regex.Split(line, "=");
double scan;
if (cols[1].Contains("-"))
{
string[] nums = Regex.Split(cols[1], "-");
scan = double.Parse(nums[0]);
}
else
{
scan = double.Parse(cols[1]);
}
ms.ScanNumber = (int)scan;
}
else if (line.StartsWith("TITLE"))
{
ms.Ilines.Add(line);
if (line.Contains("Fragmentation:hcd"))
{
ms.ActivationType = "HCD";
ms.InstrumentType = "FTMS";
}
}
else if (Regex.IsMatch(line, "^[0-9]+"))
{
//If the line begins with a number it is an ion line
string[] cols = Regex.Split(line, "\t| ");
try
{
PatternTools.MSParser.Ion i = new PatternTools.MSParser.Ion(double.Parse(cols[0]), double.Parse(cols[1]), ms.CromatographyRetentionTime, ms.ScanNumber);
ms.MSData.Add(i);
}
catch
{
throw new Exception("An inconsistency has been found in file: " + file + "\nThe line reads:\n" + line);
}
}
}
theMS2.Add(ms);
//The first one is always bogus!
if (theMS2.Count > 0)
{
theMS2.RemoveAt(0);
}
sr.Close();
return(theMS2);
}
