public static List<string> IDList(string idpXML, string mzML)
{
List<string> list = new List<string>();
MSDataFile foo = new MSDataFile(mzML);
SpectrumList sl = foo.run.spectrumList;
IDPicker.Workspace workspace = new IDPicker.Workspace();
Package.loadWorkspace(ref workspace, idpXML);
foreach (IDPicker.SourceGroupList.MapPair groupItr in workspace.groups)
foreach (IDPicker.SourceInfo source in groupItr.Value.getSources(true))
foreach (IDPicker.SpectrumList.MapPair sItr in source.spectra)
{
IDPicker.ResultInstance ri = sItr.Value.results[1];
IDPicker.VariantInfo vi = ri.info.peptides.Min;
bool boolCharge = sItr.Value.id.charge.Equals(3);
if (boolCharge)
{
string rawPepSequence = vi.ToString();
string interpretation = vi.ToSimpleString();
// Look up the index with nativeID
object idOrIndex = null;
if (sItr.Value.nativeID != null && sItr.Value.nativeID.Length > 0)
{
idOrIndex = sItr.Value.nativeID;
list.Add(idOrIndex.ToString());
}
}//end if (boolcharge)
}//end foreach
return list;
}
public SpectraPeptides(string path)
{
string name = Path.GetFileNameWithoutExtension(path);
IDPicker.Workspace workspace = new IDPicker.Workspace();
Package.loadWorkspace(ref workspace, path);
foreach (IDPicker.SourceGroupList.MapPair groupItr in workspace.groups)
foreach (IDPicker.SourceInfo source in groupItr.Value.getSources(true))
foreach (IDPicker.SpectrumList.MapPair sItr in source.spectra)
{
IDPicker.ResultInstance ri = sItr.Value.results[1];
IDPicker.VariantInfo vi = ri.info.peptides.Min;
string rawPepSequence = vi.ToString();
string pepSequence = vi.peptide.sequence;
string index = sItr.Value.id.index.ToString();
string spectraID = name + "." + index;
peptideslist.Add(pepSequence);
spectralist.Add(spectraID);
}
peptideslist = Package.removeDuplicate(peptideslist);
spectralist = Package.removeDuplicate(spectralist);
peptides = peptideslist.Count;
spectra = spectralist.Count;
}
///<summary>
///original code for grabbing the required ions
///get pepSequence,source,scanID, write into a csv file.
///have tons of information: theretical ion intensity, pep info, chargeLabel...
///</summary>
public static List<string> idpReader_original(string idpXMLFile, string mzMLFile, double TicCutoffPercentage, int z, List<string> pepList, List<string> output)
{
//get the path and filename of output csv file:
string fileName = Path.GetFileNameWithoutExtension(idpXMLFile);
string filePath = Path.GetDirectoryName(idpXMLFile);
IDPicker.Workspace workspace = new IDPicker.Workspace();
Package.loadWorkspace(ref workspace, idpXMLFile);
MSDataFile foo = new MSDataFile(mzMLFile);
SpectrumList sl = foo.run.spectrumList;
foreach (IDPicker.SourceGroupList.MapPair groupItr in workspace.groups)
foreach (IDPicker.SourceInfo source in groupItr.Value.getSources(true))
foreach (IDPicker.SpectrumList.MapPair sItr in source.spectra)
{
IDPicker.ResultInstance ri = sItr.Value.results[1];
IDPicker.VariantInfo vi = ri.info.peptides.Min;
string ss = vi.ToString() + "," + sItr.Value.id.source.name + "," + sItr.Value.nativeID;
bool boolCharge = sItr.Value.id.charge.Equals(z);
if (boolCharge)
{
string rawPepSequence = vi.ToString();
string pepSequence = vi.peptide.sequence;
int len = pepSequence.Length;
//make sure that the peptide is what we want
if (pepList.Contains(pepSequence))
{
// Look up the index with nativeID
object idOrIndex = null;
if (sItr.Value.nativeID != null && sItr.Value.nativeID.Length > 0)
idOrIndex = sItr.Value.nativeID;
int spectrumIndex = sl.find(idOrIndex as string);
// Trust the local index, if the nativeID lookup fails
if (spectrumIndex >= sl.size())
spectrumIndex = sItr.Value.id.index;
// Bail of the loca index is larger than the spectrum list size
if (spectrumIndex >= sl.size())
throw new Exception("Can't find spectrum associated with the index.");
//Console.WriteLine(idOrIndex.ToString());
//get base peak and TIC and converted to string
Spectrum spec1 = sl.spectrum(spectrumIndex, true);
MZIntensityPairList peaks = new MZIntensityPairList();
spec1.getMZIntensityPairs(ref peaks);
Set<Peak> peakList = new Set<Peak>();
//get base peak and TIC
double basePeakValue = 0;
double TICValue = 0;
CVParamList list = spec1.cvParams;
foreach (CVParam CVP in list)
{
if (CVP.name == "base peak intensity")
{
basePeakValue = CVP.value;
}
if (CVP.name == "total ion current")
{
TICValue = CVP.value;
}
}
string basePeak = basePeakValue.ToString();
string TIC = TICValue.ToString();
//very important. Surendra put them here
//to change those with modifications {} into a format
//that fragment method will accept.
string interpretation = vi.ToSimpleString();
Peptide peptide = new Peptide(interpretation, ModificationParsing.ModificationParsing_Auto, ModificationDelimiter.ModificationDelimiter_Brackets);
Fragmentation fragmentation = peptide.fragmentation(true, true);
//prepare the qualified peaklist
double intenThreshhold = 0;
int totalIntenClass = 0;
double[] intenArray = new double[peaks.Count];
//used during the foreach loop
int indexPeaks = 0;
//get all the peaks no matter how small they are
//then calculate the threshhold TIC
//test here
foreach (MZIntensityPair mzIntensity in peaks)
{
//Peak p = new Peak(mzIntensity.mz, mzIntensity.intensity);
//peakList.Add(p);
intenArray[indexPeaks] = mzIntensity.intensity;
indexPeaks++;
}
Array.Sort(intenArray);
Array.Reverse(intenArray, 0, peaks.Count);
//if currTIC>=cutoff, then break
double currTIC = 0;
double cutOffTIC = TicCutoffPercentage * TICValue;
foreach (double inten in intenArray)
{
currTIC = currTIC + inten;
if (currTIC < cutOffTIC)
{
intenThreshhold = inten;
totalIntenClass++;
}
else break;
}
//then based on that, generate a new peaklist that contains only ABC peaks
//then calculate the intensity classes
foreach (MZIntensityPair mzIntensity in peaks)
{
if (mzIntensity.intensity >= intenThreshhold)
{
Peak p = new Peak(mzIntensity.mz, mzIntensity.intensity);
peakList.Add(p);
}
}
//rowDic contains row information of each peptide bond
Dictionary<int, string> rowDic = new Dictionary<int, string>();
//intensityDic contains intensity information of fragment ions for each peptide bond
Dictionary<int, List<double>> intensityDic = new Dictionary<int, List<double>>();
//commonList contains the common intensities
List<double> duplicateList = new List<double>();
//call the method
List<double> completeIntensityList = new List<double>();
for (int k = 1; k < len; k++)
{
List<double> intensityList = new List<double>();
string bion = pepSequence.Substring(0, k);
string yion = pepSequence.Substring(k, len - k);
int NR = Package.parseAAResidues(bion, 'R');
int NK = Package.parseAAResidues(bion, 'K');
int NH = Package.parseAAResidues(bion, 'H');
int NL = k;
int CR = Package.parseAAResidues(yion, 'R');
int CK = Package.parseAAResidues(yion, 'K');
int CH = Package.parseAAResidues(yion, 'H');
int CL = len - k;
int R = NR - CR;
int K = NK - CK;
int H = NH - CH;
int L = NL - CL;
int pepBond = k;
int NBasicAA = NR + NK + NH;
int CBasicAA = CR + CK + CH;
string AA = NBasicAA + "," + CBasicAA + "," + NR + "," + NK + "," + NH + "," + NL + "," + CR + "," + CK + "," + CH + "," + CL + "," + R + "," + K + "," + H + "," + L;
double[] bCharge = new double[z + 1];
double[] yCharge = new double[z + 1];
double[] bIntensity = new double[z + 1];
double[] yIntensity = new double[z + 1];
string bIonIntensity = "";
string yIonIntensity = "";
//to judge if the sum of intensities are 0
//so to exclude the case with all "0s"
double sumIntensity = 0;
//return the b ion charge 1 Intensity, if matched
for (int i = 1; i <= z; i++)
{
bCharge[i] = fragmentation.b(k, i);
yCharge[i] = fragmentation.y(len - k, i);
//change for Q-star purposes.
Peak bmatched = Package.findClose(peakList, bCharge[i], 70 * bCharge[i] * Math.Pow(10, -6));
Peak ymatched = Package.findClose(peakList, yCharge[i], 70 * yCharge[i] * Math.Pow(10, -6));
if (bmatched != null)
{
bIntensity[i] = bmatched.rankOrIntensity;
intensityList.Add(bmatched.rankOrIntensity);
completeIntensityList.Add(bmatched.rankOrIntensity);
}
else bIntensity[i] = 0;
if (ymatched != null)
{
yIntensity[i] = ymatched.rankOrIntensity;
intensityList.Add(ymatched.rankOrIntensity);
completeIntensityList.Add(ymatched.rankOrIntensity);
}
else yIntensity[i] = 0;
sumIntensity = sumIntensity + bIntensity[i] + yIntensity[i];
//record b/y ion intensity information into a string
bIonIntensity = bIonIntensity + "," + bIntensity[i];
yIonIntensity = yIonIntensity + "," + yIntensity[i];
}
intensityDic.Add(pepBond, intensityList);
//to determine charge label, need to split by precursor charge
//first need to make a metric to determine if all intensities are "0"
if (z == 3)
{
if (sumIntensity != 0)
{
////////////////////////////////////////////////
//set the ambiguity label as follows:
//-3: (0/+3) y3 only
//-2: (0/+3, +1/+2) y3, b1y2
//-1: (+1/+2): b1y2
//0: (+1/+2, +2/+1): b1y2, b2y1
//1: (+2/+1): b2y1
//2: (+2/+1, +3/0): b2y1, b3
//3: (+3/0): b3 only
////////////////////////////////////////////////
double b1 = bIntensity[1];
double b2 = bIntensity[2];
double y1 = yIntensity[1];
double y2 = yIntensity[2];
double b3 = bIntensity[3];
double y3 = yIntensity[3];
double b1y2 = b1 + y2;
double b2y1 = b2 + y1;
string ambiguityLabel = "";
//first part: set the intensity group: y3, b1y2, b2y1, b3
//if one group was found, set the label
//if two were found, but adjacent to each other, then ambiguity label is set
if (y3 != 0 && (b1y2 + b2y1 + b3) == 0) ambiguityLabel = "-3";
else if (y3 != 0 && b1y2 != 0 && (b2y1 + b3) == 0) ambiguityLabel = "-2";
else if (b1y2 != 0 && (y3 + b2y1 + b3) == 0) ambiguityLabel = "-1";
else if (b1y2 != 0 && b2y1 != 0 && (y3 + b3) == 0) ambiguityLabel = "0";
else if (b2y1 != 0 && (y3 + b1y2 + b3) == 0) ambiguityLabel = "1";
else if (b2y1 != 0 && b3 != 0 && (y3 + b1y2) == 0) ambiguityLabel = "2";
else if (b3 != 0 && (y3 + b1y2 + b2y1) == 0) ambiguityLabel = "3";
else ambiguityLabel = "error";
string finalString = idOrIndex + "," + pepSequence + "," + basePeak + "," + TIC + bIonIntensity + yIonIntensity + "," + len + "," + pepBond + "," + AA + "," + ambiguityLabel;
rowDic.Add(pepBond, finalString);
}
}
}//end for each peptide bond
//now we have: rowDic, intensityDic for each pep bond
//and we have: and completeIntensityList for each peptide
//the purpose of this is to remove such rows with duplicate matches
duplicateList = Package.findCommon(completeIntensityList);
foreach (int bond in rowDic.Keys)
{
bool unique = true;
foreach (double inten in duplicateList)
{
if (intensityDic[bond].Contains(inten))
{
unique = false;
Console.WriteLine("kick");
break;
}
}
if (unique)
{
output.Add(rowDic[bond]);
}
}
}//end of if peplist contains pepsequence
}//end if z==3
}//end foreach peptide
return output;
}
//the speccharge2score get each spectrum-peptide, and then map it to scores.
public static Dictionary<string, double> SpecCharge2Score(string idpxml, string qonversion, List<string> list)
{
///<summary>
///given qonversion and idpxml files, find the scores, DecoyState, making a dictionary.
///make it peptide-eccentric.
///</summary>
//get peptide sequence
string name = Path.GetFileNameWithoutExtension(idpxml);
Dictionary<string, string> dic_spectrum = new Dictionary<string, string>();
IDPicker.Workspace workspace = new IDPicker.Workspace();
Package.loadWorkspace(ref workspace, idpxml);
int repeatedIndex = 0;
foreach (IDPicker.SourceGroupList.MapPair groupItr in workspace.groups)
foreach (IDPicker.SourceInfo source in groupItr.Value.getSources(true))
foreach (IDPicker.SpectrumList.MapPair sItr in source.spectra)
{
IDPicker.ResultInstance ri = sItr.Value.results[1];
IDPicker.VariantInfo vi = ri.info.peptides.Min;
string rawPepSequence = vi.ToString();
string pepSequence = vi.peptide.sequence;
string index = sItr.Value.id.index.ToString();
int z = sItr.Value.id.charge;
string id = name + "." + index + "," + z.ToString();
if (!dic_spectrum.ContainsKey(id))
{
dic_spectrum.Add(id, pepSequence);
}
else repeatedIndex++;
}
//Console.WriteLine("repeated index (comes from spectrum with ambiguous charge state) is: " + repeatedIndex);
//then read qonverion.txt
//get index, TotalScore, DecoyState
//add pep and RNASeq_support information
//dic_peptide is what we wanna return
Dictionary<string, double> finaldic = new Dictionary<string, double>();
string t;
int unmatchedPep = 0;
TextReader file = new StreamReader(qonversion);
while ((t = file.ReadLine()) != null)
{
if (t.Contains("("))
{
Regex r = new Regex(" +");
string[] str = r.Split(t);
string Index = str[2];
double TotalScore = Convert.ToDouble(str[8]);
string charge = str[3];
string DecoyState = str[4];
string id = name + "." + Index + "," + charge;
if (dic_spectrum.ContainsKey(id))
{
string pep = dic_spectrum[id];
string key = id + "," + pep + "," + DecoyState;
finaldic.Add(key, TotalScore);
list.Add(key);
}
else unmatchedPep++;
}
}
//Console.WriteLine("unmatched peptide (peptide that was not found in the csv file) is: " + unmatchedPep);
Console.WriteLine("--one set xml-qonversion done with unmatched (unexpected): " + unmatchedPep);
return finaldic;
}
//given a assembly.xml file, grab peptide ids based on charge state.
//z=0, output all the peptides
public static List<string> PepSecurity(string xml, int z)
{
List<string> peptideList = new List<string>();
IDPicker.Workspace workspace = new IDPicker.Workspace();
Package.loadWorkspace(ref workspace, xml);
foreach (IDPicker.SourceGroupList.MapPair groupItr in workspace.groups)
foreach (IDPicker.SourceInfo source in groupItr.Value.getSources(true))
foreach (IDPicker.SpectrumList.MapPair sItr in source.spectra)
{
IDPicker.ResultInstance ri = sItr.Value.results[1];
IDPicker.VariantInfo vi = ri.info.peptides.Min;
string rawPepSequence = vi.ToString();
string pepSequence = vi.peptide.sequence;
<<<<<<< HEAD
if (z == 0)
{
if (sItr.Value.id.charge.Equals(2) || sItr.Value.id.charge.Equals(3) || sItr.Value.id.charge.Equals(4))
{
peptideList.Add(pepSequence);
}
}
//if (z == 0) peptideList.Add(pepSequence);
=======
//if (z == 0)
//{
// if (sItr.Value.id.charge.Equals(2) || sItr.Value.id.charge.Equals(3) || sItr.Value.id.charge.Equals(4))
// {
// peptideList.Add(pepSequence);
// }
//}
if (z == 0) peptideList.Add(pepSequence);
>>>>>>> 80fc9e47b4dafd28bcc96f478ae26543036b9ff6
else
{
if (sItr.Value.id.charge.Equals(z)) peptideList.Add(pepSequence);
}
}
List<string> uniPepList = Package.removeDuplicate(peptideList);
return uniPepList;
}
public static Dictionary<string, string> spectra2RNASeq(string idpxml, string qonversion, Dictionary<string, string> dic_RNASeq)
{
///<summary>
///now the problem is:
///d a idpQonvert file, read in all the peptides from pepxml without any fdr cutoff
///ead an idpxml file, read in all peptides that passed the fdr cutoff
///ow I restrict my rows to fdr-cutoff peptides, which might generate a limited pool
///ext is to consider fdr 0.1 cutoff, it will generate enough "0"s in the RNASeq_support column
///ifferent cutoff values here might have different merits. we'll find out which is the best
///</summary>
///next to check 1. whether 0.1 cutoff has the same idpqonvert file or not; Yes, the same idpqonvert file
///2. generate new idpxml with fdr cutoff=0.1 done.
///file path: X:\wangd5\idpXML_FDR1.00\myrimatch\FDR0.1
//then read idpxml file
//get peptide sequence
string name = Path.GetFileNameWithoutExtension(idpxml);
Dictionary<string, string> dic_peptide = new Dictionary<string, string>();
//dic_spectra is what we want to return
Dictionary<string, string> dic_spectra = new Dictionary<string, string>();
IDPicker.Workspace workspace = new IDPicker.Workspace();
Package.loadWorkspace(ref workspace, idpxml);
int repeatedIndex = 0;
foreach (IDPicker.SourceGroupList.MapPair groupItr in workspace.groups)
foreach (IDPicker.SourceInfo source in groupItr.Value.getSources(true))
foreach (IDPicker.SpectrumList.MapPair sItr in source.spectra)
{
IDPicker.ResultInstance ri = sItr.Value.results[1];
IDPicker.VariantInfo vi = ri.info.peptides.Min;
string rawPepSequence = vi.ToString();
string pepSequence = vi.peptide.sequence;
string index = sItr.Value.id.index.ToString();
int z = sItr.Value.id.charge;
string id = name + "." + index + "." + z.ToString() ;
if (!dic_peptide.ContainsKey(id))
{
bool Nspecificity = vi.peptide.NTerminusIsSpecific;
bool Cspecitificy = vi.peptide.CTerminusIsSpecific;
string value = pepSequence + "," + Nspecificity.ToString() + "," + Cspecitificy.ToString();
dic_peptide.Add(id, pepSequence);
dic_spectra.Add(id, value);
}
else repeatedIndex++;
}
Console.WriteLine("repeated index (comes from spectrum with ambiguous charge state) is: " + repeatedIndex);
//then read qonverion.txt
//get index, charges , TotalScore, DecoyState
//add pep and RNASeq_support information
string t;
int unmatchedPep = 0;
TextReader file = new StreamReader(qonversion);
while ((t = file.ReadLine()) != null)
{
if (t.Contains("("))
{
Regex r = new Regex(" +");
string[] str = r.Split(t);
string NativeID = str[1];
string Index = str[2];
string charge = str[3];
string DecoyState = str[4];
string TotalScore = str[8];
string FDR = str[9];
string id = name + "." + Index + "." + charge;
if (dic_peptide.ContainsKey(id))
{
string pep = dic_peptide[id];
if (dic_RNASeq.ContainsKey(pep))
{
string RNASeq = dic_RNASeq[pep];
dic_spectra[id] = NativeID + "," + dic_spectra[id] + "," + charge + "," + DecoyState + "," + TotalScore + "," + FDR + "," + RNASeq;
}
else unmatchedPep++;
}
}
}
Console.WriteLine("unmatched peptide (peptide that was not found in the csv file) is: " + unmatchedPep);
return dic_spectra;
}
//get pepSequence,source,scanID, write into a csv file.
//have tons of information: theretical ion intensity, pep info, chargeLabel...
public static void idpReader(string idpXMLFile, string mzMLFile, double TicCutoffPercentage, int z, int model)
{
//get the path and filename of output csv file:
string fileName = Path.GetFileNameWithoutExtension(idpXMLFile);
string filePath = Path.GetDirectoryName(idpXMLFile);
string csvFile = Path.Combine(filePath, fileName) + "_" + z.ToString() + ".csv";
IDPicker.Workspace workspace = new IDPicker.Workspace();
Package.loadWorkspace(ref workspace, idpXMLFile);
using (StreamWriter file = new StreamWriter(csvFile))
{
//TODO idOrIndex + "," + pepSequence + "," + pepBond + "," + AA + "," + bIons + "," + yIons;
file.WriteLine("nativeID,pepSequence,bond,b1,b2,y1,y2");
MSDataFile foo = new MSDataFile(mzMLFile);
SpectrumList sl = foo.run.spectrumList;
foreach (IDPicker.SourceGroupList.MapPair groupItr in workspace.groups)
foreach (IDPicker.SourceInfo source in groupItr.Value.getSources(true))
foreach (IDPicker.SpectrumList.MapPair sItr in source.spectra)
{
IDPicker.ResultInstance ri = sItr.Value.results[1];
IDPicker.VariantInfo vi = ri.info.peptides.Min;
string ss = vi.ToString() + "," + sItr.Value.id.source.name + "," + sItr.Value.nativeID;
bool boolCharge = sItr.Value.id.charge.Equals(z);
if (boolCharge)
{
string rawPepSequence = vi.ToString();
string pepSequence = vi.peptide.sequence;
int len = pepSequence.Length;
// Look up the index with nativeID
object idOrIndex = null;
if( sItr.Value.nativeID != null && sItr.Value.nativeID.Length > 0 )
idOrIndex = sItr.Value.nativeID;
int spectrumIndex = sl.find(idOrIndex as string);
// Trust the local index, if the nativeID lookup fails
if( spectrumIndex >= sl.size() )
spectrumIndex = sItr.Value.id.index;
// Bail of the loca index is larger than the spectrum list size
if( spectrumIndex >= sl.size() )
throw new Exception( "Can't find spectrum associated with the index." );
//get base peak and TIC and converted to string
Spectrum spec1 = sl.spectrum(spectrumIndex, true);
MZIntensityPairList peaks = new MZIntensityPairList();
spec1.getMZIntensityPairs(ref peaks);
Set<Peak> peakList = new Set<Peak>();
//get base peak and TIC
double basePeakValue = 0;
double TICValue = 0;
CVParamList list = spec1.cvParams;
foreach (CVParam CVP in list)
{
if (CVP.name == "base peak intensity")
{
basePeakValue = CVP.value;
}
if (CVP.name == "total ion current")
{
TICValue = CVP.value;
}
}
string basePeak = basePeakValue.ToString();
string TIC = TICValue.ToString();
//very important. Surendra put them here
//to change those with modifications {} into a format
//that fragment method will accept.
string interpretation = vi.ToSimpleString();
Peptide peptide = new Peptide(interpretation, ModificationParsing.ModificationParsing_Auto, ModificationDelimiter.ModificationDelimiter_Brackets);
Fragmentation fragmentation = peptide.fragmentation(true, true);
//prepare the qualified peaklist
double intenThreshhold = 0;
int totalIntenClass = 0;
double[] intenArray = new double[peaks.Count];
//used during the foreach loop
int indexPeaks = 0;
//get all the peaks no matter how small they are
//then calculate the threshhold TIC
//test here
foreach (MZIntensityPair mzIntensity in peaks)
{
//Peak p = new Peak(mzIntensity.mz, mzIntensity.intensity);
//peakList.Add(p);
intenArray[indexPeaks] = mzIntensity.intensity;
indexPeaks++;
}
Array.Sort(intenArray);
Array.Reverse(intenArray, 0, peaks.Count);
//if currTIC>=cutoff, then break
double currTIC = 0;
double cutOffTIC = TicCutoffPercentage * TICValue;
foreach (double inten in intenArray)
{
currTIC = currTIC + inten;
if (currTIC < cutOffTIC)
{
intenThreshhold = inten;
totalIntenClass++;
}
else break;
}
//then based on that, generate a new peaklist that contains only ABC peaks
//then calculate the intensity classes
foreach (MZIntensityPair mzIntensity in peaks)
{
if (mzIntensity.intensity >= intenThreshhold)
{
//note 0 here. This is to tell people that the orbiorbi fragment charge is unknown.
//the peaklist will be updated later.
Peak p = new Peak(mzIntensity.mz, mzIntensity.intensity,0);
peakList.Add(p);
}
}
Console.WriteLine("nativeID: =============" + idOrIndex);
//update peaklist for charge states.
peakList = Package.chargeAssignment(peakList);
//int ones = 0;
//int twos = 0;
//foreach (var peak in peakList)
//{
// //Console.WriteLine(peak.fragmentCharge);
// if (peak.fragmentCharge == 1) ones++;
// else if (peak.fragmentCharge == 2) twos++;
//}
//Console.WriteLine("charge 1: " + ones);
//Console.WriteLine("charge 2: " + twos);
//rowDic contains row information of each peptide bond
Dictionary<int, string> rowDic = new Dictionary<int, string>();
//intensityDic contains intensity information of fragment ions for each peptide bond
Dictionary<int, List<double>> intensityDic = new Dictionary<int, List<double>>();
//commonList contains the common intensities
List<double> duplicateList = new List<double>();
//call the method
List<double> completeIntensityList = new List<double>();
for (int k = 1; k < len; k++)
{
List<double> intensityList = new List<double>();
string bion = pepSequence.Substring(0, k);
string yion = pepSequence.Substring(k, len - k);
int NR = Package.parseAAResidues(bion, 'R');
int NK = Package.parseAAResidues(bion, 'K');
int NH = Package.parseAAResidues(bion, 'H');
int NL = k;
int CR = Package.parseAAResidues(yion, 'R');
int CK = Package.parseAAResidues(yion, 'K');
int CH = Package.parseAAResidues(yion, 'H');
int CL = len - k;
int pepBond = k;
int NBasicAA = NR + NK + NH;
int CBasicAA = CR + CK + CH;
string AA = NR + "," + NK + "," + NH + "," + NL + "," + CR + "," + CK + "," + CH + "," + CL;
double[] bCharge = new double[z + 1];
double[] yCharge = new double[z + 1];
//double[] bIntensity = new double[z + 1];
//double[] yIntensity = new double[z + 1];
//add variable for "real" charges
if (model == 0) //naive model
{
int[] bFragmentCharge = new int[z + 1];
int[] yFragmentCharge = new int[z + 1];
//return the b ion charge 1 Intensity, if matched
for (int i = 1; i < z; i++)
{
bCharge[i] = fragmentation.b(k, i);
yCharge[i] = fragmentation.y(len - k, i);
//change for ORBI-ORBI purposes.
Peak bmatched = Package.findClose(peakList, bCharge[i], bCharge[i] * 30 * Math.Pow(10, -6));
Peak ymatched = Package.findClose(peakList, yCharge[i], yCharge[i] * 30 * Math.Pow(10, -6));
if (bmatched != null)
{
//bIntensity[i] = bmatched.rankOrIntensity;
int fragmentCharge = bmatched.fragmentCharge;
if (fragmentCharge == i)
bFragmentCharge[i] = 3;
else bFragmentCharge[i] = 2;
//intensityList.Add(bmatched.rankOrIntensity);
//completeIntensityList.Add(bmatched.rankOrIntensity);
}
else bFragmentCharge[i] = 1;
//else bIntensity[i] = 0;
if (ymatched != null)
{
//yIntensity[i] = ymatched.rankOrIntensity;
//yFragmentCharge[i] = ymatched.fragmentCharge;
int fragmentCharge = ymatched.fragmentCharge;
if (fragmentCharge == i)
yFragmentCharge[i] = 3;
else yFragmentCharge[i] = 2;
//intensityList.Add(ymatched.rankOrIntensity);
//completeIntensityList.Add(ymatched.rankOrIntensity);
}
else yFragmentCharge[i] = 1;
//else yIntensity[i] = 0;
}
string finalString = idOrIndex + "," + pepSequence + "," + pepBond + "," + bFragmentCharge[1] + "," + bFragmentCharge[2] + "," + yFragmentCharge[1] + "," + yFragmentCharge[2];
file.WriteLine(finalString);
}
else if (model == 1) //my binary logistic regression basophile model
{
int b1=0, b2=0, y1=0, y2 = 0;
double y1_logit = 0.1098112 * NR + 0.2085831 * NK + 0.1512109 * NH + 0.0460839 * NL
- 0.3872417 * CR - 0.3684911 * CK - 0.1634741 * CH - 0.1693931 * CL + 1.2632997;
double y2_logit =-0.6345364 * NR - 0.3365917 * NK - 0.4577882 * NH - 0.1492703 * NL
+ 0.7738133 * CR + 0.6036758 * CK + 0.5942542 * CH + 0.0701467 * CL + 0.0806280;
double b1_logit = 0.0801432 * NR - 0.1088081 * NK - 0.1338220 * NH - 0.1413059 * NL
- 0.3157957 * CR - 0.2708274 * CK - 0.3703136 * CH + 0.0157418 * CL + 1.2124699;
double b2_logit = 0.8606449 * NR + 0.2763119 * NK + 0.4969152 * NH + 0.0685712 * NL
- 1.3346995 * CR - 1.0977316 * CK - 1.0973677 * CH - 0.2028884 * CL + 1.9355980;
if (b1_logit > -0.5)
{
double mz_b1 = fragmentation.b(k, 1);
Peak matched = Package.findClose(peakList, mz_b1, mz_b1 * 30 * Math.Pow(10, -6));
if (matched != null)
{
if (matched.fragmentCharge == 1) b1 = 3;
else b1 = 2;
}
else b1 = 1;
}
else b1 = 0;
if (b2_logit > 0)
{
double mz_b2 = fragmentation.b(k, 2);
Peak matched = Package.findClose(peakList, mz_b2, mz_b2 * 30 * Math.Pow(10, -6));
if (matched != null)
{
if (matched.fragmentCharge == 2) b2 = 3;
else b2 = 2;
}
else b2 = 1;
}
else b2 = 0;
if (y1_logit > -0.5)
{
double mz_y1 = fragmentation.y(len - k, 1);
Peak matched = Package.findClose(peakList, mz_y1, mz_y1 * 30 * Math.Pow(10, -6));
if (matched != null)
{
if (matched.fragmentCharge == 1) y1 = 3;
else y1 = 2;
}
else y1 = 1;
}
else y1 = 0;
if (y2_logit > -0.5)
{
double mz_y2 = fragmentation.y(len - k, 2);
Peak matched = Package.findClose(peakList, mz_y2, mz_y2 * 30 * Math.Pow(10, -6));
if (matched != null)
{
if (matched.fragmentCharge == 2) y2 = 3;
else y2 = 2;
}
else y2 = 1;
}
else y2 = 0;
string finalString = idOrIndex + "," + pepSequence + "," + pepBond + "," + b1 + "," + b2 + "," + y1 + "," + y2;
file.WriteLine(finalString);
}
else if (model == 2) //Surendra's ordinal model
{
int b1 = 0, b2 = 0, y1 = 0, y2 = 0;
double logit = NR * 0.9862 + NH * 0.8772 + NK * 0.7064 + NL * 0.4133
- CR * 1.1688 - CH * 0.3948 - CK * 0.6710 - CL * 0.4859;
//charge Label = "1", generate b+,y++
if (logit < -2.2502)
{
double mz_b1 = fragmentation.b(k, 1);
Peak b1matched = Package.findClose(peakList, mz_b1, mz_b1 * 30 * Math.Pow(10, -6));
if (b1matched != null)
{
if (b1matched.fragmentCharge == 1) b1 = 3; //matched, charge agree
else b1 = 2; //matched, charge does not agree
}
else b1 = 1; //unmatched, but predicted.
double mz_y2 = fragmentation.y(len - k, 2);
Peak y2matched = Package.findClose(peakList, mz_y2, mz_y2 * 30 * Math.Pow(10, -6));
if (y2matched != null)
{
if (y2matched.fragmentCharge == 2) y2 = 3; //matched, charge agree
else y2 = 2; //matched, charge does not agree
}
else y2 = 1; //unmatched, but predicted.
}
//ambiLabel = "2", generate b+, y+, b++, y++
else if (logit < 0.7872)
{
double mz_b1 = fragmentation.b(k, 1);
Peak b1matched = Package.findClose(peakList, mz_b1, mz_b1 * 30 * Math.Pow(10, -6));
if (b1matched != null)
{
if (b1matched.fragmentCharge == 1) b1 = 3; //matched, charge agree
else b1 = 2; //matched, charge does not agree
}
else b1 = 1; //unmatched, but predicted.
double mz_y1 = fragmentation.y(len - k, 1);
Peak y1matched = Package.findClose(peakList, mz_y1, mz_y1 * 30 * Math.Pow(10, -6));
if (y1matched != null)
{
if (y1matched.fragmentCharge == 1) y1 = 3; //matched, charge agree
else y1 = 2; //matched, charge does not agree
}
else y1 = 1; //unmatched, but predicted.
double mz_b2 = fragmentation.b(k, 2);
Peak b2matched = Package.findClose(peakList, mz_b2, mz_b2 * 30 * Math.Pow(10, -6));
if (b2matched != null)
{
if (b2matched.fragmentCharge == 2) b2 = 3; //matched, charge agree
else b2 = 2; //matched, charge does not agree
}
else b2 = 1; //unmatched, but predicted.
double mz_y2 = fragmentation.y(len - k, 2);
Peak y2matched = Package.findClose(peakList, mz_y2, mz_y2 * 30 * Math.Pow(10, -6));
if (y2matched != null)
{
if (y2matched.fragmentCharge == 2) y2 = 3; //matched, charge agree
else y2 = 2; //matched, charge does not agree
}
else y2 = 1; //unmatched, but predicted.
}
//ambiLabel = "3", generate b++,y+
else
{
double mz_b2 = fragmentation.b(k, 2);
Peak b2matched = Package.findClose(peakList, mz_b2, mz_b2 * 30 * Math.Pow(10, -6));
if (b2matched != null)
{
if (b2matched.fragmentCharge == 2) b2 = 3; //matched, charge agree
else b2 = 2; //matched, charge does not agree
}
else b2 = 1; //unmatched, but predicted.
double mz_y1 = fragmentation.y(len - k, 1);
Peak y1matched = Package.findClose(peakList, mz_y1, mz_y1 * 30 * Math.Pow(10, -6));
if (y1matched != null)
{
if (y1matched.fragmentCharge == 1) y1 = 3; //matched, charge agree
else y1 = 2; //matched, charge does not agree
}
else y1 = 1; //unmatched, but predicted.
}
string finalString = idOrIndex + "," + pepSequence + "," + pepBond + "," + b1 + "," + b2 + "," + y1 + "," + y2;
file.WriteLine(finalString);
}
//intensityDic.Add(pepBond, intensityList);
//rowDic.Add(pepBond, finalString);
}//end for each peptide bond
}//end if z==3
}//end foreach peptide
}//end using
}
static void Main(string[] args)
{
//Console.WriteLine("started");
//string naive = "Z:\\home\\dwang\\fragmentation\\UPS\\naive\\klc_031308p_cptac_study6_6_QC1.idpXML";
//string baso = "Z:\\home\\dwang\\fragmentation\\UPS\\basophilenew\\klc_031308p_cptac_study6_6_QC1.idpXML";
//for (int z=0; z<=4; z++)
//{
// List<string> pep_naive = Package.PepSecurity(naive, z);
// List<string> pep_baso = Package.PepSecurity(baso, z);
// List<string> common = Package.findCommon(pep_baso, pep_naive);
// Console.WriteLine("z==: " + z);
// Console.WriteLine("pep in naive: " + pep_naive.Count);
// Console.WriteLine("pep in baso: " + pep_baso.Count);
// Console.WriteLine("common: " + common.Count);
//}
///////////////////////////////////////////////////////////
//start myrimatch
///////////////////////////////////////////////////////////
Dictionary<string, string> peptideDic = new Dictionary<string, string>();
TextReader file_temp = new StreamReader("X:\\wangd5\\idpXML_FDR1.00\\score evaluation\\merge.csv");
DataTable table_temp = CSV.CsvParser.Parse(file_temp, true);
foreach (DataRow dr in table_temp.Rows)
{
string pep = dr[2].ToString();
string decoy = dr[3].ToString();
if (!peptideDic.ContainsKey(pep))
peptideDic.Add(pep, decoy);
}
string xml = "X:\\wangd5\\idpXML_FDR1.00\\myrimatch\\Assemble_MM.xml";
Dictionary<string,string> PSM = new Dictionary<string,string>();
Console.WriteLine("preparing reading idpXML");
int index = 0;
IDPicker.Workspace workspace = new IDPicker.Workspace();
Package.loadWorkspace(ref workspace, xml);
foreach (IDPicker.SourceGroupList.MapPair groupItr in workspace.groups)
foreach (IDPicker.SourceInfo source in groupItr.Value.getSources(true))
foreach (IDPicker.SpectrumList.MapPair sItr in source.spectra)
{
IDPicker.ResultInstance ri = sItr.Value.results[1];
IDPicker.VariantInfo vi = ri.info.peptides.Min;
string pepSequence = vi.peptide.sequence;
var scores = ri.searchScores.Values;
float[] scoreArr = new float[2];
scores.CopyTo(scoreArr, 0);
float mvh = scoreArr[0];
float mzfidelity = scoreArr[1];
string z = sItr.Value.id.charge.ToString();
string key = pepSequence + "." + index;
PSM.Add(key, mvh + "," + mzfidelity + "," + z);
index++;
}
Console.WriteLine("preparing reading RNA-seq");
Dictionary<string, string> dic_RNASeq = new Dictionary<string, string>();
TextReader file_csv = new StreamReader("X:\\wangd5\\idpXML_FDR1.00\\score evaluation\\fdr1.0.csv");
DataTable table = CSV.CsvParser.Parse(file_csv, true);
foreach (DataRow dr in table.Rows)
{
string pep = dr[0].ToString();
string RNASeq = dr[4].ToString();
dic_RNASeq.Add(pep, RNASeq);
}
List<string> finalList = new List<string>();
int unmatched = 0;
foreach (string key in PSM.Keys)
{
string pep = key.Split('.')[0];
if (dic_RNASeq.ContainsKey(pep) && peptideDic.ContainsKey(pep))
{
string rna = dic_RNASeq[pep];
finalList.Add(pep + "," + PSM[key] + "," + peptideDic[pep] + "," + rna);
}
else unmatched++;
}
Console.WriteLine("information: unmatched number is: " + unmatched);
Console.WriteLine("preparing writing files...");
string output = "X:\\wangd5\\idpXML_FDR1.00\\score evaluation\\myrimatchscorecombination.csv";
TextWriter file = new StreamWriter(output);
file.WriteLine("pep,mvh,mz,z,rna");
foreach (string ss in finalList)
{
file.WriteLine(ss);
}
file.Close();
/////////////////////////////////////////////////////////////////////////
//for 3 search engines
////////////////////////////////////////////////////////////////////////
/*
//===========================================================
string test_mm = "X:\\wangd5\\SW480\\MM\\FDR0.05\\mm.xml";
string test_sq = "X:\\wangd5\\SW480\\SQ\\FDR0.05\\sq.xml";
string test_xt = "X:\\wangd5\\SW480\\XT\\FDR0.05\\xt.xml";
string test_p = "X:\\wangd5\\SW480\\evaluation\\p.csv";
//string test_mm = "Z:\\home\\dwang\\fragmentation\\RNA-Seq\\RKO\\Assemble_MM.xml";
//string test_sq = "Z:\\home\\dwang\\fragmentation\\RNA-Seq\\RKO\\Assemble_SQ.xml";
//string test_xt = "Z:\\home\\dwang\\fragmentation\\RNA-Seq\\RKO\\Assemble_XT.xml";
//get peptides in p
List<string> p = new List<string>();
TextReader file_p = new StreamReader(test_p);
DataTable dt = CSV.CsvParser.Parse(file_p, true);
foreach (DataRow dr in dt.Rows)
{
string pep = dr[2].ToString();
p.Add(pep);
}
List<string> m = Package.PepSecurity(test_mm, 0);
List<string> x = Package.PepSecurity(test_xt, 0);
List<string> s = Package.PepSecurity(test_sq, 0);
m = Package.removeDuplicate(m);
x = Package.removeDuplicate(x);
s = Package.removeDuplicate(s);
p = Package.removeDuplicate(p);
List<string> pm = Package.findCommon(p, m);
List<string> mx = Package.findCommon(m, x);
List<string> ms = Package.findCommon(m, s);
List<string> xs = Package.findCommon(x, s);
List<string> mxs = Package.findCommon(mx, s);
///////////////////////////////////////////////////////////////////////////////////////
//read the fdr1.0.csv, to get all the peptide sequences that identified by 3 engines\
//no matter how low the score is
//also, the file contains the RNASeq information
Dictionary<string, string> dic_RNASeq = new Dictionary<string, string>();
TextReader file_csv = new StreamReader("X:\\wangd5\\SW480\\evaluation\\PepRNA_FDR1.csv");
//TextReader file_csv = new StreamReader("X:\\wangd5\\idpXML_FDR1.00\\score evaluation\\fdr1.0.csvn");
//TextReader file_csv = new StreamReader("X:\\wangd5\\idpXML_FDR1.00\\score evaluation\\fdr1.0.csv");
DataTable table = CSV.CsvParser.Parse(file_csv, true);
foreach (DataRow dr in table.Rows)
{
string pep = dr[0].ToString();
string RNASeq = dr[1].ToString();
dic_RNASeq.Add(pep, RNASeq);
}
int m_n = 0;
int x_n = 0;
int s_n = 0;
int p_n = 0;
int ms_n = 0;
int mx_n = 0;
int xs_n = 0;
int mxs_n = 0;
int pm_n = 0;
foreach (string ss in m)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") m_n++;
}
else Console.WriteLine("crap");
}
foreach (string ss in x)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") x_n++;
}
}
foreach (string ss in s)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") s_n++;
}
}
foreach (string ss in p)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") p_n++;
}
}
foreach (string ss in ms)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") ms_n++;
}
}
foreach (string ss in mx)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") mx_n++;
}
}
foreach (string ss in xs)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") xs_n++;
}
}
foreach (string ss in mxs)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") mxs_n++;
}
}
foreach (string ss in pm)
{
if (dic_RNASeq.ContainsKey(ss))
{
if (dic_RNASeq[ss] == "1") pm_n++;
}
}
Console.WriteLine("m: " + m.Count + "=" + m_n);
Console.WriteLine("x: " + x.Count + "=" + x_n);
Console.WriteLine("s: " + s.Count + "=" + s_n);
Console.WriteLine("p: " + p.Count + "=" + p_n);
Console.WriteLine("mx: " + mx.Count + "=" + mx_n);
Console.WriteLine("ms: " + ms.Count + "=" + ms_n);
Console.WriteLine("xs: " + xs.Count + "=" + xs_n);
Console.WriteLine("mxs: " + mxs.Count + "=" + mxs_n);
Console.WriteLine("pm: " + pm.Count + "=" + pm_n);
string path_mm = "X:\\wangd5\\SW480\\MM\\FDR1.00\\mam_012808n_SW480_200ug_";
string path_sq = "X:\\wangd5\\SW480\\SQ\\FDR1.00\\mam_012808n_SW480_200ug_";
string path_xt = "X:\\wangd5\\SW480\\XT\\FDR1.00\\mam_012808n_SW480_200ug_";
List<string> keyslist = new List<string>();
Dictionary<string, double> dic_mm = Package.RKO(path_mm, keyslist);
Dictionary<string, double> dic_sq = Package.RKO(path_sq, keyslist);
Dictionary<string, double> dic_xt = Package.RKO(path_xt, keyslist);
keyslist = Package.removeDuplicate(keyslist);
//merge the three dictionaries into one.
Dictionary<string, string> dic_merge = new Dictionary<string,string>();
int misses = 0;
foreach (string key in keyslist)
{
string mm = "";
string sq = "";
string xt = "";
string[] str = key.Split(',');
string pep = str[2];
if (dic_RNASeq.ContainsKey(pep))
{
string RNA = dic_RNASeq[pep];
if (dic_mm.ContainsKey(key))
{
mm = dic_mm[key].ToString();
}
else mm = "0";
if (dic_xt.ContainsKey(key))
{
xt = dic_xt[key].ToString();
}
else xt = "0";
if (dic_sq.ContainsKey(key))
{
sq = dic_sq[key].ToString();
}
else sq = "0";
dic_merge.Add(key, mm + "," + xt + "," + sq + "," + RNA);
}
else misses++;
}
Console.WriteLine("the number of pep-rna misses is: " + misses);
//write into file
string output = "X:\\wangd5\\SW480\\evaluation\\merge.csv";
TextWriter file = new StreamWriter(output);
file.WriteLine("spectrum,charge,pep,decoystate,mm,xt,sq,rna");
foreach (string key in dic_merge.Keys)
{
file.WriteLine(key + "," + dic_merge[key]);
}
}
*/
}
