//method was originally written recursively, but large peptides result in stackoverflow exceptions
public static void MassMatch(string B, string Y, NeoPsm psm, int BIndex, int YIndex) //this is the workhorse of SpliceFragments
{
double experimentalMass = psm.expMass;
string bFrag = IonCrop(B, experimentalMass, BIndex, ProductType.B, false); //returns a B ion sequence that has a mass smaller than the experimental mass by cleaving C term AA
if (bFrag.Length < ionsUsedMassVer) //If the number of AA from the N-term peptide is less than desired amount, start over loop and remove a single aa from the C-term
{
return;
}
string yFrag = IonCrop(Y, experimentalMass, YIndex, ProductType.Y, false); //returns a Y ion sequence that has a mass smaller than the experimental mass by cleaving N term AA
if (yFrag.Length < ionsUsedMassVer) //If the number of AA from the C-term peptide is less than desired amount, end recursion.
{
return;
}
double theoreticalMass = NeoMassCalculator.MonoIsoptopicMass(bFrag) + NeoMassCalculator.MonoIsoptopicMass(yFrag) - NeoConstants.WATER_MONOISOTOPIC_MASS + fixedModMass; //water added once in b and once in y
if (NeoMassCalculator.IdenticalMasses(experimentalMass, theoreticalMass, NeoFindAmbiguity.precursorMassTolerancePpm)) //if match
{
string novelSeq = bFrag + yFrag;
if (!psm.candidates.Any(x => x.seq.Equals(novelSeq))) //if fusion sequence was not previously assigned to this psm
{
psm.candidates.Add(new FusionCandidate(novelSeq));
}
}
}
public static List <NeoPsm> ImportNeoPsms(string nFileName, string cFileName)
{
string[] nInput = File.ReadAllLines(nFileName);
string[] cInput = File.ReadAllLines(cFileName);
List <NeoPsm> psms = new List <NeoPsm>();
ParseHeader(nInput[0]);
List <InitialID> nAssignment = new List <InitialID>();
List <InitialID> cAssignment = new List <InitialID>();
for (int i = 1; i < nInput.Count(); i++)
{
string[] line = nInput[i].Split('\t').ToArray();
InitialID id = new InitialID(Convert.ToInt32(line[scanNumberIndex]), Convert.ToDouble(line[scanPrecursorMassIndex]), line[accessionIndex], line[fullIndex], line[matchedIonsIndex], line[scoreIndex]);
nAssignment.Add(id);
}
for (int i = 1; i < cInput.Count(); i++)
{
string[] line = cInput[i].Split('\t').ToArray();
InitialID id = new InitialID(Convert.ToInt32(line[scanNumberIndex]), Convert.ToDouble(line[scanPrecursorMassIndex]), line[accessionIndex], line[fullIndex], line[matchedIonsIndex], line[scoreIndex]);
cAssignment.Add(id);
}
//sort by scan number
List <InitialID> nAssignmentSorted = nAssignment.OrderBy(o => o.scanNumber).ToList();
List <InitialID> cAssignmentSorted = cAssignment.OrderBy(o => o.scanNumber).ToList();
//remove scans not found in both files
double maxCount = nAssignmentSorted.Count();
for (int i = 0; i < maxCount; i++)
{
if (i < cAssignmentSorted.Count())
{
if (nAssignmentSorted[i].scanNumber.Equals(cAssignmentSorted[i].scanNumber))
{
NeoPsm psm = new NeoPsm(nAssignmentSorted[i].scanNumber, nAssignmentSorted[i].expMass, nAssignmentSorted[i], cAssignmentSorted[i]);
psms.Add(psm);
continue;
}
else if (nAssignmentSorted[i].scanNumber < cAssignmentSorted[i].scanNumber) //no information was found for the b scan using y ions, so remove it
{
nAssignmentSorted.Remove(nAssignmentSorted[i]);
maxCount--;
i--;
}
else //no information was found for the y scan using b ions, so remove it
{
cAssignmentSorted.Remove(cAssignmentSorted[i]);
i--;
}
}
}
return(psms);
}
public static void AutoFill(NeoPsm psm)
{
double experimentalMass = psm.expMass;
for (int i = 0; i < 2; i++)
{
string ionSequence;
ProductType ion;
if (i == 0)
{
ionSequence = psm.nInfo.seq;
ion = ProductType.B;
}
else
{
ionSequence = psm.cInfo.seq;
ion = ProductType.Y;
}
int fragNumber = 0;
string ionFrag = ionSequence;
while (ionFrag.Length > 1)
{
if (ion == ProductType.B)
{
ionFrag = ionSequence.Substring(0, (ionSequence.Length - fragNumber));
if (ionFrag.Substring(ionSequence.Length - fragNumber - 1, 1) == "]") //if end of a PTM annotation
{
while (ionFrag.Substring(ionSequence.Length - fragNumber - 1, 1) != "[")
{
fragNumber++;
ionFrag = ionSequence.Substring(0, (ionSequence.Length - fragNumber));
}
fragNumber += 2; //removes "(" and the AA the PTM was attached to
ionFrag = ionSequence.Substring(0, (ionSequence.Length - fragNumber));
}
}
else //Ion==Y
{
ionFrag = ionSequence.Substring((0 + fragNumber), (ionSequence.Length - fragNumber));
if (ionFrag.Substring(0, 1) == "[") //if start of a PTM annotation
{
while (ionFrag.Substring(0, 1) != "]")
{
fragNumber++;
ionFrag = ionSequence.Substring((0 + fragNumber), (ionSequence.Length - fragNumber));
}
fragNumber++; //removes ")"
ionFrag = ionSequence.Substring((0 + fragNumber), (ionSequence.Length - fragNumber));
}
}
double theoreticalMass = NeoMassCalculator.MonoIsoptopicMass(ionFrag) - NeoConstants.WATER_MONOISOTOPIC_MASS;
if (theoreticalMass < experimentalMass)
{
double key = experimentalMass - theoreticalMass;
List <string> combinations = new List <string>();
double closestPeak = double.NaN;
var ipos = Array.BinarySearch(NeoFindAmbiguity.keys, key);
if (ipos < 0)
{
ipos = ~ipos;
}
if (ipos > 0)
{
var downIpos = ipos - 1;
// Try down
while (downIpos >= 0)
{
closestPeak = NeoFindAmbiguity.keys[downIpos];
if (NeoMassCalculator.IdenticalMasses(experimentalMass, closestPeak + theoreticalMass, NeoFindAmbiguity.precursorMassTolerancePpm))
{
foreach (string frag in NeoFindAmbiguity.massDict[closestPeak])
{
combinations.Add(frag);
}
}
else
{
break;
}
downIpos--;
}
}
if (ipos < NeoFindAmbiguity.keys.Length)
{
var upIpos = ipos;
// Try here and up
while (upIpos < NeoFindAmbiguity.keys.Length)
{
closestPeak = NeoFindAmbiguity.keys[upIpos];
if (NeoMassCalculator.IdenticalMasses(experimentalMass, closestPeak + theoreticalMass, NeoFindAmbiguity.precursorMassTolerancePpm))
{
foreach (string frag in NeoFindAmbiguity.massDict[closestPeak])
{
combinations.Add(frag);
}
}
else
{
break;
}
upIpos++;
}
}
if (combinations.Count != 0)
{
foreach (string s in combinations)
{
if (ion == ProductType.B)
{
psm.candidates.Add(new FusionCandidate(ionFrag + s));
}
else
{
psm.candidates.Add(new FusionCandidate(s + ionFrag));
}
}
break;
}
}
fragNumber++;
}
}
}