public string cleanSeq(string seq, out string error_message)
{
error_message = "";
bool ModificationOn = false;
string ModificationName = "";
int aaIndex = 0;
string cleanedSeq = "";
foreach (char amino_acid in seq)
{
if (amino_acid == ')') //only occurs at end of mod
{
ModificationOn = false;
double modMass = MassCalculator.getPTMMass(ModificationName, out string e);
error_message += e;
PTM ptm = new PTM(ModificationName, aaIndex, modMass);
this.ptms.Add(ptm);
}
if (ModificationOn == true) //only occurs if "(" already found
{
ModificationName += amino_acid;
}
if (amino_acid == '(') //start collecting PTM name
{
ModificationOn = true;
}
if (ModificationOn == false && amino_acid != ')')
{
cleanedSeq += amino_acid;
}
}
return(cleanedSeq);
}
public static string cleanSeq(string fullSeq)
{
string temp = fullSeq.Replace("[Common Fixed:Carbamidomethyl of C]", "");
temp = temp.Replace("ptmlist:", "");
temp = MassCalculator.RemoveNestedParentheses(temp, false);
temp = temp.Replace("'", "");
temp = temp.Replace("\"", "");
return(temp);
}
//This method removes the number of amino acids specified by FragNumber from the respecitve terminus specified by ion of IonSequence
//If checkToRemoveExtraAA is true, additional AA will be removed to achieve a theoretical mass less than the Experimental mass
public string IonCrop(string IonSequence, double ExperimentalMass, int FragNumber, IonType ion, bool checkToRemoveExtraAA, out string error_message)
{
error_message = "";
string IonFrag;
if (ion == IonType.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++; //removes "("
FragNumber++; //removes 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));
}
}
if (checkToRemoveExtraAA == false)
{
return(IonFrag);
}
else
{
double IonMass = MassCalculator.MonoIsoptopicMass(IonFrag, out string e);
error_message += e;
if (IonMass < ExperimentalMass) //end if the mass of the fragment is lower than the experimental
{
return(IonFrag);
}
else //call the function again to remove another amino acid.
{
FragNumber++;
string x = IonCrop(IonSequence, ExperimentalMass, FragNumber, ion, true, out string e2);
error_message += e2;
return(x);
}
}
}
}; //20 common AA, ordered by mass assuming carbamido
public string PopulateMassDictionary()
{
string error_message = "";
List <double> tempKeys = new List <double>();
double maxMass = (maxMissingConsecutivePeaks + 1) * 186.079313 + (Constants.WATER_MONOISOTOPIC_MASS + 3); //If one sequence knocks it out of range, the next might not (VVVVW vs VVVVV where the first is out of range but the second is not) W-G prevents this //peaks+1 for converting missed peaks into number of ambiguous aa
int mer = (4) * (maxMissingConsecutivePeaks + 1); //maximum length allowed (4>W/G>3) //peaks+1 for converting missed peaks into number of ambiguous aa
//int mer = (2*(maxMissingConsecutivePeaks+1)); //only two aa should fit into one... work around to reduce computational demand
int[] indexes = new int[mer];
for (int i = 0; i < mer; i++)
{
indexes[i] = 0;
}
string seq = "";
int length = 1;
/* List<char> firstAA = new List<char>();
* foreach(char aa in AANames)
* {
* firstAA.Add(aa);
* }
* Parallel.ForEach(Partitioner.Creat(0,))*/
while (indexes[0] < AANames.Count())
{
//get new seq
seq = "";
if (indexes[2] == 15)
{
}
for (int n = 0; n < length; n++)
{
seq += AANames[indexes[n]];
}
//if new seq is within range
double fragMass = MassCalculator.MonoIsoptopicMass(seq, out string e);
error_message += e;
if (fragMass < maxMass)
{
var rounded = Math.Round(fragMass, decimalDigitsForFragmentMassRounding);
List <string> value;
if (massDict.TryGetValue(rounded, out value))
{
if (!value.Contains(seq))
{
value.Add(seq);
}
}
else
{
massDict.Add(rounded, new List <string> {
seq
});
tempKeys.Add(rounded);
}
if (mer != length && fragMass + 57.0214 < maxMass) //if not last position
{
length++; //allow m to increase
}
else //don't increment length, we're happy right now!
{
if (indexes[length - 1] < AANames.Count() - 1) //if not last aa in possible aa
{
indexes[length - 1]++;
}
else //if it is, we need to go back a bit
{
indexes[length - 1] = 0;
indexes[length - 2]++; //could cause crashing with weird aa
length--;
}
}
}
else //if not in range, move back one
{
indexes[length - 1] = 0;
indexes[length - 2]++; //could cause crashing with weird aa
length--;
}
//important catch to make sure going back doesn't result in an out of index exception
for (int i = indexes.Count() - 1; i >= 0; i--)
{
if (indexes[i] == AANames.Count())
{
if (i > 0)
{
length--;
indexes[i] = 0;
indexes[i - 1]++;
}
}
}
}
tempKeys.Sort();
keys = new double[tempKeys.Count()];
for (int k = 0; k < tempKeys.Count(); k++)
{
keys[k] = tempKeys[k];
}
using (System.IO.StreamWriter file = new System.IO.StreamWriter(@"C:\Users\Zach Rolfs\Desktop\Chemistry\Smith Research\Fusion Peptides\Dictionary" + "3" + ".txt"))
{
foreach (double key in keys)
{
List <string> value = new List <string>();
massDict.TryGetValue(key, out value);
bool satisfyRule = false;
foreach (string s in value)
{
if (s.Length <= maxMissingConsecutivePeaks + 1)
{
satisfyRule = true;
}
}
if (satisfyRule)
{
string output = key.ToString() + '\t';
foreach (string s in value)
{
output += s + ';';
}
output = output.Substring(0, output.Length - 1);
file.WriteLine(output);
}
}
}
//test
/* foreach(double key in keys)
* {
* List<string> value;
* if (massDict.TryGetValue(key, out value))
* {
* //MessageBox.Show(key.ToString() + " with " + value.Count());
* foreach(string v in value)
* {
* // MessageBox.Show(v);
* }
* }
* else
* {
* //MessageBox.Show("No sequences found for " + key);
* }
* }*/
return(error_message);
}
//compare the 6 first and last aa of each fusion candidate with database and determine if precursor mass can be achieved within 5 ppm. If it can, remove the psm from the list
public void removeTranslatedPeptides(List <PSM> psms, List <TheoreticalProtein> database, out string error_message)
{
error_message = "";
for (int i = 0; i < psms.Count(); i++)
{
this.worker.ReportProgress(Convert.ToInt16((Convert.ToDouble(i) / Convert.ToDouble(psms.Count())) * 100));
bool removed = false;
int fcIndex = 0;//FusionCandidate
while (!removed && fcIndex < psms[i].getFusionCandidates().Count())
{
FusionCandidate fc = psms[i].getFusionCandidates()[fcIndex];
string seq = fc.seq;
if (seq.Length >= ionsUsedDigFilter)
{
string Nterm = seq.Substring(0, ionsUsedDigFilter);
string Cterm = seq.Substring(seq.Length - ionsUsedDigFilter, ionsUsedDigFilter);
//N-TERMINUS SEARCHING
List <TheoreticalProtein> matches = database.AsParallel().Where(x => x.seq.Contains(Nterm)).ToList();
int protIndex = 0; //TheoreticalProtein
while (!removed && protIndex < matches.Count())
{
TheoreticalProtein prot = matches[protIndex];
double prodMass = 0;
//use below code to catch multiple appearances of a frag in a parent protein
List <int> indexes = new List <int>();
string subProt = prot.seq;
int pastIndex = 0;
while (subProt.Contains(Nterm))
{
int newIndex = subProt.IndexOf(Nterm);
indexes.Add(newIndex + pastIndex);
subProt = subProt.Substring(newIndex + 1, subProt.Length - newIndex - 1); //need to remove old match
pastIndex += newIndex + 1;
}
string protProd = "";
int sameProtIndex = 0;
while (!removed && sameProtIndex < indexes.Count())
{
Boolean hitEndOfProt = false;
int numAAused = ionsUsedDigFilter;
while (prodMass < psms[i].getExpMass() + 200 && hitEndOfProt == false)
{
try
{
// for (int PTMInclude = 0; PTMInclude < PTMIncludeMax; PTMInclude++) //loop twice, once without incorporating identified ptm masses and once with
{
//Obtain the Product mass
//if (Ion == 0) //if B main
{
protProd = prot.seq.Substring(indexes[sameProtIndex], numAAused);
prodMass = MassCalculator.MonoIsoptopicMass(protProd, out string e2);
error_message += e2;
//MessageBox.Show("1 "+ProtProd+" "+FASTARow[0].ToString() + " " + ProdMass.ToString());
/* if (PTMInclude == 1)
* {
* ProdMass += BPTMMass;
* }*/
}
//MissedCleavage and NonSpecific Cleavage/autolysis catch
if (generateDecoys)
{
if (((prodMass) > (psms[i].getExpMass() - 9.5) && (prodMass) < (psms[i].getExpMass() - 4.5)) | ((prodMass) > (psms[i].getExpMass() + 5.5) && (prodMass) < (psms[i].getExpMass() + 7.5))) //if match, add it!
{
psms.Remove(psms[i]);
i--;
removed = true;
}
}
else
{
if ((prodMass) > (psms[i].getExpMass() * (1 - precursorMassTolerancePpm / 1000000)) && (prodMass) < (psms[i].getExpMass() * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psms.Remove(psms[i]);
i--;
removed = true;
}
}
}
}
catch //sloppy patch used for when hitting the end of a protein
{
hitEndOfProt = true;
}
numAAused++;
}
sameProtIndex++;
}
protIndex++;
}
//C-TERMINUS SEARCHING
matches = database.AsParallel().Where(x => x.seq.Contains(Cterm)).ToList();
protIndex = 0; //TheoreticalProtein
while (!removed && protIndex < matches.Count())
{
TheoreticalProtein prot = matches[protIndex];
double prodMass = 0;
//use below code to catch multiple appearances of a frag in a parent protein
List <int> indexes = new List <int>();
string subProt = prot.seq;
int pastIndex = 0;
while (subProt.Contains(Cterm))
{
int newIndex = subProt.IndexOf(Cterm);
indexes.Add(newIndex + pastIndex);
subProt = subProt.Substring(newIndex + 1, subProt.Length - newIndex - 1); //need to remove old match
pastIndex += newIndex + 1;
}
string protProd = "";
int sameProtIndex = 0;
while (!removed && sameProtIndex < indexes.Count())
{
Boolean hitEndOfProt = false;
int numAAused = ionsUsedDigFilter;
while (prodMass < psms[i].getExpMass() + 200 && hitEndOfProt == false)
{
try
{
// for (int PTMInclude = 0; PTMInclude < PTMIncludeMax; PTMInclude++) //loop twice, once without incorporating identified ptm masses and once with
{
//Obtain the Product mass
//else //if y main
{
protProd = prot.seq.Substring(indexes[sameProtIndex] - numAAused + ionsUsedDigFilter, numAAused);
//MessageBox.Show(BProt);
prodMass = MassCalculator.MonoIsoptopicMass(protProd, out string e);
error_message += e;
//MessageBox.Show(Prot.IndexOf(bigFrag).ToString() + " " + AASearchLength + " "+bigFrag.Count().ToString());
//MessageBox.Show("2 "+bigFrag+" "+ProtProd + " " + ProdMass.ToString());
/* if (PTMInclude == 1)
* {
* ProdMass += YPTMMass;
* }*/
}
//MissedCleavage and NonSpecific Cleavage/autolysis catch
if (generateDecoys)
{
if (((prodMass) > (psms[i].getExpMass() - 9.5) && (prodMass) < (psms[i].getExpMass() - 4.5)) | ((prodMass) > (psms[i].getExpMass() + 5.5) && (prodMass) < (psms[i].getExpMass() + 7.5))) //if match, add it!
{
psms.Remove(psms[i]);
i--;
removed = true;
}
}
else
{
if ((prodMass) > (psms[i].getExpMass() * (1 - precursorMassTolerancePpm / 1000000)) && (prodMass) < (psms[i].getExpMass() * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psms.Remove(psms[i]);
i--;
removed = true;
}
}
}
}
catch //sloppy patch used for when hitting the end of a protein
{
hitEndOfProt = true;
}
numAAused++;
}
sameProtIndex++;
}
protIndex++;
}
}
fcIndex++;
}
}
}
public void FindCommonFalsePositives(List <PSM> psms, List <TheoreticalProtein> database, out string error_message)
{
error_message = "";
int i = 0;
foreach (PSM psm in psms)
{
this.worker.ReportProgress(Convert.ToInt16((Convert.ToDouble(i) / Convert.ToDouble(psms.Count())) * 100));
i++;
foreach (FusionCandidate fusionCandidate in psm.getFusionCandidates())
{
foreach (ParentInfo info in fusionCandidate.parentInfo)
{
if (info.fragFound.Length >= 6)
{
foreach (TheoreticalProtein protein in info.theoreticalProteins)
{
string protSeq = protein.seq;
char[] candidateSeq = fusionCandidate.seq.ToCharArray();
int index = protSeq.IndexOf(info.fragFound);
int fragLength = info.fragFound.Length;
string possibleTranslatedSequence = protSeq.Substring(index, fragLength);
if (!possibleTranslatedSequence.Equals(fusionCandidate.seq)) //if not already found as translated
{
double fragMass = MassCalculator.MonoIsoptopicMass(possibleTranslatedSequence, out string e);
error_message += e;
double expMass = psm.getExpMass();
while (fragMass < expMass + 187.079 - 57.021 + 1)
{
//Find SNPs
if (candidateSeq.Count() == possibleTranslatedSequence.Length)
{
char[] possibleSeqArray = possibleTranslatedSequence.ToCharArray();
if (IsSNP(candidateSeq, possibleSeqArray))
{
psm.variants.Add(new Variant(possibleTranslatedSequence, Variant.variantType.SNP, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
psm.variants.Add(new Variant(fusionCandidate.seq, Variant.variantType.SNP, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
}
if (generateDecoys)
{
//search for unmodified sequences
if (((fragMass) > (expMass - 9.5) && (fragMass) < (expMass - 4.5)) | ((fragMass) > (expMass + 5.5) && (fragMass) < (expMass + 7.5))) //if match, add it!
{
psm.variants.Add(new Variant(possibleTranslatedSequence, Variant.variantType.UM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
//fragMass, expMass
//search for PTMs
else if (((fragMass + .984016) > (expMass - 9.5) && (fragMass + .98402) < (expMass - 4.5)) | ((fragMass + .98402) > (expMass + 5.5) && (fragMass + .98402) < (expMass + 7.5))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Deamidation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if (((fragMass + 15.99491463) > (expMass - 9.5) && (fragMass + 15.99491463) < (expMass - 4.5)) | ((fragMass + 15.99491463) > (expMass + 5.5) && (fragMass + 15.99491463) < (expMass + 7.5))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Oxidation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if (((fragMass + 15.99491463 * 2) > (expMass - 9.5) && (fragMass + 15.99491463 * 2) < (expMass - 4.5)) | ((fragMass + 15.99491463 * 2) > (expMass + 5.5) && (fragMass + 15.99491463 * 2) < (expMass + 7.5))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+DiOxidation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if (((fragMass + 15.99491463 * 3) > (expMass - 9.5) && (fragMass + 15.99491463 * 3) < (expMass - 4.5)) | ((fragMass + 15.99491463 * 3) > (expMass + 5.5) && (fragMass + 15.99491463 * 3) < (expMass + 7.5))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+TriOxidation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
//else if ((fragMass + 14.01565) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 14.01565) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
//{
// psm.variants.Add(new Variant(Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
//} else if (((TheoreticalMass + .984016) > (ExperimentalMass - 9.5) && (TheoreticalMass + .98402) < (ExperimentalMass - 4.5)) | ((TheoreticalMass + .98402) > (ExperimentalMass + 5.5) && (TheoreticalMass) < (ExperimentalMass + 7.5))) //if match
else if (((fragMass + 42.01056) > (expMass - 9.5) && (fragMass + 42.01056) < (expMass - 4.5)) | ((fragMass + 42.01056) > (expMass + 5.5) && (fragMass + 42.01056) < (expMass + 7.5))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Acetylation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if (((fragMass + 79.96633) > (expMass - 9.5) && (fragMass + 79.96633) < (expMass - 4.5)) | ((fragMass + 79.96633) > (expMass + 5.5) && (fragMass + 79.96633) < (expMass + 7.5))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Phosphorylation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
}
else
{
//search for unmodified sequences
if ((fragMass) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match, add it!
{
psm.variants.Add(new Variant(possibleTranslatedSequence, Variant.variantType.UM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
//search for PTMs
else if ((fragMass + .984016) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + .98402) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Deamidation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 15.99491463) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 15.99491) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Oxidation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 15.99491463 * 2) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 15.99491 * 2) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Dioxidation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 15.99491463 * 3) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 15.99491 * 3) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Trioxidation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
//else if ((fragMass + 14.01565) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 14.01565) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
//{
// psm.variants.Add(new Variant(Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
//}
else if ((fragMass + 42.01056) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 42.01056) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Acetyl", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 79.96633) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 79.96633) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Phospho", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 61.913495) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 61.913495) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Zinc", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 37.955588) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 37.955588) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Potassium", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 21.981944) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 21.981944) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Sodium", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 57.021464) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 57.021464) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Carbamidomethyl", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 79.956815) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 79.956815) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Sulfonation", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 14.01565) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 14.01565) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Methyl", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 28.0313) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 28.0313) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+DiMethyl", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass + 42.04695) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass + 42.04695) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+TriMethyl", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
else if ((fragMass - -17.026549) > (expMass * (1 - precursorMassTolerancePpm / 1000000)) && (fragMass - -17.026549) < (expMass * (1 + precursorMassTolerancePpm / 1000000))) //if match
{
psm.variants.Add(new Variant(possibleTranslatedSequence + "+Ammonia loss", Variant.variantType.PTM, protein.id, protSeq, protSeq.IndexOf(info.fragFound), info.fragFound.Length));
}
}
//-17, 22,
//update the sequence
fragLength++;
if (info.parentType.Equals(ParentInfo.terminal.N))
{
if (index + fragLength < protSeq.Length)
{
possibleTranslatedSequence = protSeq.Substring(index, fragLength);
fragMass = MassCalculator.MonoIsoptopicMass(possibleTranslatedSequence, out string e2);
error_message += e2;
}
else
{
fragMass = expMass + 187.079 - 57.021 + 1;
}
}
else
{
index--;
if (index >= 0)
{
possibleTranslatedSequence = protSeq.Substring(index, fragLength);
fragMass = MassCalculator.MonoIsoptopicMass(possibleTranslatedSequence, out string e2);
error_message += e2;
}
else
{
fragMass = expMass + 187.079 - 57.021 + 1;
}
}
}
}
}
}
}
}
List <Variant> variants = psm.variants;
for (int v = variants.Count() - 1; v >= 0; v--)
{
for (int v2 = 0; v2 < v; v2++)
{
if (variants[v].pepSeq.Equals(variants[v2].pepSeq) && variants[v].id.Equals(variants[v2].id))
{
variants.Remove(variants[v]);
v2 = variants.Count();
}
}
}
}
}
//method was originally written recursively, but large peptides result in stackoverflow exceptions
public void MassMatch(string B, string Y, PSM psm, int BIndex, int YIndex, out string error_message) //this is the workhorse of SpliceFragments
{
error_message = "";
test = psm.getScan().ToString();
double ExperimentalMass = psm.getExpMass();
string BFrag = IonCrop(B, ExperimentalMass, BIndex, IonType.b, false, out string e4); //returns a B ion sequence that has a mass smaller than the experimental mass by cleaving C term AA
//BIndex = B.Length - BFrag.Length; //added 11/8/16 Useful first pass to record how many AA have been cleaved from C term
string YFrag = IonCrop(Y, ExperimentalMass, YIndex, IonType.y, false, out string e3); //returns a Y ion sequence that has a mass smaller than the experimental mass by cleaving N term AA
//YIndex = Y.Length - YFrag.Length; //added 11/8/16 Useful first pass to record how many AA have been cleaved from N term
double TheoreticalMass = MassCalculator.MonoIsoptopicMass(BFrag, out string e) + MassCalculator.MonoIsoptopicMass(YFrag, out string e2) - Constants.WATER_MONOISOTOPIC_MASS + fixedModMass; //water added once in b and once in y
error_message += e3 + e4 + e + e2;
//add PTM masses
foreach (PTM ptm in psm.getNInfo().getPTMs())
{
if (ptm.index < BFrag.Length)
{
TheoreticalMass += ptm.mass;
}
}
foreach (PTM ptm in psm.getCInfo().getPTMs())
{
if (Y.Length - ptm.index < YFrag.Length)
{
TheoreticalMass += ptm.mass;
}
}
if (YFrag.Length < ionsUsedMassVer) //If the number of AA from the C-term peptide is less than desired amount, end recursion.
{
//we're done
}
else 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
{
// MassMatch(B, Y, psm, 0, YIndex+1);
}
//if match
//bool elif = true; //"else if" where not a match==true
else if (FalsePositives.generateDecoys)
{
//else if (((TheoreticalMass - Constants.PEPTIDE_N_TERMINAL_MONOISOTOPIC_MASS * 7) > (ExperimentalMass - 1 * Constants.PEPTIDE_N_TERMINAL_MONOISOTOPIC_MASS) && (TheoreticalMass - Constants.PEPTIDE_N_TERMINAL_MONOISOTOPIC_MASS * 7) < (ExperimentalMass + 1 * Constants.PEPTIDE_N_TERMINAL_MONOISOTOPIC_MASS)) | ((TheoreticalMass + Constants.PEPTIDE_N_TERMINAL_MONOISOTOPIC_MASS * 7) > (ExperimentalMass - 1 * Constants.PEPTIDE_N_TERMINAL_MONOISOTOPIC_MASS) && (TheoreticalMass + Constants.PEPTIDE_N_TERMINAL_MONOISOTOPIC_MASS * 7) < (ExperimentalMass + 1 * Constants.PEPTIDE_N_TERMINAL_MONOISOTOPIC_MASS)))//if match //if ((TheoreticalMass) > (ExperimentalMass+ i -PrecursorMassToleranceDa) && (TheoreticalMass) < (ExperimentalMass+i +PrecursorMassToleranceDa)) //if match
if (((TheoreticalMass) > (ExperimentalMass - 9.5) && (TheoreticalMass) < (ExperimentalMass - 4.5)) | ((TheoreticalMass) > (ExperimentalMass + 5.5) && (TheoreticalMass) < (ExperimentalMass + 7.5)))//if match //if ((TheoreticalMass) > (ExperimentalMass+ i -PrecursorMassToleranceDa) && (TheoreticalMass) < (ExperimentalMass+i +PrecursorMassToleranceDa)) //if match //else if(Math.Abs(ExperimentalMass - TheoreticalMass)<40 && (ExperimentalMass - TheoreticalMass)-Math.Floor(ExperimentalMass-TheoreticalMass)>0.3 && (ExperimentalMass - TheoreticalMass) - Math.Floor(ExperimentalMass - TheoreticalMass) < 0.8)
{
// elif = false;
bool previouslyFound = false;
foreach (FusionCandidate oldCandidate in psm.getFusionCandidates())
{
if ((BFrag + YFrag).Equals(oldCandidate.seq)) //see if that sequence was already recorded
{
previouslyFound = true;
}
}
if (!previouslyFound) //if fusion sequence was not previously assigned to this psm
{
FusionCandidate candidate = new FusionCandidate(BFrag + YFrag);
psm.addFusionCandidate(candidate);
// MassMatch(B, Y, psm, BIndex + 1, YIndex);
}
}
}
else
{
if ((TheoreticalMass) > (ExperimentalMass * (1 - FalsePositives.precursorMassTolerancePpm / 1000000)) && (TheoreticalMass) < (ExperimentalMass * (1 + FalsePositives.precursorMassTolerancePpm / 1000000))) //if match //if ((TheoreticalMass) > (ExperimentalMass+ i -PrecursorMassToleranceDa) && (TheoreticalMass) < (ExperimentalMass+i +PrecursorMassToleranceDa)) //if match
{
// elif = false;
bool previouslyFound = false;
foreach (FusionCandidate oldCandidate in psm.getFusionCandidates())
{
if ((BFrag + YFrag).Equals(oldCandidate.seq)) //see if that sequence was already recorded
{
previouslyFound = true;
}
}
if (!previouslyFound) //if fusion sequence was not previously assigned to this psm
{
FusionCandidate candidate = new FusionCandidate(BFrag + YFrag);
psm.addFusionCandidate(candidate);
// MassMatch(B, Y, psm, BIndex + 1, YIndex);
}
}
}
// if(elif) //not a match
{
/* if (TheoreticalMass < ExperimentalMass && BIndex == 0) //first pass, theo less than exp and can't take away more ions
* {
* //we're done
* }
* else
* {
* if (TheoreticalMass < ExperimentalMass) //if b out of ions, but y not, crop off a y and start again
* {
* BIndex = 0;
* YIndex++;
* MassMatch(B,Y, psm, BIndex, YIndex);
* }
* else
* { //crop off a b ion
* MassMatch(B, Y, psm, BIndex + 1, YIndex);
* }
* }*/
}
}
public bool GeneratePossibleSequences(PSM psm, out string error_message) //returns false if over the specified number of sequences are generated
{
error_message = "";
List <string> foundSeq = new List <string>(); //get list of all FP sequences
foreach (FusionCandidate fusionCandidate in psm.getFusionCandidates())
{
findIons(fusionCandidate, psm, out string error_message1); //populate the foundIons array
error_message += error_message1;
foundSeq.Add(fusionCandidate.seq);
}
bool done = false;
int globalIndex = 0;
while (!done)
{
done = true; //let's assume we're done and correct it later if we're not
if (psm.getFusionCandidates().Count() > maxNumPossibleSequences) //if there are more than a set number of possible sequences, this is junk and we are not searching them all
{
return(false);
}
for (int fc = 0; fc < psm.getFusionCandidates().Count(); fc++)
{
FusionCandidate fusionCandidate = psm.getFusionCandidates()[fc];
if (fusionCandidate.getFoundIons().Count() > globalIndex) //prevent crashing, use to tell when done by hitting end of fc
{
List <FusionCandidate> tempCandidates = new List <FusionCandidate>(); //fill with possible sequences
done = false; //We're not done, because at least one fusion candidate sequence length is still greater than the global index
string fusionSeq = fusionCandidate.seq;
bool[] IonFound = fusionCandidate.getFoundIons();
if (IonFound[globalIndex]) //only look for ambiguity if a peak was found to provide the stop point.
{
int mostRecent = -1; //most recent Ion found prior to this one (start point)
for (int i = 0; i < globalIndex; i++) //identify start point
{
if (IonFound[i])
{
mostRecent = i; //save most recent hit, exclusive of the current index
}
}
string ambiguousFrag = fusionSeq.Substring(mostRecent + 1, globalIndex - mostRecent);
double key = MassCalculator.MonoIsoptopicMass(ambiguousFrag, out string error_message2);
error_message += error_message2;
List <string> combinations = new List <string>();
double closestPeak = double.NaN;
var ipos = Array.BinarySearch(keys, key);
if (ipos < 0)
{
ipos = ~ipos;
}
if (ipos > 0)
{
var downIpos = ipos - 1;
// Try down
while (downIpos >= 0)
{
closestPeak = keys[downIpos];
if (closestPeak > key - productMassToleranceDa && closestPeak < key + productMassToleranceDa)
{
string[] value;
if (massDict.TryGetValue(closestPeak, out value))
{
foreach (string frag in value)
{
combinations.Add(frag);
}
}
}
else
{
break;
}
downIpos--;
}
}
if (ipos < keys.Length)
{
var upIpos = ipos;
// Try here and up
while (upIpos < keys.Length)
{
closestPeak = keys[upIpos];
if (closestPeak > key - productMassToleranceDa && closestPeak < key + productMassToleranceDa)
{
string[] value;
if (massDict.TryGetValue(closestPeak, out value))
{
foreach (string frag in value)
{
combinations.Add(frag);
}
}
}
else
{
break;
}
upIpos++;
}
}
foreach (string str in combinations)
{
string nTermSeq = fusionSeq.Substring(0, mostRecent + 1);
string cTermSeq = fusionSeq.Substring(globalIndex + 1, fusionSeq.Length - globalIndex - 1);
string novelSeq = nTermSeq + str + cTermSeq;
FusionCandidate tempCandidate = new FusionCandidate(novelSeq);
tempCandidates.Add(tempCandidate);
}
}
foreach (FusionCandidate newfc in tempCandidates)
{
if (!foundSeq.Contains(newfc.seq)) //if new FP sequence, add it.
{
foundSeq.Add(newfc.seq);
findIons(newfc, psm, out string error_message3);
error_message += error_message3;
psm.getFusionCandidates().Add(newfc);
}
}
}
}
globalIndex++;
if (psm.getFusionCandidates().Count() > maxNumPossibleSequences)
{
return(false);
}
}
return(true);
}
//use ion hits to know where peaks have been found by morpheus and where there is ambiguity
public static void findIons(FusionCandidate fusionCandidate, PSM psm, out string error_message)
{
error_message = "";
double[] nPeaks = psm.getNInfo().getPeakHits(); //get peaks
double[] cPeaks = psm.getCInfo().getPeakHits();
fusionCandidate.makeFoundIons();
string candSeq = fusionCandidate.seq;
bool[] foundIons = fusionCandidate.getFoundIons();
//find which aa have peaks
for (int i = 0; i < foundIons.Count() - 1; i++)
{
//B IONS//
if (ionsUsed.Contains(IonType.b))
{
double bTheoMass = MassCalculator.MonoIsoptopicMass(candSeq.Substring(0, 1 + i), out string error_message2) - Constants.WATER_MONOISOTOPIC_MASS;
error_message += error_message2;
foreach (PTM ptm in psm.getNInfo().getPTMs())
{
if (ptm.index <= i)
{
bTheoMass += ptm.mass;
}
}
foreach (double expPeak in nPeaks)
{
if (expPeak > bTheoMass - productMassToleranceDa && expPeak < bTheoMass + productMassToleranceDa)
{
foundIons[i] = true;
}
}
}
//Y IONS//
if (ionsUsed.Contains(IonType.y))
{
double yTheoMass = MassCalculator.MonoIsoptopicMass(candSeq.Substring(candSeq.Length - 1 - i, i + 1), out string error_message3);
error_message += error_message3;
foreach (PTM ptm in psm.getCInfo().getPTMs())
{
if (ptm.index >= candSeq.Length - 2 - i)
{
yTheoMass += ptm.mass;
}
}
foreach (double expPeak in cPeaks)
{
if (expPeak > yTheoMass - productMassToleranceDa && expPeak < yTheoMass + productMassToleranceDa)
{
foundIons[foundIons.Count() - 2 - i] = true;
}
}
}
//C IONS//
if (ionsUsed.Contains(IonType.c))
{
double cTheoMass = MassCalculator.MonoIsoptopicMass(candSeq.Substring(0, 1 + i), out string error_message4) - Constants.WATER_MONOISOTOPIC_MASS + Constants.nitrogenMonoisotopicMass + 3 * Constants.hydrogenMonoisotopicMass;
error_message += error_message4;
foreach (PTM ptm in psm.getNInfo().getPTMs())
{
if (ptm.index <= i)
{
cTheoMass += ptm.mass;
}
}
foreach (double expPeak in nPeaks)
{
if (expPeak > cTheoMass - productMassToleranceDa && expPeak < cTheoMass + productMassToleranceDa)
{
foundIons[i] = true;
}
}
}
//ZDOT IONS//
if (ionsUsed.Contains(IonType.zdot))
{
double zdotTheoMass = MassCalculator.MonoIsoptopicMass(candSeq.Substring(candSeq.Length - 1 - i, i + 1), out string error_message5) - Constants.nitrogenMonoisotopicMass - 2 * Constants.hydrogenMonoisotopicMass;
error_message += error_message5;
foreach (PTM ptm in psm.getCInfo().getPTMs())
{
if (ptm.index >= candSeq.Length - 2 - i)
{
zdotTheoMass += ptm.mass;
}
}
foreach (double expPeak in cPeaks)
{
if (expPeak > zdotTheoMass - productMassToleranceDa && expPeak < zdotTheoMass + productMassToleranceDa)
{
foundIons[foundIons.Count() - 2 - i] = true;
}
}
}
}
//foundIons[0] = true; //AspN always starts with a D
foundIons[foundIons.Count() - 1] = true;//A|B|C|D|E|F|K| where the whole peptide peak is always placed arbitrarly at the c term
}