public IEnumerable <Tuple <Peptide, int> > DigestProteomeOnTheFlyFast(List <Protein> proteins, bool allowSNP, Queries AllQueries)
{
foreach (Peptide peptide in ProteinSearcher.ProteinDigest(options, proteins, allowSNP))
{
//int firstIndex = AllQueries.BinarySearch(MassTolerance.MzFloor(peptide.MonoisotopicMass, options.precursorMassTolerance));
//if (firstIndex >= 0 && firstIndex < AllQueries.Count)
// yield return new Tuple<Peptide, int>(peptide, firstIndex);
//foreach (Peptide peptide in ProteinSearcher.ProteinDigestNoEnzyme(dbOptions, proteins, AllQueries))
//if (!TargetPeptides.Contains(peptide.BaseSequence))
//{
foreach (Peptide modPeptide in peptide.GetVariablyModifiedPeptides(options.variableModifications, options.maximumVariableModificationIsoforms))
{
modPeptide.SetFixedModifications(options.fixedModifications);
int firstIndex = AllQueries.BinarySearch(MassTolerance.MzFloor(modPeptide.MonoisotopicMass, options.precursorMassTolerance));
if (firstIndex >= 0 && firstIndex < AllQueries.Count)
{
yield return(new Tuple <Peptide, int>(modPeptide, firstIndex));
}
}
//TODO check if this favors targets over decoys since proteins are sorted target->deco
// TargetPeptides.Add(peptide.BaseSequence);
//}
}
}
public IEnumerable <Tuple <Peptide, int> > DigestProteomeOnTheFly(List <Protein> proteins, bool allowSNP, Queries AllQueries)
{
//HashSet<string> TargetPeptides = new HashSet<string>();
//Dictionary<string, int> TargetPeptides = new Dictionary<string, int>();
//Digest proteins and store peptides in a Dictionnary
//Does not fit in memory -> 360 Go ....
//dicOfPeptideSequences = new Dictionary<string, List<Protein>>();
//double minimumMonoisotopicPeakOffset = dbOptions.precursorMonoisotopicPeakCorrection ? dbOptions.minimumPrecursorMonoisotopicPeakOffset : 0;
//double maximumMonoisotopicPeakOffset = dbOptions.precursorMonoisotopicPeakCorrection ? dbOptions.maximumPrecursorMonoisotopicPeakOffset : 0;
foreach (Peptide peptide in ProteinSearcher.ProteinDigest(options, proteins, allowSNP))
{
//int firstIndex = AllQueries.BinarySearch(MassTolerance.MzFloor(peptide.MonoisotopicMass, options.precursorMassTolerance));
//if (firstIndex >= 0 && firstIndex < AllQueries.Count)
// yield return new Tuple<Peptide, int>(peptide, firstIndex);
//foreach (Peptide peptide in ProteinSearcher.ProteinDigestNoEnzyme(dbOptions, proteins, AllQueries))
//if (!TargetPeptides.Contains(peptide.BaseSequence))
//{
foreach (Peptide modPeptide in peptide.GetVariablyModifiedPeptides(options.variableModifications, options.maximumVariableModificationIsoforms))
{
modPeptide.SetFixedModifications(options.fixedModifications);
int firstIndex = AllQueries.BinarySearch(MassTolerance.MzFloor(modPeptide.MonoisotopicMass, options.precursorMassTolerance));
if (firstIndex >= 0 && firstIndex < AllQueries.Count)
{
yield return(new Tuple <Peptide, int>(modPeptide, firstIndex));
}
}
//TODO check if this favors targets over decoys since proteins are sorted target->deco
// TargetPeptides.Add(peptide.BaseSequence);
//}
}
}
/// <summary>
/// Maps all peptide sequences to potential precursors and spectrum
/// </summary>
/// <param name="queries"></param>
/// <param name="fittingPeptides"></param>
/// <param name="previousProteins"></param>
/// <returns></returns>
public Precursors Search(Queries queries, IEnumerable <Tuple <Peptide, int> > fittingPeptides)
{
queries.dbOptions.ConSole.WriteLine("Mapping " + queries.Count + " queries to the digested proteome ... ");
long nbQueryConsidered = 0;
Parallel.ForEach <Tuple <Peptide, int> >(fittingPeptides, (Tuple <Peptide, int> hit) =>
//foreach (Tuple<Peptide, int> hit in fittingPeptides)
{
int indexPrecursor = hit.Item2;
double maximumMass = MassTolerance.MzTop(hit.Item1.MonoisotopicMass, options.precursorMassTolerance);
double minimumMass = MassTolerance.MzFloor(hit.Item1.MonoisotopicMass, options.precursorMassTolerance);
if (indexPrecursor < queries.Count && queries[indexPrecursor].precursor.Mass >= minimumMass)
{
while (indexPrecursor < queries.Count && queries[indexPrecursor].precursor.Mass <= maximumMass)
{
lock (queries[indexPrecursor].psms)
{
//if (low_index < Count && this[low_index].precursor.Mass >= minimum_precursor_mass)
// foreach (Query query in queries.GetQueryInMassRange(modified_peptide.MonoisotopicMass, options.precursorMassTolerance))
//{
//Target (or decoy with enzyme digests)
ComputePSMs(queries[indexPrecursor], hit.Item1);
//Decoy if NoEnzyme digest
if (options.DecoyFusion)
{
ComputePSMs(queries[indexPrecursor], hit.Item1.Reverse());
}
indexPrecursor++;
//foreach (Precursor isotope in query.precursor.Isotopes)
// ComputePSMs(query, modified_peptide, isotope.MassShift, previousProteins);
}
}
nbQueryConsidered += indexPrecursor - hit.Item2;
}
else
{
options.ConSole.WriteLine("WTF####");
}
});
//Push PSMs to Precursor
foreach (Query query in queries)
{
query.precursor.psms_AllPossibilities.AddRange(query.psms);//No MERGE
/*
* //Push PSMs to precursors
* if (query.precursor.psms_AllPossibilities.Count == 0)
* query.precursor.psms_AllPossibilities.AddRange(query.psms);
* else
* {
* //Merge common entries
* foreach (PeptideSpectrumMatch psmQuery in query.psms)
* {
* bool isNew = true;
* foreach (PeptideSpectrumMatch psmPrecursor in query.precursor.psms_AllPossibilities)
* {
* if (psmPrecursor != psmQuery && psmPrecursor.Peptide == psmQuery.Peptide)//Peptide object should be the same
* {
* psmPrecursor.Merge(psmQuery, options);
* isNew = false;
* break;
* }
* }
* if (isNew)
* query.precursor.psms_AllPossibilities.Add(psmQuery);
* }
* }//*/
}
//PeptideSpectrumMatches allPsms = new PeptideSpectrumMatches();
int nbAssignedPrecursor = 0;
foreach (Precursor precursor in queries.Precursors)
{
if (precursor.psms_AllPossibilities.Count > 0)
{
nbAssignedPrecursor++;
// allPsms.AddRange(precursor.psms);
}
}
//TODO Check the impact of this approach!!!!
//List<PeptideSpectrumMatch> okPSMs = allPsms.ComputeAtFDR(0.05);//TODO Add parameter for this value
//Dictionary<PeptideSpectrumMatch, int> dicOfPsm = new Dictionary<PeptideSpectrumMatch, int>();
//foreach(PeptideSpectrumMatch match in okPSMs)
// dicOfPsm.Add(match, 0);
//foreach (Precursor precursor in spectra.Precursors)
//{
// List<PeptideSpectrumMatch> newList = new List<PeptideSpectrumMatch>();
// foreach (PeptideSpectrumMatch psm in precursor.psms)
// if (dicOfPsm.ContainsKey(psm))
// newList.Add(psm);
// precursor.psms = newList;
//}
int nbAssignedQuery = 0;
foreach (Query query in queries)
{
if (query.psms.Count > 0)
{
nbAssignedQuery++;
}
}
options.ConSole.WriteLine(nbAssignedQuery + " queries matched [" + nbAssignedPrecursor + " precursors] out of " + nbQueryConsidered + " psm computed");
return(queries.Precursors);
}