public IcBottomUpScores GetScores(char pre, string sequence, char post, AminoAcid nTerm, AminoAcid cTerm, Composition composition, int charge, int ms2ScanNum)
{
ScoredSpectrum scoredSpectrum;
var index = GetChargetScanNumPairIndex(charge, ms2ScanNum);
if (!_scoredSpectra.TryGetValue(index, out scoredSpectrum))
{
var spec = Run.GetSpectrum(ms2ScanNum) as ProductSpectrum;
if (spec == null) return null;
scoredSpectrum = new ScoredSpectrum(spec, _rankScorer, charge, composition.Mass, Tolerance);
_scoredSpectra.Add(index, scoredSpectrum);
}
var seqGraph = SequenceGraph.CreateGraph(AminoAcidSet, nTerm, sequence, cTerm);
if (seqGraph == null)
{
return null;
}
Tuple<double, string> scoreAndModifications = null;
var bestScore = double.NegativeInfinity;
var protCompositions = seqGraph.GetSequenceCompositions();
for (var modIndex = 0; modIndex < protCompositions.Length; modIndex++)
{
seqGraph.SetSink(modIndex);
var protCompositionWithH2O = seqGraph.GetSinkSequenceCompositionWithH2O();
if (!protCompositionWithH2O.Equals(composition)) continue;
var curScoreAndModifications = seqGraph.GetFragmentScoreAndModifications(scoredSpectrum);
var curScore = curScoreAndModifications.Item1;
if (curScore > bestScore)
{
scoreAndModifications = curScoreAndModifications;
bestScore = curScore;
}
}
if (scoreAndModifications == null) return null;
var ms2Score = scoreAndModifications.Item1;
// TODO: This assumes enzyme is trypsin
const double probN = 0.99999;
const double probC = 0.99999;
const double sumAAProbabilities = 0.1;
var creditN = Math.Log(probN / sumAAProbabilities);
var penaltyN = Math.Log((1.0 - probN) / (1.0 - sumAAProbabilities));
var creditC = Math.Log(probC / sumAAProbabilities);
var penaltyC = Math.Log((1.0 - probC) / (1.0 - sumAAProbabilities));
if (pre == 'K' || pre == 'R' || pre == FastaDatabase.Delimiter || pre == '-') ms2Score += creditN;
else ms2Score += penaltyN;
var lastResidue = sequence[sequence.Length - 1];
if (lastResidue == 'K' || lastResidue == 'R' || post == FastaDatabase.Delimiter || post == '-') ms2Score += creditC;
else ms2Score += penaltyC;
var modifications = scoreAndModifications.Item2;
return new IcBottomUpScores(ms2Score, modifications);
}
public ShiftedSequenceGraph(AminoAcidSet aminoAcidSet, double shiftedMass, bool isForward, int maxSequenceLength, double maxSequenceMass = 50000.0)
{
_aminoAcidSet = aminoAcidSet;
_modificationParams = aminoAcidSet.GetModificationParams();
_isForward = isForward;
_index = 0;
_maxSeqIndex = maxSequenceLength + 2; // shift + Term + length
_maxSequenceMass = maxSequenceMass;
_aminoAcidSequence = new AminoAcid[_maxSeqIndex];
var shiftAa = new AminoAcid('\0', "Shift", new CompositionWithDeltaMass(shiftedMass));
_aminoAcidSequence[0] = shiftAa;
ShiftMass = shiftedMass;
_fragmentComposition = new Composition.Composition[_maxSeqIndex];
_fragmentComposition[0] = shiftAa.Composition;
_graph = new Node[_maxSeqIndex][];
_graph[0] = new[] { new Node(0) };
_nodeComposition = new Composition.Composition[_maxSeqIndex][];
_compNodeComposition = new Composition.Composition[_maxSeqIndex][];
for (var i = 0; i < _maxSeqIndex; i++)
{
_compNodeComposition[i] = new Composition.Composition[_modificationParams.NumModificationCombinations];
_nodeComposition[i] = new Composition.Composition[_modificationParams.NumModificationCombinations];
}
IsValid = true;
}
public SequenceTagGenerator(LcMsRun run, Tolerance tolerance, int minTagLength = 5, int maxTagLength = 8,
AminoAcid[] aminoAcidsArray = null)
{
_run = run;
_tolerance = tolerance;
_minTagLen = minTagLength;
_maxTagLen = maxTagLength;
_aminoAcids = aminoAcidsArray ?? AminoAcid.StandardAminoAcidArr;
_ms2ScanToTagMap = new Dictionary<int, IList<SequenceTag>>();
}
public ModifiedAminoAcid(AminoAcid aa, Modification modification)
: base(aa.Residue, aa.Name + "+" + modification.Name, aa.Composition + modification.Composition)
{
var modAa = aa as ModifiedAminoAcid;
if (modAa == null)
{
_modification = modification; // aa is not modified
}
else // aa is already modified
{
_modification = Modification.RegisterAndGetModification(modAa.Modification.Name + "+" + modification.Name, Composition);
}
}
/// <summary>
/// Generate a sequence graph for the provided data
/// </summary>
/// <param name="aminoAcidSet"></param>
/// <param name="nTerm"></param>
/// <param name="sequence"></param>
/// <param name="cTerm"></param>
protected SequenceGraph(AminoAcidSet aminoAcidSet, AminoAcid nTerm, string sequence, AminoAcid cTerm)
{
_aminoAcidSet = aminoAcidSet;
_sequence = sequence;
_nTerm = nTerm;
_modificationParams = aminoAcidSet.GetModificationParams();
_maxSeqIndex = sequence.Length + 3; // init + C-term + sequence length + N-term
_index = 0;
_aminoAcidSequence = new AminoAcid[_maxSeqIndex];
_aminoAcidSequence[0] = AminoAcid.Empty;
_suffixComposition = new Composition.Composition[_maxSeqIndex];
_suffixComposition[0] = Composition.Composition.Zero;
_graph = new Node[_maxSeqIndex][];
_graph[0] = new[] { new Node(0) };
_nodeComposition = new Composition.Composition[_maxSeqIndex][]; //, _modificationParams.NumModificationCombinations];
_compNodeComposition = new Composition.Composition[_maxSeqIndex][]; //, _modificationParams.NumModificationCombinations];
for (var i = 0; i < _maxSeqIndex; i++)
{
_compNodeComposition[i] = new Composition.Composition[_modificationParams.NumModificationCombinations];
_nodeComposition[i] = new Composition.Composition[_modificationParams.NumModificationCombinations];
}
NumNTermCleavages = 0;
IsValid = true;
SetNTerminalAminoAcid(nTerm);
AddAminoAcid(cTerm.Residue);
for (var i = sequence.Length - 1; i >= 0; i--)
{
if (AddAminoAcid(sequence[i]) == false)
{
IsValid = false;
break;
}
}
if (IsValid)
{
AddAminoAcid(nTerm.Residue);
}
}
internal ScoringGraph(AminoAcid[] aminoAcidSequence, Composition sequenceComposition, ScoringGraphNode rootNode,
int minPrecursorCharge, int maxPrecursorCharge)
{
_aminoAcidSequence = aminoAcidSequence;
_sequenceComposition = sequenceComposition;
_rootNode = rootNode;
_minPrecursorCharge = minPrecursorCharge;
_maxPrecursorCharge = maxPrecursorCharge;
_precursorIon = new Dictionary<int, Ion>();
for (var precursorCharge = _minPrecursorCharge; precursorCharge <= _maxPrecursorCharge; precursorCharge++)
{
_precursorIon[precursorCharge] = new Ion(_sequenceComposition, precursorCharge);
}
// store all nodes in an array
var nodes = new HashSet<ScoringGraphNode>();
var curNodes = new HashSet<ScoringGraphNode> { _rootNode };
while (curNodes.Any())
{
var newNodes = new HashSet<ScoringGraphNode>();
foreach (var node in curNodes)
{
if (nodes.Add(node)) // if node is new
{
foreach (var nextNode in node.GetNextNodes())
{
newNodes.Add(nextNode);
}
}
}
curNodes = newNodes;
}
_nodes = nodes.ToArray();
}
private void SetNTerminalAminoAcid(AminoAcid nTerm)
{
_aminoAcidSequence[_aminoAcidSequence.Length - 1 - NumNTermCleavages] = nTerm;
}
/// <summary>
/// Create a graph representing the sequence. Sequence is reversed.
/// </summary>
/// <param name="aaSet">amino acid set</param>
/// <param name="nTerm">N-term amino acid</param>
/// <param name="sequence">sequence</param>
/// <param name="cTerm">C-term amino acid</param>
/// <returns>sequence graph</returns>
public static SequenceGraph CreateGraph(AminoAcidSet aaSet, AminoAcid nTerm, string sequence, AminoAcid cTerm)
{
var seqGraph = new SequenceGraph(aaSet, nTerm, sequence, cTerm);
return(seqGraph.IsValid ? seqGraph : null);
}
public void AddMatchedAminoAcid(AminoAcid aa, double massError)
{
AminoAcidList.Add(aa);
}
internal ScoringGraph(AminoAcid[] aminoAcidSequence, Composition sequenceComposition, ScoringGraphNode rootNode)
: this(aminoAcidSequence, sequenceComposition, rootNode, DefaultMinPrecursorCharge, DefaultMaxPrecursorCharge)
{
}
public ModifiedAminoAcid(AminoAcid aa, Modification modification)
: base(aa.Residue, aa.Name+"+"+modification.Name, aa.Composition+modification.Composition)
{
var modAa = aa as ModifiedAminoAcid;
if(modAa == null) _modification = modification; // aa is not modified
else // aa is already modified
{
_modification = Modification.RegisterAndGetModification(modAa.Modification.Name+"+"+modification.Name, Composition);
}
}
public IcScores GetScores(AminoAcid nTerm, string seqStr, AminoAcid cTerm, Composition composition, int charge, int ms2ScanNum)
{
var spec = Run.GetSpectrum(ms2ScanNum) as ProductSpectrum;
if (spec == null) return null;
return GetScores(spec, seqStr, composition, charge, ms2ScanNum);
}
private void SetNTerminalAminoAcid(AminoAcid nTerm)
{
_aminoAcidSequence[_aminoAcidSequence.Length - 1 - NumNTermCleavages] = nTerm;
}
/// <summary>
/// Create a graph representing the sequence. Sequence is reversed.
/// </summary>
/// <param name="aaSet">amino acid set</param>
/// <param name="nTerm">N-term amino acid</param>
/// <param name="sequence">sequence</param>
/// <param name="cTerm">C-term amino acid</param>
/// <returns>sequence graph</returns>
public static SequenceGraph CreateGraph(AminoAcidSet aaSet, AminoAcid nTerm, string sequence, AminoAcid cTerm)
{
var seqGraph = new SequenceGraph(aaSet, nTerm, sequence, cTerm);
return seqGraph.IsValid ? seqGraph : null;
}
protected SequenceGraph(AminoAcidSet aminoAcidSet, AminoAcid nTerm, string sequence, AminoAcid cTerm)
{
_aminoAcidSet = aminoAcidSet;
_sequence = sequence;
_nTerm = nTerm;
_modificationParams = aminoAcidSet.GetModificationParams();
_maxSeqIndex = sequence.Length + 3; // init + C-term + sequence length + N-term
_index = 0;
_aminoAcidSequence = new AminoAcid[_maxSeqIndex];
_aminoAcidSequence[0] = AminoAcid.Empty;
_suffixComposition = new Composition.Composition[_maxSeqIndex];
_suffixComposition[0] = Composition.Composition.Zero;
_graph = new Node[_maxSeqIndex][];
_graph[0] = new[] { new Node(0) };
_nodeComposition = new Composition.Composition[_maxSeqIndex][]; //, _modificationParams.NumModificationCombinations];
_compNodeComposition = new Composition.Composition[_maxSeqIndex][]; //, _modificationParams.NumModificationCombinations];
for (var i = 0; i < _maxSeqIndex; i++)
{
_compNodeComposition[i] = new Composition.Composition[_modificationParams.NumModificationCombinations];
_nodeComposition[i] = new Composition.Composition[_modificationParams.NumModificationCombinations];
}
NumNTermCleavages = 0;
IsValid = true;
SetNTerminalAminoAcid(nTerm);
AddAminoAcid(cTerm.Residue);
for (var i = sequence.Length - 1; i >= 0; i--)
{
if (AddAminoAcid(sequence[i]) == false)
{
IsValid = false;
break;
}
}
if (IsValid) AddAminoAcid(nTerm.Residue);
}