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;
}
/// <summary>
/// Add an amino acid residue to this generator.
/// </summary>
/// <param name="index">index to add the amino acid. 0 is C-term. 1 is the C-term amino acid.</param>
/// <param name="residue">amino acid residue to add.</param>
/// <param name="loc">location of the residue</param>
/// <returns>true if residue is a valid amino acid; false otherwise.</returns>
private bool PutAminoAcid(int index, char residue, SequenceLocation loc)
{
_index = index + 1;
var aminoAcid = AminoAcidSet.GetAminoAcid(residue, loc);
if (aminoAcid == null) // residue is not valid
{
return false;
}
var fragmentComposition = _fragmentComposition[_index - 1] + aminoAcid.Composition;
if (fragmentComposition.Mass > _maxSequenceMass) return false;
_aminoAcidSequence[_index] = aminoAcid;
_fragmentComposition[_index] = fragmentComposition;
var modIndices = AminoAcidSet.GetModificationIndices(residue, loc);
if (!modIndices.Any()) // No modification
{
_graph[_index] = new Node[_graph[_index - 1].Length];
for (var i = 0; i < _graph[_index - 1].Length; i++)
{
_graph[_index][i] = new Node(_graph[_index - 1][i].ModificationCombinationIndex, i);
}
}
else
{
var modCombIndexToNodeMap = new Dictionary<int, Node>();
for (var i = 0; i < _graph[_index - 1].Length; i++)
{
var prevNodeIndex = i;
var prevNode = _graph[_index - 1][i];
var prevModCombIndex = prevNode.ModificationCombinationIndex;
Node newNode;
// unmodified edge
if (modCombIndexToNodeMap.TryGetValue(prevModCombIndex, out newNode))
{
newNode.AddPrevNodeIndex(prevNodeIndex);
}
else
{
modCombIndexToNodeMap.Add(prevModCombIndex, new Node(prevModCombIndex, prevNodeIndex));
}
// modified edge
foreach (var modIndex in modIndices)
{
var modCombIndex = ModificationParams.GetModificationCombinationIndex(
prevNode.ModificationCombinationIndex, modIndex);
if (modCombIndex < 0) // too many modifications
continue;
if (modCombIndexToNodeMap.TryGetValue(modCombIndex, out newNode))
{
newNode.AddPrevNodeIndex(prevNodeIndex);
}
else
{
modCombIndexToNodeMap.Add(modCombIndex, new Node(modCombIndex, prevNodeIndex));
}
}
_graph[_index] = modCombIndexToNodeMap.Values.ToArray();
}
}
return true;
}
/// <summary>
/// Add an amino acid residue to this generator.
/// </summary>
/// <param name="index">index to add the amino acid. 0 is C-term. 1 is the C-term amino acid.</param>
/// <param name="residue">amino acid residue to add.</param>
/// <returns>true if residue is a valid amino acid; false otherwise.</returns>
private bool PutAminoAcid(int index, char residue)
{
_index = index + 1;
SequenceLocation? location = null;
if(_index == 1) // C-term residue
{
if (residue == AminoAcid.PeptideCTerm.Residue) location = SequenceLocation.PeptideCTerm;
else if (residue == AminoAcid.ProteinCTerm.Residue) location = SequenceLocation.ProteinCTerm;
}
else if (_index == _aminoAcidSequence.Length - 1 - NumNTermCleavages) // N-term residue
{
if (residue == AminoAcid.PeptideNTerm.Residue) location = SequenceLocation.PeptideNTerm;
else if (residue == AminoAcid.ProteinNTerm.Residue) location = SequenceLocation.ProteinNTerm;
}
else if(_index == 2) // Amino acid at the C-term
{
if (_aminoAcidSequence[1] == AminoAcid.PeptideCTerm) location = SequenceLocation.PeptideCTerm;
else if (_aminoAcidSequence[1] == AminoAcid.ProteinCTerm) location = SequenceLocation.ProteinCTerm;
}
else if (_index == _aminoAcidSequence.Length - 2 - NumNTermCleavages) // Amino acid at the N-term
{
if (_aminoAcidSequence[_aminoAcidSequence.Length - 1] == AminoAcid.PeptideNTerm) location = SequenceLocation.PeptideNTerm;
else if (_aminoAcidSequence[_aminoAcidSequence.Length - 1] == AminoAcid.ProteinNTerm) location = SequenceLocation.ProteinNTerm;
}
else
{
location = SequenceLocation.Everywhere;
}
if (location == null) return false;
var loc = (SequenceLocation) location;
var aminoAcid = AminoAcidSet.GetAminoAcid(residue, loc);
if (aminoAcid == null) // residue is not valid
{
return false;
}
_aminoAcidSequence[_index] = aminoAcid;
_suffixComposition[_index] = _suffixComposition[_index - 1] + aminoAcid.Composition;
var modIndices = AminoAcidSet.GetModificationIndices(residue, loc);
if (!modIndices.Any()) // No modification
{
_graph[_index] = new Node[_graph[_index - 1].Length];
for (var i = 0; i < _graph[_index - 1].Length; i++)
{
_graph[_index][i] = new Node(_graph[_index - 1][i].ModificationCombinationIndex, i);
}
}
else
{
var modCombIndexToNodeMap = new Dictionary<int, Node>();
for (var i = 0; i < _graph[_index - 1].Length; i++)
{
var prevNodeIndex = i;
var prevNode = _graph[_index - 1][i];
var prevModCombIndex = prevNode.ModificationCombinationIndex;
Node newNode;
// unmodified edge
if (modCombIndexToNodeMap.TryGetValue(prevModCombIndex, out newNode))
{
newNode.AddPrevNodeIndex(prevNodeIndex);
}
else
{
modCombIndexToNodeMap.Add(prevModCombIndex, new Node(prevModCombIndex, prevNodeIndex));
}
// modified edge
foreach (var modIndex in modIndices)
{
var modCombIndex = ModificationParams.GetModificationCombinationIndex(
prevNode.ModificationCombinationIndex, modIndex);
if (modCombIndex < 0) // too many modifications
continue;
if (modCombIndexToNodeMap.TryGetValue(modCombIndex, out newNode))
{
newNode.AddPrevNodeIndex(prevNodeIndex);
}
else
{
modCombIndexToNodeMap.Add(modCombIndex, new Node(modCombIndex, prevNodeIndex));
}
}
_graph[_index] = modCombIndexToNodeMap.Values.ToArray();
}
}
return true;
}
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);
}
