/// <summary>
/// Generates a sequence based on final best peptide sequence.
/// </summary>
/// <param name="seq">unmodified sequence</param>
/// <param name="myParam">ascore parameters</param>
/// <param name="peptideMods">peptide modification position array</param>
/// <returns></returns>
private string GenerateFinalSequences(string seq, ParameterFileManager myParam, IReadOnlyList <int> peptideMods)
{
var sbFinalSeq = new System.Text.StringBuilder(seq.Length);
for (var i = 0; i < seq.Length; i++)
{
if (i >= peptideMods.Count)
{
// Invalid index for i; assume the residue is not modified
sbFinalSeq.Append(seq[i]);
}
else if (peptideMods[i] == 0)
{
sbFinalSeq.Append(seq[i]);
}
else
{
foreach (var dynamicMod in myParam.DynamicMods)
{
if (peptideMods[i] == dynamicMod.UniqueID)
{
sbFinalSeq.Append(seq[i] + dynamicMod.ModSymbol.ToString(CultureInfo.InvariantCulture));
}
}
}
}
return(sbFinalSeq.ToString());
}
private static List <double> GetChargeList(ParameterFileManager ascoreParameters, double mzmax, double mzmin, TheoreticalSpectra theoMono,
TheoreticalSpectra theoAve, int[] myPositions, out Dictionary <int, ChargeStateIons> toWrite)
{
Dictionary <int, ChargeStateIons> mySpectraMono = theoMono.GetTempSpectra(myPositions,
ascoreParameters.DynamicMods, MassType.Monoisotopic);
Dictionary <int, ChargeStateIons> mySpectraAverage = null;
if (ascoreParameters.FragmentMassTolerance <= 0.05)
{
mySpectraAverage = theoAve.GetTempSpectra(myPositions,
ascoreParameters.DynamicMods, MassType.Average);
}
//Get ions within m/z range
Dictionary <int, ChargeStateIons> mySpectra = new Dictionary <int, ChargeStateIons>();
if (ascoreParameters.FragmentMassTolerance <= 0.05)
{
mySpectra.Add(1, mySpectraMono[1]);
foreach (int charge in mySpectraAverage.Keys)
{
if (charge != 1)
{
mySpectra.Add(charge, mySpectraAverage[charge]);
}
}
}
else
{
mySpectra = mySpectraMono;
}
toWrite = mySpectra;
List <double> myIons = GetCurrentComboTheoreticalIons(mzmax, mzmin, mySpectra);
return(myIons);
}
/// <summary>
/// Gets a clean sequence initializes dynamic modifications
/// </summary>
/// <param name="seq">input protein sequence including mod characters, but without the prefix or suffix residues</param>
/// <param name="ascoreParams">ascore parameters reference</param>
/// <returns>protein sequence without mods as well as changing ascoreParams</returns>
private string GetCleanSequence(string seq, ref ParameterFileManager ascoreParams)
{
foreach (var dynamicMod in ascoreParams.DynamicMods)
{
var newSeq = seq.Replace(dynamicMod.ModSymbol.ToString(), string.Empty);
dynamicMod.Count = seq.Length - newSeq.Length;
seq = newSeq;
}
return(seq);
}
/// <summary>
/// Configure and run the AScore algorithm, optionally can add protein mapping information
/// </summary>
/// <param name="spectraManager"></param>
/// <param name="psmResultsManager"></param>
/// <param name="ascoreParams"></param>
/// <param name="outputFilePath">Name of the output file</param>
/// <param name="fastaFilePath">Path to FASTA file. If this is empty/null, protein mapping will not occur</param>
/// <param name="outputDescriptions">Whether to include protein description line in output or not.</param>
public void RunAScoreOnSingleFile(
SpectraManagerCache spectraManager,
PsmResultsManager psmResultsManager,
ParameterFileManager ascoreParams,
string outputFilePath,
string fastaFilePath = "",
bool outputDescriptions = false
)
{
var ascoreOptions = new AScoreOptions
{
FastaFilePath = fastaFilePath,
OutputProteinDescriptions = outputDescriptions
};
ascoreOptions.SetAScoreResultsFilePath(outputFilePath);
RunAScoreOnSingleFile(ascoreOptions, spectraManager, psmResultsManager, ascoreParams);
}
/// <summary>
/// Configure and run the AScore algorithm, optionally can add protein mapping information
/// </summary>
/// <param name="ascoreOptions"></param>
/// <param name="spectraManager"></param>
/// <param name="psmResultsManager"></param>
/// <param name="ascoreParams"></param>
public void RunAScoreWithMappingFile(
AScoreOptions ascoreOptions,
SpectraManagerCache spectraManager,
PsmResultsManager psmResultsManager,
ParameterFileManager ascoreParams)
{
var requiredColumns = new List <string>
{
"Job",
"Dataset"
};
OnStatusEvent("Reading Job to Dataset Map File: " + PathUtils.CompactPathString(ascoreOptions.JobToDatasetMapFile, 80));
ReadJobToDatasetMapFile(ascoreOptions, requiredColumns, out var jobToDatasetNameMap);
RunAScoreOnPreparedData(jobToDatasetNameMap, spectraManager, psmResultsManager, ascoreParams, ascoreOptions, false);
ProteinMapperTestRun(ascoreOptions);
}
/// <summary>
/// Configure and run the AScore algorithm, optionally can add protein mapping information
/// </summary>
/// <param name="ascoreOptions"></param>
/// <param name="spectraManager"></param>
/// <param name="psmResultsManager"></param>
/// <param name="ascoreParams"></param>
public void RunAScoreOnSingleFile(
AScoreOptions ascoreOptions,
SpectraManagerCache spectraManager,
PsmResultsManager psmResultsManager,
ParameterFileManager ascoreParams)
{
var jobToDatasetNameMap = new Dictionary <string, DatasetFileInfo>
{
{
psmResultsManager.JobNum,
new DatasetFileInfo(spectraManager.SpectrumFilePath, spectraManager.ModSummaryFilePath)
}
};
if (spectraManager == null || !spectraManager.Initialized)
{
throw new Exception(
"spectraManager must be instantiated and initialized before calling RunAScoreOnSingleFile for a single source file");
}
RunAScoreOnPreparedData(jobToDatasetNameMap, spectraManager, psmResultsManager, ascoreParams, ascoreOptions, true);
ProteinMapperTestRun(ascoreOptions);
}
/// <summary>
/// Gets a list of ions for matching
/// </summary>
/// <param name="ascoreParams">AScore parameters</param>
/// <param name="mzMax"></param>
/// <param name="mzMin"></param>
/// <param name="theoreticalMonoMassSpectra"></param>
/// <param name="theoreticalAverageMassSpectra"></param>
/// <param name="myPositions"></param>
/// <returns></returns>
private List <double> GetChargeList(ParameterFileManager ascoreParams, double mzMax, double mzMin,
TheoreticalSpectra theoreticalMonoMassSpectra,
TheoreticalSpectra theoreticalAverageMassSpectra,
int[] myPositions)
{
const double FRAGMENT_MASS_TOLERANCE = 0.0501;
var mySpectraMono = theoreticalMonoMassSpectra.GetTempSpectra(myPositions,
ascoreParams.DynamicMods, MassType.Monoisotopic);
var mySpectra = new Dictionary <int, ChargeStateIons>();
if (ascoreParams.FragmentMassTolerance <= FRAGMENT_MASS_TOLERANCE)
{
var mySpectraAverage = theoreticalAverageMassSpectra.GetTempSpectra(myPositions,
ascoreParams.DynamicMods, MassType.Average);
//Get ions within m/z range
mySpectra.Add(1, mySpectraMono[1]);
foreach (var charge in mySpectraAverage.Keys)
{
if (charge != 1)
{
mySpectra.Add(charge, mySpectraAverage[charge]);
}
}
}
else
{
mySpectra = mySpectraMono;
}
var myIons = GetCurrentComboTheoreticalIons(mzMax, mzMin, mySpectra);
return(myIons);
}
/// <summary>
/// Generates a sequence based on final best peptide sequence.
/// </summary>
/// <param name="seq">unmodified sequence</param>
/// <param name="myParam">ascore parameters</param>
/// <param name="peptides">peptide modification position array</param>
/// <returns></returns>
private static string GenerateFinalSequences(string seq, ParameterFileManager myParam, int[] peptides)
{
string finalSeq = "";
char[] c = seq.ToCharArray();
for (int i = 0; i < c.Length; i++)
{
if (peptides[i] == 0)
{
finalSeq += "" + c[i];
}
else
{
foreach (Mod.DynamicModification dmod in myParam.DynamicMods)
{
if (peptides[i] == dmod.UniqueID)
{
finalSeq += "" + c[i] + "" + dmod.ModSymbol;
}
}
}
}
return(finalSeq);
}
/// <summary>
/// Gets a clean sequence intitializes dynamic modifications
/// </summary>
/// <param name="seq">input protein sequence including mod characters</param>
/// <param name="ascoreParameterss">ascore parameters reference</param>
/// <returns>protein sequence without mods as well as changing ascoreParameterss</returns>
private static string GetCleanSequence(string seq, ref ParameterFileManager ascoreParameterss)
{
seq = seq.Split('.')[1];
foreach (Mod.DynamicModification dmod in ascoreParameterss.DynamicMods)
{
string s = "";
//xml hates sequest
if (dmod.ModSymbol == '*')
{
s = @"\*";
}
else if (dmod.ModSymbol == '^')
{
s = @"\^";
}
else
{
s = dmod.ModSymbol.ToString();
}
dmod.Count = Regex.Matches(seq, s).Count;
seq = seq.Replace(dmod.ModSymbol.ToString(), string.Empty);
}
return(seq);
}
/// <summary>
/// Configure and run the AScore algorithm
/// </summary>
/// <param name="ascoreOptions"></param>
/// <returns></returns>
public int RunAScore(AScoreOptions ascoreOptions)
{
var paramManager = new ParameterFileManager(ascoreOptions.AScoreParamFile);
RegisterEvents(paramManager);
Console.WriteLine();
if (paramManager.DynamicMods.Count > 0 || paramManager.StaticMods.Count > 0)
{
OnStatusEvent("Loaded modifications from: " + ascoreOptions.AScoreParamFile);
foreach (var mod in paramManager.StaticMods)
{
OnStatusEvent(Utilities.GetModDescription("Static, ", mod));
}
foreach (var mod in paramManager.DynamicMods)
{
OnStatusEvent(Utilities.GetModDescription("Dynamic, ", mod));
}
foreach (var mod in paramManager.TerminiMods)
{
OnStatusEvent(Utilities.GetModDescription("Terminus, ", mod));
}
Console.WriteLine();
}
PsmResultsManager psmResultsManager;
switch (ascoreOptions.SearchType)
{
case AScoreOptions.SearchMode.XTandem:
OnStatusEvent("Caching data in " + PathUtils.CompactPathString(ascoreOptions.DbSearchResultsFile, 80));
psmResultsManager = new XTandemFHT(ascoreOptions.DbSearchResultsFile);
break;
case AScoreOptions.SearchMode.Sequest:
OnStatusEvent("Caching data in " + PathUtils.CompactPathString(ascoreOptions.DbSearchResultsFile, 80));
psmResultsManager = new SequestFHT(ascoreOptions.DbSearchResultsFile);
break;
case AScoreOptions.SearchMode.Inspect:
OnStatusEvent("Caching data in " + PathUtils.CompactPathString(ascoreOptions.DbSearchResultsFile, 80));
psmResultsManager = new InspectFHT(ascoreOptions.DbSearchResultsFile);
break;
case AScoreOptions.SearchMode.Msgfdb:
case AScoreOptions.SearchMode.Msgfplus:
OnStatusEvent("Caching data in " + PathUtils.CompactPathString(ascoreOptions.DbSearchResultsFile, 80));
if (ascoreOptions.SearchResultsType == AScoreOptions.DbSearchResultsType.Mzid)
{
if (ascoreOptions.CreateUpdatedDbSearchResultsFile)
{
psmResultsManager = new MsgfMzidFull(ascoreOptions.DbSearchResultsFile);
}
else
{
psmResultsManager = new MsgfMzid(ascoreOptions.DbSearchResultsFile);
}
}
else
{
psmResultsManager = new MsgfdbFHT(ascoreOptions.DbSearchResultsFile);
}
break;
default:
OnErrorEvent(string.Format(
"Incorrect search type: {0} , supported values are {1}",
ascoreOptions.SearchType,
string.Join(", ", Enum.GetNames(typeof(AScoreOptions.SearchMode)))
));
return(-13);
}
var peptideMassCalculator = new PeptideMassCalculator();
var spectraManager = new SpectraManagerCache(peptideMassCalculator);
RegisterEvents(spectraManager);
OnStatusEvent("Output directory: " + ascoreOptions.OutputDirectoryInfo.FullName);
var ascoreEngine = new AScoreAlgorithm();
RegisterEvents(ascoreEngine);
// Initialize the options
FilterOnMSGFScore = ascoreOptions.FilterOnMSGFScore;
// Run the algorithm
if (ascoreOptions.MultiJobMode)
{
RunAScoreWithMappingFile(ascoreOptions, spectraManager, psmResultsManager, paramManager);
}
else
{
spectraManager.OpenFile(ascoreOptions.MassSpecFile, ascoreOptions.ModSummaryFile);
RunAScoreOnSingleFile(ascoreOptions, spectraManager, psmResultsManager, paramManager);
}
OnStatusEvent("AScore Complete");
if (ascoreOptions.CreateUpdatedDbSearchResultsFile)
{
if (ascoreOptions.SearchResultsType == AScoreOptions.DbSearchResultsType.Fht)
{
CreateUpdatedFirstHitsFile(ascoreOptions);
}
else if (psmResultsManager is MsgfMzidFull mzidFull)
{
mzidFull.WriteToMzidFile(ascoreOptions.UpdatedDbSearchResultsFileName);
OnStatusEvent("Results merged; new file: " + PathUtils.CompactPathString(ascoreOptions.UpdatedDbSearchResultsFileName, 80));
}
}
return(0);
}
/// <summary>
/// Runs the all the tools necessary to perform an ascore run
/// </summary>
/// <param name="jobToDatasetNameMap">Keys are job numbers (stored as strings); values are Dataset Names or the path to the _dta.txt file</param>
/// <param name="spectraManager">Manager for reading _dta.txt or .mzML files; must have already been initialized by the calling class</param>
/// <param name="psmResultsManager"></param>
/// <param name="ascoreParams"></param>
/// <param name="ascoreOptions"></param>
/// <param name="spectraFileOpened">Set to true if processing a single dataset, and spectraManager.OpenFile() has already been called</param>
private void RunAScoreOnPreparedData(
IReadOnlyDictionary <string, DatasetFileInfo> jobToDatasetNameMap,
SpectraManagerCache spectraManager,
PsmResultsManager psmResultsManager,
ParameterFileManager ascoreParams,
AScoreOptions ascoreOptions,
bool spectraFileOpened)
{
var totalRows = psmResultsManager.GetRowLength();
var dctPeptidesProcessed = new Dictionary <string, int>();
if (jobToDatasetNameMap == null || jobToDatasetNameMap.Count == 0)
{
const string errorMessage = "Error in AlgorithmRun: jobToDatasetNameMap cannot be null or empty";
OnErrorEvent(errorMessage);
throw new ArgumentException(errorMessage);
}
ISpectraManager spectraFile = null;
string spectraManagerCurrentJob = null; // Force open after first read from fht
var modSummaryManager = new ModSummaryFileManager();
RegisterEvents(modSummaryManager);
var peptideMassCalculator = new PeptideMassCalculator();
if (FilterOnMSGFScore)
{
OnStatusEvent("Filtering using MSGF_SpecProb <= " + ascoreParams.MSGFPreFilter.ToString("0.0E+00"));
}
Console.WriteLine();
var statsByType = new int[4];
var ascoreAlgorithm = new AScoreAlgorithm();
RegisterEvents(ascoreAlgorithm);
while (psmResultsManager.CurrentRowNum < totalRows)
{
// Console.Clear();
if (psmResultsManager.CurrentRowNum % 100 == 0)
{
Console.Write("\rPercent Completion " + Math.Round((double)psmResultsManager.CurrentRowNum / totalRows * 100) + "%");
}
int scanNumber;
int scanCount;
int chargeState;
string peptideSeq;
double msgfScore;
if (FilterOnMSGFScore)
{
psmResultsManager.GetNextRow(out scanNumber, out scanCount, out chargeState, out peptideSeq, out msgfScore, ref ascoreParams);
}
else
{
psmResultsManager.GetNextRow(out scanNumber, out scanCount, out chargeState, out peptideSeq, ref ascoreParams);
msgfScore = 1;
}
switch (ascoreParams.FragmentType)
{
case FragmentType.CID:
statsByType[(int)FragmentType.CID]++;
break;
case FragmentType.ETD:
statsByType[(int)FragmentType.ETD]++;
break;
case FragmentType.HCD:
statsByType[(int)FragmentType.HCD]++;
break;
default:
statsByType[(int)FragmentType.Unspecified]++;
break;
}
if (string.IsNullOrEmpty(spectraManagerCurrentJob) || !string.Equals(spectraManagerCurrentJob, psmResultsManager.JobNum))
{
// New dataset
// Get the correct spectrum file for the match
if (!jobToDatasetNameMap.TryGetValue(psmResultsManager.JobNum, out var datasetInfo))
{
var errorMessage = "Input file refers to job " + psmResultsManager.JobNum +
" but jobToDatasetNameMap does not contain that job; unable to continue";
OnWarningEvent(errorMessage);
if (!psmResultsManager.JobColumnDefined)
{
OnWarningEvent(
"If the input file includes results from multiple jobs, the first column must be job number with Job as the column heading");
}
throw new Exception(errorMessage);
}
var datasetName = GetDatasetName(datasetInfo.SpectrumFilePath);
OnStatusEvent("Dataset name: " + datasetName);
if (!spectraFileOpened)
{
// This method was called from RunAScoreWithMappingFile
// Open the spectrum file for this dataset
spectraFile = spectraManager.GetSpectraManagerForFile(
psmResultsManager.PSMResultsFilePath,
datasetName,
datasetInfo.ModSummaryFilePath);
}
else
{
spectraFile = spectraManager.GetCurrentSpectrumManager();
}
spectraManagerCurrentJob = string.Copy(psmResultsManager.JobNum);
Console.Write("\r");
if (string.IsNullOrWhiteSpace(datasetInfo.ModSummaryFilePath) && !string.IsNullOrWhiteSpace(ascoreOptions.ModSummaryFile))
{
datasetInfo.ModSummaryFilePath = ascoreOptions.ModSummaryFile;
}
if (psmResultsManager is MsgfMzid mzid)
{
mzid.SetModifications(ascoreParams);
}
else if (psmResultsManager is MsgfMzidFull mzidFull)
{
mzidFull.SetModifications(ascoreParams);
}
else
{
if (string.IsNullOrEmpty(datasetInfo.ModSummaryFilePath))
{
modSummaryManager.ReadModSummary(spectraFile.DatasetName, psmResultsManager.PSMResultsFilePath, ascoreParams);
}
else
{
var modSummaryFile = new FileInfo(datasetInfo.ModSummaryFilePath);
modSummaryManager.ReadModSummary(modSummaryFile, ascoreParams);
}
}
Console.WriteLine();
Console.Write("\rPercent Completion " + Math.Round((double)psmResultsManager.CurrentRowNum / totalRows * 100) + "%");
}
// perform work on the match
var peptideParts = peptideSeq.Split('.');
string sequenceWithoutSuffixOrPrefix;
string front;
string back;
if (peptideParts.Length >= 3)
{
front = peptideParts[0];
sequenceWithoutSuffixOrPrefix = peptideParts[1];
back = peptideParts[2];
}
else
{
front = "?";
sequenceWithoutSuffixOrPrefix = string.Copy(peptideSeq);
back = "?";
}
var sequenceClean = GetCleanSequence(sequenceWithoutSuffixOrPrefix, ref ascoreParams);
var skipPSM = FilterOnMSGFScore && msgfScore > ascoreParams.MSGFPreFilter;
var scanChargePeptide = scanNumber + "_" + chargeState + "_" + sequenceWithoutSuffixOrPrefix;
if (dctPeptidesProcessed.ContainsKey(scanChargePeptide))
{
// We have already processed this PSM
skipPSM = true;
}
else
{
dctPeptidesProcessed.Add(scanChargePeptide, 0);
}
if (skipPSM)
{
psmResultsManager.IncrementRow();
continue;
}
//Get experimental spectra
if (spectraFile == null)
{
const string errorMessage = "spectraFile is uninitialized in RunAScoreOnPreparedData; this indicates a programming bug";
OnErrorEvent(errorMessage);
throw new Exception(errorMessage);
}
var expSpec = spectraFile.GetExperimentalSpectra(scanNumber, scanCount, chargeState);
if (expSpec == null)
{
OnWarningEvent("Scan " + scanNumber + " not found in spectra file for peptide " + peptideSeq);
psmResultsManager.IncrementRow();
continue;
}
// Assume monoisotopic for both hi res and low res spectra
MolecularWeights.MassType = MassType.Monoisotopic;
// Compute precursor m/z value
var precursorMZ = peptideMassCalculator.ConvoluteMass(expSpec.PrecursorMass, 1, chargeState);
// Set the m/z range
var mzMax = maxRange;
var mzMin = precursorMZ * lowRangeMultiplier;
if (ascoreParams.FragmentType != FragmentType.CID)
{
mzMax = maxRange;
mzMin = minRange;
}
//Generate all combination mixtures
var modMixture = new Combinatorics.ModMixtureCombo(ascoreParams.DynamicMods, sequenceClean);
var myPositionsList = GetMyPositionList(sequenceClean, modMixture);
//If I have more than 1 modifiable site proceed to calculation
if (myPositionsList.Count > 1)
{
ascoreAlgorithm.ComputeAScore(psmResultsManager, ascoreParams, scanNumber, chargeState,
peptideSeq, front, back, sequenceClean, expSpec,
mzMax, mzMin, myPositionsList);
}
else if (myPositionsList.Count == 1)
{
// Either one or no modifiable sites
var uniqueID = myPositionsList[0].Max();
if (uniqueID == 0)
{
psmResultsManager.WriteToTable(peptideSeq, scanNumber, 0, myPositionsList[0], MOD_INFO_NO_MODIFIED_RESIDUES);
}
else
{
psmResultsManager.WriteToTable(peptideSeq, scanNumber, 0, myPositionsList[0], LookupModInfoByID(uniqueID, ascoreParams.DynamicMods));
}
}
else
{
// No modifiable sites
psmResultsManager.WriteToTable(peptideSeq, scanNumber, 0, new int[0], MOD_INFO_NO_MODIFIED_RESIDUES);
}
psmResultsManager.IncrementRow();
}
Console.WriteLine();
OnStatusEvent(string.Format("Writing {0:N0} rows to {1}", psmResultsManager.ResultsCount, PathUtils.CompactPathString(ascoreOptions.AScoreResultsFilePath, 80)));
psmResultsManager.WriteToFile(ascoreOptions.AScoreResultsFilePath);
Console.WriteLine();
if (statsByType.Sum() == 0)
{
OnWarningEvent("Input file appeared empty");
}
else
{
OnStatusEvent("Stats by fragmentation ion type:");
ReportStatsForFragType(" CID", statsByType, FragmentType.CID);
ReportStatsForFragType(" ETD", statsByType, FragmentType.ETD);
ReportStatsForFragType(" HCD", statsByType, FragmentType.HCD);
}
Console.WriteLine();
}
public void ComputeAScore(
PsmResultsManager psmResultsManager, ParameterFileManager ascoreParams, int scanNumber,
int chargeState, string peptideSeq, string front, string back, string sequenceClean, ExperimentalSpectra expSpec,
double mzMax, double mzMin, IReadOnlyList <int[]> myPositionsList)
{
// Change the charge state to 2+ if it is 1+
if (chargeState == 1)
{
chargeState = 2;
}
// Parallel lists of scores
var peptideScores = new List <List <double> >();
var weightedScores = new List <List <double> >();
try
{
var theoreticalMonoMassSpectra = new TheoreticalSpectra(sequenceClean, ascoreParams, chargeState, MassType.Monoisotopic);
var theoreticalAverageMassSpectra = new TheoreticalSpectra(sequenceClean, ascoreParams, chargeState, MassType.Average);
var peptideMassTheoretical = theoreticalMonoMassSpectra.PeptideNeutralMassWithStaticMods + GetModMassTotal(peptideSeq, ascoreParams.DynamicMods);
if (Math.Abs(peptideMassTheoretical - expSpec.PrecursorNeutralMass) > 20)
{
OnWarningEvent(string.Format(
"Scan {0}: Observed precursor mass of {1:F1} Da is more than 20 Da away from the computed mass of {2:F1} Da; DeltaMass = {3:F1} Da",
scanNumber,
expSpec.PrecursorNeutralMass,
peptideMassTheoretical,
expSpec.PrecursorNeutralMass - peptideMassTheoretical));
}
else
{
// Make sure the masses agree within a reasonable tolerance
var validMatch = false;
for (double chargeAdjust = 0; chargeAdjust < 0.1; chargeAdjust += 0.005)
{
for (var massAdjust = -chargeState - 3; massAdjust <= chargeState + 3; massAdjust++)
{
var delM = peptideMassTheoretical - expSpec.PrecursorNeutralMass + massAdjust * MASS_C13;
if (Math.Abs(delM) < 0.15 + chargeState * chargeAdjust)
{
validMatch = true;
break;
}
}
if (validMatch)
{
break;
}
}
if (!validMatch)
{
OnWarningEvent(string.Format(
"Scan {0}: Observed precursor mass of {1:F1} Da is not a reasonable match for computed mass of {2:F1} Da; " +
"DeltaMass = {3:F1} Da; Peptide = {4}",
scanNumber,
expSpec.PrecursorNeutralMass,
peptideMassTheoretical,
expSpec.PrecursorNeutralMass - peptideMassTheoretical,
peptideSeq
));
}
}
var modNumber = 0;
foreach (var myPositions in myPositionsList)
{
//Generate spectra for a modification combination
var myIons = GetChargeList(ascoreParams, mzMax, mzMin, theoreticalMonoMassSpectra, theoreticalAverageMassSpectra, myPositions);
peptideScores.Add(new List <double>());
weightedScores.Add(new List <double>());
for (var peakDepth = 1; peakDepth < 11; ++peakDepth)
{
var peakDepthSpectra = expSpec.GetPeakDepthSpectra(peakDepth);
peakDepthSpectra.Sort();
var matchedIons = GetMatchedMZ(ascoreParams.FragmentMassTolerance, myIons, peakDepthSpectra);
//Adjusted peptide score to score based on tolerance window.
var score = PeptideScoresManager.GetPeptideScore(
peakDepth * ascoreParams.FragmentMassTolerance * 2 / 100.0, myIons.Count, matchedIons.Count);
// Check if there were any negative scores
peptideScores[modNumber].Add(score);
weightedScores[modNumber].Add(score * ScoreWeights[peakDepth - 1]);
}
modNumber++;
}
var sortedSumScore = new List <ValueIndexPair <double> >();
for (var seq = 0; seq < peptideScores.Count; ++seq)
{
var score = 0.0;
for (var depth = 0; depth < peptideScores[seq].Count; ++depth)
{
score += weightedScores[seq][depth];
}
sortedSumScore.Add(new ValueIndexPair <double>(score, seq));
}
sortedSumScore.Sort();
var topPeptideScore = sortedSumScore[0].Value;
// Need the phosphorylation sites for the top peptide
var topPeptidePTMSites = myPositionsList[sortedSumScore[0].Index];
var ascoreResults = CalculateAScoreForSite(ascoreParams, expSpec, mzMax, mzMin, myPositionsList, topPeptidePTMSites, peptideScores, theoreticalMonoMassSpectra,
theoreticalAverageMassSpectra, sortedSumScore);
foreach (var ascoreResult in ascoreResults)
{
ascoreResult.SecondSequence = front + "." +
GenerateFinalSequences(sequenceClean, ascoreParams, ascoreResult.PeptideMods) + "." + back;
}
//Put scores into our table
var bestSeq = front + "." + GenerateFinalSequences(sequenceClean, ascoreParams, topPeptidePTMSites) + "." + back;
foreach (var ascoreResult in ascoreResults)
{
psmResultsManager.WriteToTable(peptideSeq, bestSeq, scanNumber, topPeptideScore, ascoreResult);
}
}
catch (Exception ex)
{
OnErrorEvent("Exception in ComputeAScore: " + ex.Message);
throw;
}
}
/// <summary>
/// Calculate the AScore for the given site information
/// </summary>
/// <param name="ascoreParams"></param>
/// <param name="expSpec"></param>
/// <param name="mzMax"></param>
/// <param name="mzMin"></param>
/// <param name="myPositionsList"></param>
/// <param name="topPeptidePTMSites"></param>
/// <param name="peptideScores"></param>
/// <param name="theoreticalMonoMassSpectra"></param>
/// <param name="theoreticalAverageMassSpectra"></param>
/// <param name="sortedSumScore"></param>
/// <returns></returns>
private List <AScoreResult> CalculateAScoreForSite(ParameterFileManager ascoreParams, ExperimentalSpectra expSpec,
double mzMax, double mzMin,
IReadOnlyList <int[]> myPositionsList,
int[] topPeptidePTMSites,
IReadOnlyList <List <double> > peptideScores,
TheoreticalSpectra theoreticalMonoMassSpectra,
TheoreticalSpectra theoreticalAverageMassSpectra,
IReadOnlyList <ValueIndexPair <double> > sortedSumScore)
{
// Initialize AScore results storage
var lstResults = new List <AScoreResult>();
var siteInfo = GetSiteDict(topPeptidePTMSites);
// Get the top sequence theoretical spectra
var topTheoreticalIons = GetChargeList(ascoreParams, mzMax, mzMin, theoreticalMonoMassSpectra, theoreticalAverageMassSpectra, topPeptidePTMSites);
for (var indSite = 0; indSite < siteInfo.Count; ++indSite)
{
var ascoreResult = new AScoreResult();
lstResults.Add(ascoreResult);
ascoreResult.ModInfo = LookupModInfoByID(siteInfo.Values[indSite], ascoreParams.DynamicMods);
int secondPeptide;
for (secondPeptide = 0; secondPeptide < sortedSumScore.Count; ++secondPeptide)
{
var secondDict = GetSiteDict(myPositionsList[sortedSumScore[secondPeptide].Index]);
var othersMatch = true;
if (!secondDict.ContainsKey(siteInfo.Keys[indSite]))
{
var sites = siteInfo.Keys.ToList();
for (var i = 0; i < sites.Count; i++)
{
if (i != indSite)
{
othersMatch = othersMatch && secondDict.ContainsKey(sites[i]);
}
}
if (othersMatch)
{
ascoreResult.PeptideMods = myPositionsList[sortedSumScore[secondPeptide].Index];
break;
}
}
else
{
if (secondDict[siteInfo.Keys[indSite]] != siteInfo.Values[indSite])
{
ascoreResult.PeptideMods = myPositionsList[sortedSumScore[secondPeptide].Index];
break;
}
}
}
if (secondPeptide == sortedSumScore.Count)
{
ascoreResult.AScore = 1000;
ascoreResult.NumSiteIons = 0;
ascoreResult.SiteDetermineMatched = 0;
continue;
}
var secondTopPeptidePTMSites = myPositionsList[sortedSumScore[secondPeptide].Index];
// Get the second best scoring spectra
var secondTopTheoreticalIons = GetChargeList(ascoreParams,
mzMax, mzMin,
theoreticalMonoMassSpectra,
theoreticalAverageMassSpectra,
secondTopPeptidePTMSites);
// Calculate the diff score between the top and second sites
var diffScore = new List <ValueIndexPair <double> >();
for (var i = 0; i < peptideScores[0].Count; ++i)
{
diffScore.Add(new ValueIndexPair <double>(
peptideScores[sortedSumScore[0].Index][i] -
peptideScores[sortedSumScore[secondPeptide].Index][i], i));
}
// Sort in descending order
diffScore.Sort();
// Find the peak depth for the diff score
var peakDepthForAScore = 1;
if (diffScore[0].Value > 0)
{
peakDepthForAScore = diffScore[0].Index + 1;
}
var siteIons1 = GetSiteDeterminingIons(topTheoreticalIons, secondTopTheoreticalIons);
var siteIons2 = GetSiteDeterminingIons(secondTopTheoreticalIons, topTheoreticalIons);
var peakDepthSpectraFinal = expSpec.GetPeakDepthSpectra(peakDepthForAScore);
peakDepthSpectraFinal.Sort();
var bestDeterminingCount = GetMatchedMZ(ascoreParams.FragmentMassTolerance, siteIons1, peakDepthSpectraFinal).Count;
var secondBestDeterminingCount = GetMatchedMZ(ascoreParams.FragmentMassTolerance, siteIons2, peakDepthSpectraFinal).Count;
var a1 = PeptideScoresManager.GetPeptideScore(peakDepthForAScore * ascoreParams.FragmentMassTolerance * 2 / 100,
siteIons1.Count, bestDeterminingCount);
var a2 = PeptideScoresManager.GetPeptideScore(peakDepthForAScore * ascoreParams.FragmentMassTolerance * 2 / 100,
siteIons2.Count, secondBestDeterminingCount);
// Add the results to the list
ascoreResult.AScore = Math.Abs(a1 - a2);
ascoreResult.NumSiteIons = siteIons1.Count; // numSiteIonsPoss
ascoreResult.SiteDetermineMatched = bestDeterminingCount; // numSiteIonsMatched
}
return(lstResults);
}
/// <summary>
/// Runs the all the tools necessary to perform an ascore run
/// </summary>
/// <param name="dtaFileName">dta file path</param>
/// <param name="parameterFile">parameter file path</param>
/// <param name="datasetFileName">dataset file path</param>
/// <param name="outputFilePath">output file path</param>
public static void AlgorithmRun(string dtaFileName, string parameterFile, string datasetFileName, string outputFilePath)
{
// IonWriter myIonWriter = new IonWriter(System.IO.Path.Combine(System.IO.Path.GetDirectoryName(datasetFileName), "IonLog.txt"));
System.Data.DataTable dt = Utilities.TextFileToDataTableAssignTypeString(datasetFileName, false);
DtaManager dtaManager = new DtaManager(dtaFileName);
ParameterFileManager ascoreParameters = new ParameterFileManager(parameterFile);
//string datasetName = "Syne_Glyco-2_degly-S_10Jul11_Andromeda_11-06-19";
//int scanNumber = 7315;
//int scanCount = 1;
//int chargeState = 3;
//string sequence = "IVNDELESLGYGENLLNLSTINR";
string datasetName = System.IO.Path.GetFileName(dtaFileName).Substring(0, System.IO.Path.GetFileName(dtaFileName).Length - 8);
//adds columns to the datatable corresponding to ascore info
dt.Columns.Add("BestSequence", typeof(string));
dt.Columns.Add("PeptideScore", typeof(double));
dt.Columns.Add("AScore", typeof(double));
dt.Columns.Add("numSiteIons", typeof(int));
dt.Columns.Add("SecondSequence", typeof(string));
int totalRows = dt.Rows.Count;
int rowsCounts = 0;
//Where all the action happens
for (int t = 0; t < totalRows; t++)
{
Console.WriteLine(rowsCounts++ + " / " + totalRows);
//Use array of column names specific to other id forms
int scanNumber = int.Parse((string)dt.Rows[t]["ScanNum"]);
int scanCount = int.Parse((string)dt.Rows[t]["ScanCount"]);
int chargeState = int.Parse((string)dt.Rows[t]["ChargeState"]);
string peptideSeq = (string)dt.Rows[t]["Peptide"];
string[] splittedPep = peptideSeq.Split('.');
string front = splittedPep[0];
string back = splittedPep[2];
//TODO:what this does and change name of ascoreParameters
string sequence = GetCleanSequence(peptideSeq, ref ascoreParameters);
//Generate combinations for this sequence/mod seto
//Dictionary of list Now part of an object
//List<List<int>> mySites = GetSiteLocation(ascoreParameters.DynamicMods, sequence);
//List<List<List<int>>> myCombos = GenerateCombosToCheck(mySites, ascoreParameters.DynamicMods);
//Get experimental spectra
ExperimentalSpectra expSpec = dtaManager.GetExperimentalSpectra(GetDtaFileName(datasetName, scanNumber,
scanCount, chargeState));
//I assume monoisotopic here, nobody uses average anymore.
MolecularWeights.MassType = MassType.Monoisotopic;
//Get precursor
double precursorMZ = (expSpec.PrecursorMass +
((expSpec.PrecursorChargeState - 1) *
MolecularWeights.Hydrogen)) / chargeState;
//Set the m/z range
//Remove magic numbers parameterize
double mzmax = 2000.0;
double mzmin = precursorMZ * 0.28;
if (ascoreParameters.FragmentType == FragmentType.CID)
{
mzmax = 2000.0;
mzmin = precursorMZ * 0.28;
}
else
{
mzmax = 2000.0;
mzmin = 50.0;
}
//initialize ascore variable storage
List <double> vecAScore = new List <double>();
List <int> vecNumSiteIons = new List <int>();
List <List <int> > AScorePeptide = new List <List <int> >();
//Generate all combination mixtures
Combinatorics.ModMixtureCombo modMixture = new Combinatorics.ModMixtureCombo(ascoreParameters.DynamicMods, sequence);
List <int[]> myPositionsList = new List <int[]>();
foreach (List <int> mycom in modMixture.FinalCombos)
{
int[] myPositions = new int[sequence.Length];
for (int i = 0; i < mycom.Count; i++)
{
myPositions[modMixture.AllSite[i]] = mycom[i];
}
myPositionsList.Add(myPositions);
}
//If I have more than 1 modifiable site proceed to calculation
if (myPositionsList.Count > 1)
{
List <List <double> > peptideScores = new List <List <double> >();
List <List <double> > weightedScores = new List <List <double> >();
TheoreticalSpectra theo = new TheoreticalSpectra(sequence, ascoreParameters, chargeState,
new List <Mod.DynamicModification>(), MassType.Monoisotopic);
int modNumber = 0;
foreach (int[] myPositions in myPositionsList)
{
//Generate spectra for a modification combination
Dictionary <int, ChargeStateIons> mySpectra = theo.GetTempSpectra(myPositions,
ascoreParameters.DynamicMods, MassType.Monoisotopic);
//Get ions within m/z range
List <double> myIons = GetCurrentComboTheoreticalIons(mzmax, mzmin, mySpectra);
peptideScores.Add(new List <double>());
weightedScores.Add(new List <double>());
for (int peakDepth = 1; peakDepth < 11; ++peakDepth)
{
List <ExperimentalSpectraEntry> peakDepthSpectra = expSpec.GetPeakDepthSpectra(peakDepth);
List <double> matchedIons = GetMatchedMZ(
peakDepth, ascoreParameters.FragmentMassTolerance,
myIons, peakDepthSpectra);
//if (scanNumber == 19446)
//{
// myIonWriter.MatchList(matchedIons);
//}
//Adjusted peptide score to score based on tolerance window.
double score = PeptideScoresManager.GetPeptideScore(
((double)peakDepth * ascoreParameters.FragmentMassTolerance * 2) / 100.0, myIons.Count, matchedIons.Count);
// Check if there were any negative scores
peptideScores[modNumber].Add(score);
weightedScores[modNumber].Add(
score * ScoreWeights[peakDepth - 1]);
}
modNumber++;
}
List <ValueIndexPair <double> > sortedSumScore = new List <ValueIndexPair <double> >();
for (int seq = 0; seq < peptideScores.Count; ++seq)
{
double score = 0.0;
for (int depth = 0; depth < peptideScores[seq].Count; ++depth)
{
score += weightedScores[seq][depth];
}
sortedSumScore.Add(new ValueIndexPair <double>(score, seq));
}
sortedSumScore.Sort(new ValueIndexPair <double> .SortValueDescend());
double topPeptideScore = sortedSumScore[0].Value;
// Need the phosphorylation sites for the top peptide
int[] topPeptidePTMsites =
myPositionsList[sortedSumScore[0].Index];
// Get the top sequence theoretical spectra
Dictionary <int, ChargeStateIons> topSpectra = theo.GetTempSpectra(topPeptidePTMsites,
ascoreParameters.DynamicMods, MassType.Monoisotopic);
List <double> topTheoIons = GetCurrentComboTheoreticalIons(mzmax, mzmin, topSpectra);
int secondPeptide = 1;
int[] secondTopPeptidePTMsites = myPositionsList[sortedSumScore[secondPeptide].Index];
// Get the second best scoring spectra
Dictionary <int, ChargeStateIons> secondTopSpectra = theo.GetTempSpectra(
secondTopPeptidePTMsites, ascoreParameters.DynamicMods, MassType.Monoisotopic);
List <double> secondTopTheoIons = GetCurrentComboTheoreticalIons(mzmax, mzmin, secondTopSpectra);
// Calculate the diff score between the top and second sites
List <ValueIndexPair <double> > diffScore = new List <ValueIndexPair <double> >();
for (int i = 0; i < peptideScores[0].Count; ++i)
{
diffScore.Add(new ValueIndexPair <double>(
peptideScores[sortedSumScore[0].Index][i] -
peptideScores[sortedSumScore[secondPeptide].Index][i], i));
}
// Sort in descending order
diffScore.Sort(new ValueIndexPair <double> .SortValueDescend());
// Find the peak depth for the diff score
int peakDepthForAScore = 1;
if (diffScore[0].Value > 0)
{
peakDepthForAScore = diffScore[0].Index + 1;
}
List <double> siteIons1 = GetSiteDeterminingIons(topTheoIons, secondTopTheoIons);
List <double> siteIons2 = GetSiteDeterminingIons(secondTopTheoIons, topTheoIons);
List <ExperimentalSpectraEntry> peakDepthSpectraFinal = expSpec.GetPeakDepthSpectra(peakDepthForAScore);
int bestDterminingCount = GetMatchedMZ(peakDepthForAScore,
ascoreParameters.FragmentMassTolerance, siteIons1, peakDepthSpectraFinal).Count;
double a1 = PeptideScoresManager.GetPeptideScore(
((double)peakDepthForAScore * ascoreParameters.FragmentMassTolerance * 2) / 100,
siteIons1.Count, bestDterminingCount);
double a2 = PeptideScoresManager.GetPeptideScore(
((double)peakDepthForAScore * ascoreParameters.FragmentMassTolerance * 2) / 100,
siteIons2.Count, GetMatchedMZ(peakDepthForAScore,
ascoreParameters.FragmentMassTolerance, siteIons2, peakDepthSpectraFinal).Count);
// Add the results to the list
vecAScore.Add(Math.Abs(a1 - a2));
vecNumSiteIons.Add(siteIons1.Count);
//Put scores into our table
dt.Rows[t]["BestSequence"] = front + "." + GenerateFinalSequences(sequence,
ascoreParameters, topPeptidePTMsites) + "." + back;
dt.Rows[t]["PeptideScore"] = "" + topPeptideScore;
dt.Rows[t]["AScore"] = "" + vecAScore[0];
dt.Rows[t]["numSiteIons"] = "" + bestDterminingCount;
dt.Rows[t]["SecondSequence"] = front + "." + GenerateFinalSequences(sequence,
ascoreParameters, secondTopPeptidePTMsites) + "." + back;
}
else
{
List <double> weightedScores = new List <double>();
TheoreticalSpectra theo = new TheoreticalSpectra(sequence, ascoreParameters, chargeState,
new List <Mod.DynamicModification>(), MassType.Monoisotopic);
foreach (int[] myPositions in myPositionsList)
{
Dictionary <int, ChargeStateIons> mySpectra = theo.GetTempSpectra(myPositions,
ascoreParameters.DynamicMods, MassType.Monoisotopic);
List <double> myIons = GetCurrentComboTheoreticalIons(mzmax, mzmin, mySpectra);
for (int peakDepth = 1; peakDepth < 11; ++peakDepth)
{
List <ExperimentalSpectraEntry> peakDepthSpectra = expSpec.GetPeakDepthSpectra(peakDepth);
List <double> matchedIons = GetMatchedMZ(
peakDepth, ascoreParameters.FragmentMassTolerance,
myIons, peakDepthSpectra);
//Adjusted peptide score to score based on tolerance window.
double score = PeptideScoresManager.GetPeptideScore(
((double)peakDepth * ascoreParameters.FragmentMassTolerance * 2) / 100.0, myIons.Count, matchedIons.Count);
// Check if there were any negative scores
weightedScores.Add(
score * ScoreWeights[peakDepth - 1]);
}
}
double pScore = 0.0;
for (int depth = 0; depth < weightedScores.Count; ++depth)
{
pScore += weightedScores[depth];
}
//Nothing to calculate
dt.Rows[t]["BestSequence"] = peptideSeq;
dt.Rows[t]["PeptideScore"] = "" + pScore;
if (myPositionsList[0].Count(i => i > 0) > 0)
{
dt.Rows[t]["AScore"] = "1000";
}
else
{
dt.Rows[t]["AScore"] = "-1";
}
dt.Rows[t]["numSiteIons"] = "0";
dt.Rows[t]["SecondSequence"] = "---";
}
}
Utilities.WriteDataTableToText(dt, outputFilePath);
}
/// <summary>
/// Runs the all the tools necessary to perform an ascore run
/// </summary>
/// <param name="dtaFileName">dta file path</param>
/// <param name="parameterFile">parameter file path</param>
/// <param name="datasetFileName">dataset file path</param>
/// <param name="outputFilePath">output file path</param>
public static void AlgorithmRunChargeAve(string dtaFileName, string parameterFile, string datasetFileName, string outputFilePath)
{
// IonWriter myIonWriter = new IonWriter(System.IO.Path.Combine(System.IO.Path.GetDirectoryName(datasetFileName), "IonLog.txt"));
System.Data.DataTable dt = Utilities.TextFileToDataTableAssignTypeString(datasetFileName, false);
DtaManager dtaManager = new DtaManager(dtaFileName);
ParameterFileManager ascoreParameters = new ParameterFileManager(parameterFile);
//string datasetName = "Syne_Glyco-2_degly-S_10Jul11_Andromeda_11-06-19";
//int scanNumber = 7315;
//int scanCount = 1;
//int chargeState = 3;
//string sequence = "IVNDELESLGYGENLLNLSTINR";
//Switch back to zero
string datasetName = System.IO.Path.GetFileName(dtaFileName).Substring(0, System.IO.Path.GetFileName(dtaFileName).Length - 8);
//adds columns to the datatable corresponding to ascore info
dt.Columns.Add("BestSequence", typeof(string));
dt.Columns.Add("PeptideScore", typeof(double));
for (int i = 1; i < 4; i++)
{
dt.Columns.Add("AScore" + i, typeof(double));
dt.Columns.Add("numSiteIonsMatched" + i, typeof(int));
dt.Columns.Add("numSiteIonsPoss" + i, typeof(int));
dt.Columns.Add("SecondSequence" + i, typeof(string));
dt.Columns["AScore" + i].DefaultValue = -1;
dt.Columns["numSiteIonsMatched" + i].DefaultValue = 0;
dt.Columns["numSiteIonsPoss" + i].DefaultValue = 0;
dt.Columns["SecondSequence" + i].DefaultValue = "---";
}
for (int k = 0; k < dt.Rows.Count; k++)
{
for (int p = 1; p < 4; p++)
{
dt.Rows[k]["AScore" + p] = -1;
dt.Rows[k]["numSiteIonsMatched" + p] = 0;
dt.Rows[k]["numSiteIonsPoss" + p] = 0;
dt.Rows[k]["SecondSequence" + p] = "---";
}
}
int totalRows = dt.Rows.Count;
int rowsCounts = 0;
//Where all the action happens
for (int t = 0; t < totalRows; t++)
{
Console.WriteLine(rowsCounts++ + " / " + totalRows);
//Use array of column names specific to other id forms
int scanNumber = int.Parse((string)dt.Rows[t]["ScanNum"]);
int scanCount = int.Parse((string)dt.Rows[t]["ScanCount"]);
int chargeState = int.Parse((string)dt.Rows[t]["ChargeState"]);
string peptideSeq = (string)dt.Rows[t]["Peptide"];
string[] splittedPep = peptideSeq.Split('.');
string front = splittedPep[0];
string back = splittedPep[2];
//TODO:what this does and change name of ascoreParameters
string sequence = GetCleanSequence(peptideSeq, ref ascoreParameters);
//Generate combinations for this sequence/mod seto
//Dictionary of list
//List<List<int>> mySites = GetSiteLocation(ascoreParameters.DynamicMods, sequence);
//List<List<List<int>>> myCombos = GenerateCombosToCheck(mySites, ascoreParameters.DynamicMods);
//Get experimental spectra
ExperimentalSpectra expSpec = dtaManager.GetExperimentalSpectra(GetDtaFileName(datasetName, scanNumber,
scanCount, chargeState));
if (expSpec == null)
{
continue;
}
//I assume monoisotopic here, nobody uses average anymore.
MolecularWeights.MassType = MassType.Monoisotopic;
//Get precursor
double precursorMZ = (expSpec.PrecursorMass +
((expSpec.PrecursorChargeState - 1) *
MolecularWeights.Hydrogen)) / chargeState;
//Set the m/z range
//Remove magic numbers parameterize
double mzmax = 2000.0;
double mzmin = precursorMZ * 0.28;
if (ascoreParameters.FragmentType == FragmentType.CID)
{
mzmax = 2000.0;
mzmin = precursorMZ * 0.28;
}
else
{
mzmax = 2000.0;
mzmin = 50.0;
}
//initialize ascore variable storage
List <double> vecAScore = new List <double>();
List <int> vecNumSiteIons = new List <int>();
List <int[]> AScorePeptide = new List <int[]>();
List <int> siteDetermineMatched = new List <int>();
//Generate all combination mixtures
Combinatorics.ModMixtureCombo modMixture = new Combinatorics.ModMixtureCombo(ascoreParameters.DynamicMods, sequence);
List <int[]> myPositionsList = new List <int[]>();
foreach (List <int> mycom in modMixture.FinalCombos)
{
int[] myPositions = new int[sequence.Length];
for (int i = 0; i < mycom.Count; i++)
{
myPositions[modMixture.AllSite[i]] = mycom[i];
}
myPositionsList.Add(myPositions);
}
//If I have more than 1 modifiable site proceed to calculation
if (myPositionsList.Count > 1 && chargeState > 1)
{
List <List <double> > peptideScores = new List <List <double> >();
List <List <double> > weightedScores = new List <List <double> >();
TheoreticalSpectra theoMono = new TheoreticalSpectra(sequence, ascoreParameters, chargeState,
new List <Mod.DynamicModification>(), MassType.Monoisotopic);
TheoreticalSpectra theoAve = new TheoreticalSpectra(sequence, ascoreParameters, chargeState,
new List <Mod.DynamicModification>(), MassType.Average);
int modNumber = 0;
foreach (int[] myPositions in myPositionsList)
{
//Generate spectra for a modification combination
List <double> myIons = GetChargeList(ascoreParameters, mzmax, mzmin, theoMono, theoAve, myPositions);
peptideScores.Add(new List <double>());
weightedScores.Add(new List <double>());
for (int peakDepth = 1; peakDepth < 11; ++peakDepth)
{
List <ExperimentalSpectraEntry> peakDepthSpectra = expSpec.GetPeakDepthSpectra(peakDepth);
List <double> matchedIons = GetMatchedMZ(
peakDepth, ascoreParameters.FragmentMassTolerance,
myIons, peakDepthSpectra);
//if (scanNumber == 19446)
//{
// myIonWriter.MatchList(matchedIons);
//}
//Adjusted peptide score to score based on tolerance window.
double score = PeptideScoresManager.GetPeptideScore(
((double)peakDepth * ascoreParameters.FragmentMassTolerance * 2) / 100.0, myIons.Count, matchedIons.Count);
// Check if there were any negative scores
peptideScores[modNumber].Add(score);
weightedScores[modNumber].Add(
score * ScoreWeights[peakDepth - 1]);
}
modNumber++;
}
List <ValueIndexPair <double> > sortedSumScore = new List <ValueIndexPair <double> >();
for (int seq = 0; seq < peptideScores.Count; ++seq)
{
double score = 0.0;
for (int depth = 0; depth < peptideScores[seq].Count; ++depth)
{
score += weightedScores[seq][depth];
}
sortedSumScore.Add(new ValueIndexPair <double>(score, seq));
}
sortedSumScore.Sort(new ValueIndexPair <double> .SortValueDescend());
double topPeptideScore = sortedSumScore[0].Value;
// Need the phosphorylation sites for the top peptide
int[] topPeptidePTMsites =
myPositionsList[sortedSumScore[0].Index];
SortedList <int, int> siteInfo = GetSiteDict(topPeptidePTMsites);
Dictionary <int, ChargeStateIons> bestDatatoWrite = new Dictionary <int, ChargeStateIons>();
// Get the top sequence theoretical spectra
List <double> topTheoIons = GetChargeList(ascoreParameters, mzmax, mzmin, theoMono, theoAve, topPeptidePTMsites, out bestDatatoWrite);
List <char> modChars = new List <char>();
foreach (Mod.DynamicModification m in ascoreParameters.DynamicMods)
{
modChars.Add(m.ModSymbol);
}
DataToExcelPractice.InfoForIons bestIonReport = new DataToExcelPractice.InfoForIons();
bestIonReport.Sequence = GenerateFinalSequences(sequence, ascoreParameters, topPeptidePTMsites);
bestIonReport.InitializeSequenceList(modChars.ToArray());
bestIonReport.ChargeState = chargeState;
bestIonReport.Depth = -1;
foreach (int k in bestDatatoWrite.Keys)
{
bestIonReport.BionData.Add(k, new List <double>(bestDatatoWrite[k].BIons));
bestIonReport.YionData.Add(k, new List <double>(bestDatatoWrite[k].YIons));
}
List <double> topIonIntensity = new List <double>();
bestIonReport.MatchesForThisDepth = GetMatchedMZStoreIntensity(10, ascoreParameters.FragmentMassTolerance, topTheoIons,
expSpec.GetPeakDepthSpectra(10), out topIonIntensity);
bestIonReport.MatchedIonIntensity = new List <double>(topIonIntensity);
bestIonReport.AScore = -1.0;
bestIonReport.ScanNumber = scanNumber;
List <DataToExcelPractice.InfoForIons> otherSites = new List <DataToExcelPractice.InfoForIons>();
for (int indSite = 0; indSite < siteInfo.Count; ++indSite)
{
int secondPeptide = 0;
for (secondPeptide = 0; secondPeptide < sortedSumScore.Count; ++secondPeptide)
{
SortedList <int, int> secondDict = GetSiteDict(myPositionsList[
sortedSumScore[secondPeptide].Index]);
bool othersMatch = true;
if (!secondDict.ContainsKey(siteInfo.Keys[indSite]))
{
List <int> sites = siteInfo.Keys.ToList();
for (int i = 0; i < sites.Count; i++)
{
if (i != indSite)
{
othersMatch = othersMatch && secondDict.ContainsKey(sites[i]);
}
}
if (othersMatch)
{
AScorePeptide.Add(myPositionsList[sortedSumScore[secondPeptide].Index]);
break;
}
}
else
{
if (secondDict[siteInfo.Keys[indSite]] != siteInfo.Values[indSite])
{
AScorePeptide.Add(myPositionsList[sortedSumScore[secondPeptide].Index]);
break;
}
}
}
if (secondPeptide == sortedSumScore.Count)
{
continue;
}
int[] secondTopPeptidePTMsites = myPositionsList[sortedSumScore[secondPeptide].Index];
// Get the second best scoring spectra
Dictionary <int, ChargeStateIons> secondIonWriter = new Dictionary <int, ChargeStateIons>();
List <double> secondTopTheoIons = GetChargeList(ascoreParameters, mzmax, mzmin, theoMono, theoAve, secondTopPeptidePTMsites,
out secondIonWriter);
// Calculate the diff score between the top and second sites
List <ValueIndexPair <double> > diffScore = new List <ValueIndexPair <double> >();
for (int i = 0; i < peptideScores[0].Count; ++i)
{
diffScore.Add(new ValueIndexPair <double>(
peptideScores[sortedSumScore[0].Index][i] -
peptideScores[sortedSumScore[secondPeptide].Index][i], i));
}
// Sort in descending order
diffScore.Sort(new ValueIndexPair <double> .SortValueDescend());
// Find the peak depth for the diff score
int peakDepthForAScore = 1;
if (diffScore[0].Value > 0)
{
peakDepthForAScore = diffScore[0].Index + 1;
}
DataToExcelPractice.InfoForIons secondBestIonReport = new DataToExcelPractice.InfoForIons();
secondBestIonReport.Sequence = GenerateFinalSequences(sequence, ascoreParameters, secondTopPeptidePTMsites);
secondBestIonReport.ChargeState = chargeState;
foreach (int k in secondIonWriter.Keys)
{
secondBestIonReport.BionData.Add(k, new List <double>(secondIonWriter[k].BIons));
secondBestIonReport.YionData.Add(k, new List <double>(secondIonWriter[k].YIons));
}
List <double> sTopIonIntensity = new List <double>();
secondBestIonReport.MatchesForThisDepth = GetMatchedMZStoreIntensity(10, ascoreParameters.FragmentMassTolerance, secondTopTheoIons,
expSpec.GetPeakDepthSpectra(10), out sTopIonIntensity);
secondBestIonReport.MatchedIonIntensity = new List <double>(sTopIonIntensity);
secondBestIonReport.Depth = peakDepthForAScore;
secondBestIonReport.InitializeSequenceList(modChars.ToArray());
secondBestIonReport.ScanNumber = scanNumber;
List <double> siteIons1 = GetSiteDeterminingIons(topTheoIons, secondTopTheoIons);
List <double> siteIons2 = GetSiteDeterminingIons(secondTopTheoIons, topTheoIons);
List <ExperimentalSpectraEntry> peakDepthSpectraFinal = expSpec.GetPeakDepthSpectra(peakDepthForAScore);
List <double> matched1 = GetMatchedMZ(peakDepthForAScore,
ascoreParameters.FragmentMassTolerance, siteIons1, peakDepthSpectraFinal);
List <double> matched2 = GetMatchedMZ(peakDepthForAScore,
ascoreParameters.FragmentMassTolerance, siteIons2, peakDepthSpectraFinal);
int bestDterminingCount = matched1.Count;
double a1 = PeptideScoresManager.GetPeptideScore(((double)peakDepthForAScore * ascoreParameters.FragmentMassTolerance * 2) / 100,
siteIons1.Count, matched1.Count);
double a2 = PeptideScoresManager.GetPeptideScore(((double)peakDepthForAScore * ascoreParameters.FragmentMassTolerance * 2) / 100,
siteIons2.Count, matched2.Count);
bestIonReport.AddToSiteDeterminingIons(matched1);
secondBestIonReport.SiteDeterminingIons = new List <double>(matched2);
secondBestIonReport.AScore = Math.Abs(a1 - a2);
otherSites.Add(secondBestIonReport);
// Add the results to the list
vecAScore.Add(Math.Abs(a1 - a2));
vecNumSiteIons.Add(siteIons1.Count);
siteDetermineMatched.Add(bestDterminingCount);
}
DataToExcelPractice.PeptideIonGroup myIonGroup = new DataToExcelPractice.PeptideIonGroup(bestIonReport, otherSites);
DataToExcelPractice.WriteIonsToExcel myExcel = new DataToExcelPractice.WriteIonsToExcel();
myExcel.PrintIonsToExcel(myIonGroup, System.IO.Path.GetDirectoryName(outputFilePath));
//Put scores into our table
dt.Rows[t]["BestSequence"] = front + "." + GenerateFinalSequences(sequence, ascoreParameters, topPeptidePTMsites) + "." + back;
dt.Rows[t]["PeptideScore"] = "" + topPeptideScore;
for (int i = 0; i < vecAScore.Count && i < 3; i++)
{
dt.Rows[t]["AScore" + (i + 1)] = "" + vecAScore[i];
dt.Rows[t]["numSiteIonsPoss" + (i + 1)] = vecNumSiteIons[i];
dt.Rows[t]["numSiteIonsMatched" + (i + 1)] = "" + siteDetermineMatched[i];
dt.Rows[t]["SecondSequence" + (i + 1)] = front + "." + GenerateFinalSequences(sequence, ascoreParameters, AScorePeptide[i]) + "." + back;
}
}
else if (chargeState > 1)
{
List <double> weightedScores = new List <double>();
TheoreticalSpectra theo = new TheoreticalSpectra(sequence, ascoreParameters, chargeState, new List <Mod.DynamicModification>(), MassType.Monoisotopic);
foreach (int[] myPositions in myPositionsList)
{
Dictionary <int, ChargeStateIons> mySpectra = theo.GetTempSpectra(myPositions, ascoreParameters.DynamicMods, MassType.Monoisotopic);
List <double> myIons = GetCurrentComboTheoreticalIons(mzmax, mzmin, mySpectra);
for (int peakDepth = 1; peakDepth < 11; ++peakDepth)
{
List <ExperimentalSpectraEntry> peakDepthSpectra = expSpec.GetPeakDepthSpectra(peakDepth);
List <double> matchedIons = GetMatchedMZ(
peakDepth, ascoreParameters.FragmentMassTolerance,
myIons, peakDepthSpectra);
//Adjusted peptide score to score based on tolerance window.
double score = PeptideScoresManager.GetPeptideScore(
((double)peakDepth * ascoreParameters.FragmentMassTolerance * 2) / 100.0, myIons.Count, matchedIons.Count);
// Check if there were any negative scores
weightedScores.Add(
score * ScoreWeights[peakDepth - 1]);
}
}
double pScore = 0.0;
for (int depth = 0; depth < weightedScores.Count; ++depth)
{
pScore += weightedScores[depth];
}
//Nothing to calculate
dt.Rows[t]["BestSequence"] = peptideSeq;
dt.Rows[t]["PeptideScore"] = "" + pScore;
if (myPositionsList[0].Count(i => i > 0) > 0)
{
dt.Rows[t]["AScore1"] = "1000";
}
else
{
dt.Rows[t]["AScore1"] = "-1";
}
dt.Rows[t]["numSiteIonsMatched1"] = 0;
dt.Rows[t]["numSiteIonsPoss1"] = 0;
dt.Rows[t]["SecondSequence1"] = "--";
}
else
{
dt.Rows[t]["BestSequence"] = peptideSeq;
dt.Rows[t]["PeptideScore"] = "0.0";
dt.Rows[t]["AScore1"] = "-1";
dt.Rows[t]["numSiteIonsMatched1"] = 0;
dt.Rows[t]["numSiteIonsPoss1"] = 0;
dt.Rows[t]["SecondSequence1"] = "--";
}
}
Utilities.WriteDataTableToText(dt, outputFilePath);
}
