public void TestConvoluteMass(double massMz, int currentCharge, int newCharge, double expectedMz)
{
var newMz = mPeptideMassCalculator.ConvoluteMass(massMz, currentCharge, newCharge);
var newMzAlt = mPeptideMassCalculator.ConvoluteMass(massMz, currentCharge, newCharge, PeptideMassCalculator.MASS_PROTON);
Console.WriteLine("{0} from {1}+ to {2}+ is {3:F5}; expected {4:F5}", massMz, currentCharge, newCharge, newMz, expectedMz);
Assert.AreEqual(newMz, newMzAlt, 1E-05, "The two overloads of ConvoluteMass reported conflicting mass values");
Assert.AreEqual(expectedMz, newMz, 0.0001, "Unexpected convoluted m/z");
}
/// <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();
}
private static void TestPHRPReader(string synOrFhtFile, bool blnSkipDuplicates)
{
var inputFile = new FileInfo(synOrFhtFile);
Console.WriteLine("Instantiating reader");
var startupOptions = new StartupOptions
{
LoadModsAndSeqInfo = true,
LoadMSGFResults = true,
LoadScanStatsData = false,
MaxProteinsPerPSM = 100
};
var phrpReader =
new ReaderFactory(inputFile.FullName, PeptideHitResultTypes.Unknown, startupOptions)
{
EchoMessagesToConsole = false,
SkipDuplicatePSMs = blnSkipDuplicates
};
// Check for any load errors
if (phrpReader.ErrorMessages.Count > 0)
{
Console.WriteLine("Error(s) instantiating the reader:");
foreach (var errorMessage in phrpReader.ErrorMessages)
{
Console.WriteLine(" " + errorMessage);
}
}
phrpReader.ErrorEvent += ErrorEventHandler;
phrpReader.StatusEvent += MessageEventHandler;
phrpReader.WarningEvent += WarningEventHandler;
const bool fastReadEnabled = true;
phrpReader.FastReadMode = fastReadEnabled;
var massCalculator = new PeptideMassCalculator();
if (!phrpReader.CanRead)
{
Console.WriteLine("Aborting since PHRPReader is not ready: " + phrpReader.ErrorMessage);
return;
}
var lstValues = new List <string>();
var intPSMsRead = 0;
var intModifiedPSMsRead = 0;
// ReSharper disable once CollectionNeverQueried.Local
var dctCachedValues = new Dictionary <int, PSM>();
Console.WriteLine("Reading data");
while (phrpReader.MoveNext())
{
var psm = phrpReader.CurrentPSM;
intPSMsRead += 1;
lstValues.Clear();
phrpReader.FinalizeCurrentPSM();
PeptideCleavageStateCalculator.SplitPrefixAndSuffixFromSequence(psm.Peptide, out _, out _, out _);
var strMassErrorPPM = GetCorrectedMassErrorPPM(psm, out _);
lstValues.Add(phrpReader.DatasetName + "_dta.txt"); // #SpecFile
lstValues.Add("index=" + intPSMsRead); // SpecID
lstValues.Add(psm.ScanNumber.ToString()); // ScanNum
lstValues.Add(psm.CollisionMode); // FragMethod
lstValues.Add(massCalculator.ConvoluteMass(psm.PrecursorNeutralMass, 0, psm.Charge).ToString(CultureInfo.InvariantCulture)); // Precursor m/z
lstValues.Add(strMassErrorPPM); // PrecursorError(ppm)
lstValues.Add(psm.Charge.ToString()); // Charge
lstValues.Add(psm.NumTrypticTermini.ToString()); // Tryptic state (0, 1, or 2)
lstValues.Add(CleanupPeptide(psm.PeptideWithNumericMods)); // Peptide
if (psm.SeqID <= 0)
{
lstValues.Add("**" + psm.SeqID + "**"); // SeqID is undefined
}
else
{
lstValues.Add(psm.SeqID.ToString()); // SeqID
}
lstValues.Add(psm.ProteinFirst);
if (psm.ProteinDetails.Count > 0)
{
var firstProteinDetail = psm.ProteinDetails.First();
if (!string.Equals(psm.ProteinFirst, firstProteinDetail.Key))
{
lstValues.Add(firstProteinDetail.Key);
}
else
{
lstValues.Add("<Match>");
}
lstValues.Add(firstProteinDetail.Value.ResidueStart.ToString());
lstValues.Add(firstProteinDetail.Value.ResidueEnd.ToString());
}
var strXCorr = GetScore(psm, SequestSynFileReader.GetColumnNameByID(SequestSynopsisFileColumns.XCorr), "0");
lstValues.Add(strXCorr);
lstValues.Add(GetScore(psm, SequestSynFileReader.GetColumnNameByID(SequestSynopsisFileColumns.Sp), "0"));
lstValues.Add(psm.MSGFSpecEValue);
lstValues.Add(GetScore(psm, SequestSynFileReader.GetColumnNameByID(SequestSynopsisFileColumns.DeltaCn2), "0"));
lstValues.Add(GetScore(psm, MSGFPlusSynFileReader.GetMSGFDBColumnNameByID(MSGFDBSynFileColumns.PValue), "0"));
lstValues.Add(GetScore(psm, MSGFPlusSynFileReader.GetColumnNameByID(MSGFPlusSynFileColumns.EValue), "0"));
lstValues.Add(GetScore(psm, MSGFPlusSynFileReader.GetColumnNameByID(MSGFPlusSynFileColumns.RankSpecEValue), "0"));
lstValues.Add(GetScore(psm, MSGFPlusSynFileReader.GetMSGFDBColumnNameByID(MSGFDBSynFileColumns.FDR), "1"));
lstValues.Add(GetScore(psm, MSGFPlusSynFileReader.GetColumnNameByID(MSGFPlusSynFileColumns.QValue), "0"));
lstValues.Add(GetScore(psm, MSGFPlusSynFileReader.GetColumnNameByID(MSGFPlusSynFileColumns.PepQValue), "0"));
if (psm.PeptideCleanSequence == "QQIEESTSDYDKEK")
{
Console.WriteLine(psm.Peptide + " in scan " + psm.ScanNumber);
var parentIonMZ = massCalculator.ConvoluteMass(psm.PrecursorNeutralMass, 0, psm.Charge);
Console.WriteLine("ParentIonMZ = " + parentIonMZ);
Console.WriteLine("PeptideWithNumericMods = " + psm.PeptideWithNumericMods);
}
if (psm.ModifiedResidues.Count > 0)
{
intModifiedPSMsRead += 1;
if (intModifiedPSMsRead % 500 == 0)
{
Console.WriteLine("PeptideWithNumericMods = " + psm.PeptideWithNumericMods);
foreach (var modifiedResidue in psm.ModifiedResidues)
{
Console.WriteLine(" " + modifiedResidue.Residue + modifiedResidue.EndResidueLocInPeptide + ": " + modifiedResidue.ModDefinition.ModificationMassAsText);
}
}
var dblPeptideMassRecomputed = massCalculator.ComputeSequenceMassNumericMods(psm.PeptideWithNumericMods);
if (Math.Abs(psm.PeptideMonoisotopicMass - dblPeptideMassRecomputed) > 0.1)
{
Console.WriteLine(" Peptide mass disagreement: " + (psm.PeptideMonoisotopicMass - dblPeptideMassRecomputed).ToString("0.0000000"));
}
}
var strFlattened = FlattenList(lstValues);
if (intPSMsRead % 10000 == 0)
{
Console.WriteLine(strFlattened);
}
dctCachedValues.Add(intPSMsRead, psm);
}
}