public void CountMatchedScansPerProtein()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
const int minTagLength = 6;
var proteinToScan = new Dictionary <string, HashSet <int> >();
const string fastaFilePath = @"D:\Research\Data\CommonContaminants\H_sapiens_Uniprot_SPROT_2013-05-01_withContam.fasta";
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fastaDb = new FastaDatabase(fastaFilePath);
var searchableDb = new SearchableDatabase(fastaDb);
Console.WriteLine(@"Sequence length: {0}", fastaDb.GetSequence().Length);
//const string tagFilePath = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3_seqtag.tsv";
//const string tagFilePath = @"\\protoapps\UserData\Jungkap\Co_culture\23B_pellet_TD_3Feb14_Bane_PL011402.seqtag";
const string tagFilePath = @"D:\MassSpecFiles\co_culture\23A_pellet_TD_3Feb14_Bane_PL011402.seqtag";
if (!File.Exists(tagFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, tagFilePath);
}
var isHeader = true;
var numMatchedPairs = 0;
foreach (var line in File.ReadAllLines(tagFilePath))
{
if (isHeader)
{
isHeader = false;
continue;
}
var token = line.Split('\t');
if (token.Length != 3)
{
continue;
}
var scan = Convert.ToInt32(token[0]);
var tag = token[1];
if (tag.Length < minTagLength)
{
continue;
}
foreach (var matchedProtein in searchableDb.FindAllMatchedSequenceIndices(tag)
.Select(index => fastaDb.GetProteinName(index)))
{
++numMatchedPairs;
HashSet <int> matchedScans;
if (proteinToScan.TryGetValue(matchedProtein, out matchedScans))
{
matchedScans.Add(scan);
}
else
{
matchedScans = new HashSet <int> {
scan
};
proteinToScan.Add(matchedProtein, matchedScans);
}
}
}
var numMatchedProteins = proteinToScan.Keys.Count;
var numAllProteins = fastaDb.GetNumEntries();
Console.WriteLine("NumAllProteins: {0}", numAllProteins);
Console.WriteLine("NumMatchedProteins: {0}", numMatchedProteins);
Console.WriteLine("AvgMatchedScansPerProtein: {0}", numMatchedPairs / (float)numAllProteins);
}
public void TestFasta()
{
var db = new FastaDatabase(@"\\protoapps\UserData\Jungkap\Lewy\db\ID_005140_7A170668.fasta");
Console.WriteLine(db.GetNumEntries());
}
public void WriteResultsToMzid(IEnumerable <DatabaseSearchResultData> matches, string outputFilePath)
{
var datasetName = Path.GetFileNameWithoutExtension(outputFilePath);
var creator = new IdentDataCreator("MSPathFinder_" + datasetName, "MSPathFinder_" + datasetName);
var soft = creator.AddAnalysisSoftware("Software_1", "MSPathFinder", System.Reflection.Assembly.GetCallingAssembly().GetName().Version.ToString(), CV.CVID.MS_MSPathFinder, "MSPathFinder");
var settings = creator.AddAnalysisSettings(soft, "Settings_1", CV.CVID.MS_ms_ms_search);
var searchDb = creator.AddSearchDatabase(database.GetFastaFilePath(), database.GetNumEntries(), Path.GetFileNameWithoutExtension(database.GetFastaFilePath()), CV.CVID.CVID_Unknown,
CV.CVID.MS_FASTA_format);
if (options.TargetDecoySearchMode.HasFlag(DatabaseSearchMode.Decoy))
{
searchDb.CVParams.AddRange(new CVParamObj[]
{
new CVParamObj()
{
Cvid = CV.CVID.MS_DB_composition_target_decoy,
},
new CVParamObj()
{
Cvid = CV.CVID.MS_decoy_DB_accession_regexp, Value = "^XXX",
},
//new CVParamObj() { Cvid = CV.CVID.MS_decoy_DB_type_reverse, },
new CVParamObj()
{
Cvid = CV.CVID.MS_decoy_DB_type_randomized,
},
});
}
// store the settings...
CreateMzidSettings(settings);
var path = options.SpecFilePath;
var run = lcmsRun as PbfLcMsRun;
if (run != null)
{
var rawPath = run.RawFilePath;
if (!string.IsNullOrWhiteSpace(rawPath))
{
path = rawPath;
}
}
// TODO: fix this to match correctly to the original file - May need to modify the PBF format to add an input format specifier
// TODO: Should probably? request a CV Term for the PBF format?
var nativeIdFormat = lcmsRun.NativeIdFormat;
if (nativeIdFormat == CV.CVID.CVID_Unknown)
{
nativeIdFormat = CV.CVID.MS_scan_number_only_nativeID_format;
}
var specData = creator.AddSpectraData(path, datasetName, nativeIdFormat, lcmsRun.NativeFormat);
// Get the search modifications as they were passed into the AminoAcidSet constructor, so we can retrieve masses from them
var modDict = new Dictionary <string, Modification>();
foreach (var mod in options.AminoAcidSet.SearchModifications)
{
if (!modDict.ContainsKey(mod.Modification.Name))
{
modDict.Add(mod.Modification.Name, mod.Modification);
}
else if (!modDict[mod.Modification.Name].Composition.Equals(mod.Modification.Composition))
{
throw new System.Exception(
"ERROR: Cannot have modifications with the same name and different composition/mass! Fix input modifications! Duplicated modification name: " +
mod.Modification.Name);
}
}
foreach (var match in matches)
{
var scanNum = match.ScanNum;
var spec = lcmsRun.GetSpectrum(scanNum, false);
var matchIon = new Ion(Composition.Parse(match.Composition), match.Charge);
var nativeId = spec.NativeId;
if (string.IsNullOrWhiteSpace(spec.NativeId))
{
nativeId = "scan=" + spec.ScanNum;
}
var specIdent = creator.AddSpectrumIdentification(specData, nativeId, spec.ElutionTime, match.MostAbundantIsotopeMz,
match.Charge, 1, double.NaN);
specIdent.CalculatedMassToCharge = matchIon.GetMonoIsotopicMz();
var pep = new PeptideObj(match.Sequence);
var modText = match.Modifications;
if (!string.IsNullOrWhiteSpace(modText))
{
var mods = modText.Split(',');
foreach (var mod in mods)
{
var tokens = mod.Split(' ');
var modInfo = modDict[tokens[0]];
var modObj = new ModificationObj(CV.CVID.MS_unknown_modification, modInfo.Name, int.Parse(tokens[1]), modInfo.Mass);
pep.Modifications.Add(modObj);
}
}
specIdent.Peptide = pep;
var proteinName = match.ProteinName;
var protLength = match.ProteinLength;
var proteinDescription = match.ProteinDescription;
var dbSeq = new DbSequenceObj(searchDb, protLength, proteinName, proteinDescription);
var start = match.Start;
var end = match.End;
var pepEv = new PeptideEvidenceObj(dbSeq, pep, start, end, match.Pre, match.Post, match.ProteinName.StartsWith("XXX"));
specIdent.AddPeptideEvidence(pepEv);
var probability = match.Probability;
specIdent.CVParams.Add(new CVParamObj()
{
Cvid = CV.CVID.MS_chemical_compound_formula, Value = match.Composition,
});
//specIdent.CVParams.Add(new CVParamObj() { Cvid = CV.CVID.MS_number_of_matched_peaks, Value = match.NumMatchedFragments.ToString(), });
specIdent.CVParams.Add(new CVParamObj()
{
Cvid = CV.CVID.MS_MSPathFinder_RawScore, Value = probability.ToString(CultureInfo.InvariantCulture),
});
specIdent.CVParams.Add(new CVParamObj()
{
Cvid = CV.CVID.MS_MSPathFinder_SpecEValue, Value = match.SpecEValue.ToString(CultureInfo.InvariantCulture),
});
specIdent.CVParams.Add(new CVParamObj()
{
Cvid = CV.CVID.MS_MSPathFinder_EValue, Value = match.EValue.ToString(CultureInfo.InvariantCulture),
});
if (match.HasTdaScores)
{
specIdent.CVParams.Add(new CVParamObj()
{
Cvid = CV.CVID.MS_MSPathFinder_QValue, Value = match.QValue.ToString(CultureInfo.InvariantCulture),
});
specIdent.CVParams.Add(new CVParamObj()
{
Cvid = CV.CVID.MS_MSPathFinder_PepQValue, Value = match.PepQValue.ToString(CultureInfo.InvariantCulture),
});
}
// MS-GF+ similarity: find/add isotope error?
// MS-GF+ similarity: find/add assumed dissociation method?
//specIdent.UserParams.Add(new UserParamObj() {Name = "Assumed Dissociation Method", Value = match.});
}
var identData = creator.GetIdentData();
MzIdentMlReaderWriter.Write(new MzIdentMLType(identData), outputFilePath);
}
private void WriteResultsToFile(DatabaseSequenceSpectrumMatch[] matches, string outputFilePath, FastaDatabase database)
{
using (var writer = new StreamWriter(outputFilePath))
{
writer.WriteLine("Scan\tPre\tSequence\tPost\tModifications\tComposition\tProteinName\tProteinDesc" +
"\tProteinLength\tStart\tEnd\tCharge\tMostAbundantIsotopeMz\tMass\t#MatchedFragments\tProbability\tSpecEValue\tEValue");
foreach (var scanNum in _ms2ScanNums)
{
var match = matches[scanNum];
if (match == null)
{
continue;
}
var sequence = match.Sequence;
var offset = match.Offset;
var start = database.GetOneBasedPositionInProtein(offset) + 1 + match.NumNTermCleavages;
var end = start + sequence.Length - 1;
var proteinName = database.GetProteinName(match.Offset);
var protLength = database.GetProteinLength(proteinName);
var ion = match.Ion;
var proteinDescription = database.GetProteinDescription(match.Offset);
var probability = CompositeScorer.GetProbability(match.Score);
// Note for DblToString(value, 9, true), by having "9" and "true",
// values between 100 and 999 Da will have 7 digits after the decimal place, and
// values between 1000 and 9999 will have 6 digits after the decimal place
writer.WriteLine("{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}\t{7}\t{8}\t{9}\t{10}\t{11}\t{12}\t{13}\t{14}\t{15}\t{16}\t{17}",
scanNum,
match.Pre, // Pre
sequence, // Sequence
match.Post, // Post
match.ModificationText, // Modifications
ion.Composition, // Composition
proteinName, // ProteinName
proteinDescription, // ProteinDescription
protLength, // ProteinLength
start, // Start position in protein
end, // End position in protein
ion.Charge, // precursorCharge
StringUtilities.DblToString(ion.GetMostAbundantIsotopeMz(), 9, true), // MostAbundantIsotopeMz
StringUtilities.DblToString(ion.Composition.Mass, 9, true), // Mass
match.NumMatchedFragments, // (Number of matched fragments)
StringUtilities.DblToString(probability, 4), // Probability
StringUtilities.DblToString(ExcelMinValue(match.SpecEvalue), 6, true, 0.001), // EValue; will be displayed using scientific notation if the value is less than 0.001
StringUtilities.DblToString(ExcelMinValue(match.SpecEvalue * database.GetNumEntries()), 6, true, 0.001) // SpecEValue; will be displayed using scientific notation if the value is less than 0.001
);
}
}
}