public FeatureBasedTagSearchEngine(
LcMsRun run,
Ms1FtParser featureParser,
ProductScorerBasedOnDeconvolutedSpectra ms2Scorer,
SequenceTagParser tagParser,
FastaDatabase fastaDb,
Tolerance tolerance,
AminoAcidSet aaSet,
double maxSequenceMass = 50000.0,
int minProductIonCharge = 1,
int maxProductIonCharge = 20)
{
_run = run;
_ms2Scorer = ms2Scorer;
_featureParser = featureParser;
_ms1FtFilter = new Ms1FtFilter(run, tolerance, featureParser.Ms1FtFileName);
_tagParser = tagParser;
_fastaDb = fastaDb;
_searchableDb = new SearchableDatabase(fastaDb);
_tolerance = tolerance;
_aaSet = aaSet;
_maxSequenceMass = maxSequenceMass;
_minProductIonCharge = minProductIonCharge;
_maxProductIonCharge = maxProductIonCharge;
}
public void TestSequenceEnumerationParallel2()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
var sw = new System.Diagnostics.Stopwatch();
const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_002216_235ACCEA.fasta";
var db = new FastaDatabase(dbFile);
db.Read();
var indexedDb = new IndexedDatabase(db);
var arr = db.Characters().ToArray();
sw.Start();
//var annotationsAndOffsets = indexedDb.AnnotationsAndOffsetsNoEnzyme(7, 30);
// var num = annotationsAndOffsets.AsParallel().LongCount(annotationsAndOffset => annotationsAndOffset.Annotation.IndexOf('W') >= 0);
//var num = annotationsAndOffsets.LongCount(annotationsAndOffset => annotationsAndOffset.Annotation.IndexOf('W') >= 0);
//var num = arr.AsParallel().Where(c => c == 'W').LongCount();
var num = 0;
var sum = 0L;
//foreach (var c in arr)
for (var a = 0; a < arr.Length; a++)
{
var c = arr[a];
for (var i = 0; i < c * 10000; i++) sum += i;
// Interlocked.Increment(ref num);
if (++num == 1000) break;
}
Console.WriteLine("NumPeptides: {0}", sum);
sw.Stop();
Console.WriteLine(@"{0:f4} sec", sw.Elapsed.TotalSeconds);
}
private static void TestCountingPeptides()
{
var aaSet = new AminoAcidSet();
var sw = new Stopwatch();
sw.Start();
//const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_002166_F86E3B2F.fasta";
const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_003456_9B916A8B.fasta";
// const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004208_295531A4.fasta";
var db = new FastaDatabase(dbFile);
var indexedDb = new IndexedDatabase(db);
indexedDb.Read();
//var numPeptides = indexedDb.AnnotationsAndOffsetsNoEnzyme(7, 150).LongCount();
var peptides =
indexedDb.AnnotationsAndOffsets(7, 40, 2, 2, Enzyme.Trypsin);
Parallel.ForEach(peptides, annotationAndOffset =>
//foreach(var annotationAndOffset in peptides)
{
var annotation = annotationAndOffset.Annotation;
var offset = annotationAndOffset.Offset;
var graph = SequenceGraph.CreateGraph(aaSet, annotation);
}
)
;
// Console.WriteLine("NumPeptides: {0}", numPeptides);
sw.Stop();
var sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"{0:f4} sec", sec);
}
public void TestSearching()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
var sw = new System.Diagnostics.Stopwatch();
sw.Start();
const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\Short.fasta";
if (!File.Exists(dbFile))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, dbFile);
}
var db = new FastaDatabase(dbFile);
var searchableDb = new SearchableDatabase(db);
//const string pattern = "NSGSHFCGGSLINSQWVVSAAH";
const string pattern = "FPTDDDDK";
var position = searchableDb.Search(pattern);
Assert.True(position >= 0);
Console.WriteLine("Position: {0}", position);
Console.WriteLine("Matched indices: {0}", string.Join(",", searchableDb.FindAllMatchedSequenceIndices(pattern)));
Console.WriteLine("Protein indices: {0}", string.Join(",", searchableDb.FindAllMatchedSequenceIndices(pattern).Select(i => db.GetOneBasedPositionInProtein(i))));
sw.Stop();
Console.WriteLine(@"{0:f4} sec", sw.Elapsed.TotalSeconds);
}
public ScanBasedTagSearchEngine(
LcMsRun run,
ISequenceTagFinder seqTagFinder,
LcMsPeakMatrix featureFinder,
FastaDatabase fastaDb,
Tolerance tolerance,
AminoAcidSet aaSet,
CompositeScorerFactory ms2ScorerFactory = null,
int minMatchedTagLength = DefaultMinMatchedTagLength,
double maxSequenceMass = 50000.0,
int minProductIonCharge = 1,
int maxProductIonCharge = 20)
{
_run = run;
_featureFinder = featureFinder;
_searchableDb = new SearchableDatabase(fastaDb);
_tolerance = tolerance;
_aaSet = aaSet;
_minMatchedTagLength = minMatchedTagLength;
_maxSequenceMass = maxSequenceMass;
_minProductIonCharge = minProductIonCharge;
_maxProductIonCharge = maxProductIonCharge;
MinScan = int.MinValue;
MaxScan = int.MaxValue;
_ms2ScorerFactory = ms2ScorerFactory;
_seqTagFinder = seqTagFinder;
}
public IndexedDatabase(FastaDatabase fastaDatabase)
{
FastaDatabase = fastaDatabase;
var databaseFilePath = FastaDatabase.GetFastaFilePath();
//var databaseFilePathNoExt = Path.Combine(Path.GetDirectoryName(databaseFilePath), Path.GetFileNameWithoutExtension(databaseFilePath));
//_pLcpFilePath = databaseFilePathNoExt + PermutedLongestCommonPrefixFileExtension;
_pLcpFilePath = Path.ChangeExtension(databaseFilePath, PermutedLongestCommonPrefixFileExtension);
var lastWriteTimeHash = FastaDatabase.GetLastWriteTimeHash();
if (!File.Exists(_pLcpFilePath) || !FastaDatabase.CheckHashCodeBinaryFile(_pLcpFilePath, lastWriteTimeHash))
{
Console.Write("Generating " + _pLcpFilePath + " ... ");
CreatePermutedLongestCommonPrefixFile();
Console.WriteLine("Done");
}
}
public ProteoformSpectrumMatchContainer(FastaDatabase database, int[] ms2ScanVector, int maxModifications, int maxNumMatchesPerSpectrum, int minScore = 4)
{
Database = database;
NumMatchesPerSpectrum = maxNumMatchesPerSpectrum;
_scoreCutoff = minScore;
Ms2ScanVector = ms2ScanVector;
_ms2ScanToIndexMap = new int[ms2ScanVector.Last() + 1];
for(var i = 0; i < ms2ScanVector.Length; i++)
{
var scanNum = ms2ScanVector[i];
_ms2ScanToIndexMap[scanNum] = i;
}
_matchedSet = new SortedSet<DatabaseSequenceSpectrumMatch>[maxModifications + 1][];
for(var i = 0; i <= maxModifications; i++) _matchedSet[i] = new SortedSet<DatabaseSequenceSpectrumMatch>[ms2ScanVector.Length];
_checkedOutScanNumbers = new List<int>();
}
/// <summary>
/// Constructor - build the index
/// </summary>
/// <param name="fastaDatabase"></param>
public IndexedDatabase(FastaDatabase fastaDatabase)
{
FastaDatabase = fastaDatabase;
var databaseFilePath = FastaDatabase.GetFastaFilePath();
//var databaseFilePathNoExt = Path.Combine(Path.GetDirectoryName(databaseFilePath), Path.GetFileNameWithoutExtension(databaseFilePath));
//_pLcpFilePath = databaseFilePathNoExt + PermutedLongestCommonPrefixFileExtension;
_pLcpFilePath = Path.ChangeExtension(databaseFilePath, PermutedLongestCommonPrefixFileExtension);
var lastWriteTimeHash = FastaDatabase.GetLastWriteTimeHash();
if (!File.Exists(_pLcpFilePath) || !FastaDatabase.CheckHashCodeBinaryFile(_pLcpFilePath, lastWriteTimeHash))
{
Console.Write("Generating " + _pLcpFilePath + " ... ");
CreatePermutedLongestCommonPrefixFile();
Console.WriteLine("Done");
}
}
public void TestSearchWithTagGeneration()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string rawFilePath = @"D:\MassSpecFiles\training\raw\QC_Shew_Intact_26Sep14_Bane_C2Column3.pbf";
if (!File.Exists(rawFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(rawFilePath);
const string fastaFilePath = @"D:\MSPathFinder\Fasta\ID_002216_235ACCEA.fasta";
//const string fastaFilePath = @"D:\MassSpecFiles\60k\ID_004973_9BA6912F.fasta";
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fastaDb = new FastaDatabase(fastaFilePath);
var tolerance = new Tolerance(10);
var modsFilePath = @"D:\MSPathFinder\Fasta\Mods.txt";
if (!File.Exists(modsFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, modsFilePath);
}
var aaSet = new AminoAcidSet(modsFilePath);
//TestTagBasedSearch(run, fastaDb, tolerance, aaSet);
var tagSearchEngine = new ScanBasedTagSearchEngine(run, new SequenceTagGenerator(run, new Tolerance(8)), new LcMsPeakMatrix(run), fastaDb, tolerance,aaSet);
var matchedTags = tagSearchEngine.RunSearch(4672);
foreach (var match in matchedTags)
{
Console.Write(match.Sequence);
Console.WriteLine("\t{0}\t{1}\t{2}", match.TagMatch.StartIndex, match.TagMatch.EndIndex, match.TagMatch.Mass);
}
}
public SearchableDatabase(FastaDatabase fastaDatabase)
{
FastaDatabase = fastaDatabase;
_sequence = fastaDatabase.GetSequence();
_suffixArray = new int[_sequence.Length];
SAIS.sufsort(_sequence, _suffixArray, _sequence.Length);
var neighboringLcps = new byte[_suffixArray.Length];
neighboringLcps[0] = 0;
for (var i = 1; i < _suffixArray.Length; i++)
{
var lcp = IndexedDatabase.GetLcp(_sequence, _suffixArray[i - 1], _suffixArray[i]);
neighboringLcps[i] = lcp;
}
_leftLcps = new byte[_suffixArray.Length];
_rightLcps = new byte[_suffixArray.Length];
InitializeLcps(neighboringLcps, _leftLcps, _rightLcps, 0, _suffixArray.Length-1);
}
/// <summary>
/// Constructor
/// </summary>
/// <param name="fastaDatabase"></param>
public SearchableDatabase(FastaDatabase fastaDatabase)
{
FastaDatabase = fastaDatabase;
_sequence = fastaDatabase.GetSequence();
_suffixArray = new int[_sequence.Length];
SAIS.sufsort(_sequence, _suffixArray, _sequence.Length);
var neighboringLcps = new byte[_suffixArray.Length];
neighboringLcps[0] = 0;
for (var i = 1; i < _suffixArray.Length; i++)
{
var lcp = IndexedDatabase.GetLcp(_sequence, _suffixArray[i - 1], _suffixArray[i]);
neighboringLcps[i] = lcp;
}
_leftLcps = new byte[_suffixArray.Length];
_rightLcps = new byte[_suffixArray.Length];
InitializeLcps(neighboringLcps, _leftLcps, _rightLcps, 0, _suffixArray.Length - 1);
}
public FeatureBasedTagSearchEngine(
LcMsRun run,
Ms1FtParser featureParser,
SequenceTagParser tagParser,
FastaDatabase fastaDb,
Tolerance tolerance,
AminoAcidSet aaSet,
double maxSequenceMass = 50000.0,
int minProductIonCharge = 1,
int maxProductIonCharge = 20)
: this(
run,
featureParser,
null,
tagParser,
fastaDb,
tolerance,
aaSet,
maxSequenceMass,
minProductIonCharge,
maxProductIonCharge)
{
}
public void TestSumParallel()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
//var array = Enumerable.Range(0, short.MaxValue).ToArray();
const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_002216_235ACCEA.fasta";
var db = new FastaDatabase(dbFile);
db.Read();
//var indexedDb = new IndexedDatabase(db);
//indexedDb.Read();
//var peptides = indexedDb.AnnotationsAndOffsetsNoEnzyme(7, 30);
var charArray = db.Characters().Select(c => (int)c);
// Test methods.
Console.WriteLine(SumAsParallel(charArray));
Console.WriteLine(SumDefault(charArray));
const int m = 100;
var s1 = Stopwatch.StartNew();
for (var i = 0; i < m; i++)
{
SumDefault(charArray);
}
s1.Stop();
var s2 = Stopwatch.StartNew();
for (var i = 0; i < m; i++)
{
SumAsParallel(charArray);
}
s2.Stop();
Console.WriteLine((s1.Elapsed.TotalMilliseconds * 1000000 /
m).ToString("0.00 ns"));
Console.WriteLine((s2.Elapsed.TotalMilliseconds * 1000000 /
m).ToString("0.00 ns"));
Console.Read();
}
public void TestEnumeratingProteins()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
var sw = new System.Diagnostics.Stopwatch();
sw.Start();
const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\Short.fasta";
if (!File.Exists(dbFile))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, dbFile);
}
var db = new FastaDatabase(dbFile);
var indexedDb = new IndexedDatabase(db);
foreach (var annotationAndOffset in indexedDb.IntactSequenceAnnotationsAndOffsetsWithCTermCleavagesLargerThan(100, 300, 3))
{
Console.WriteLine(annotationAndOffset.Annotation);
}
sw.Stop();
Console.WriteLine(@"{0:f4} sec", sw.Elapsed.TotalSeconds);
}
[TestCase(3, @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_005133_8491EFA2.fasta", 323719193)] // 3MB
//[TestCase(6, @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004530_B63BD900.fasta", 595227563)] // 6MB
//[TestCase(15, @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004208_295531A4.fasta", 1882434687)] // 15MB
public void TestSequenceEnumeration(double size, string dbFile, int expected)
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName, dbFile);
var sw = new System.Diagnostics.Stopwatch();
sw.Start();
var db = new FastaDatabase(dbFile);
var indexedDb = new IndexedDatabase(db);
var numSequences = 0L;
var timeDB = sw.Elapsed;
Console.WriteLine("Read DB in " + timeDB.TotalSeconds + " Seconds");
var estimatedAnnOff = indexedDb.EstimateTotalPeptides(0, 30, 250);
var timeEstimate = sw.Elapsed;
Console.WriteLine("Read Estimate in " + (timeEstimate - timeDB).TotalSeconds + " Seconds");
//int coreCount = 0;
//foreach (var item in new System.Management.ManagementObjectSearcher("Select NumberOfCores from Win32_Processor").Get())
//{
// coreCount += int.Parse(item["NumberOfCores"].ToString());
//}
//Console.WriteLine("Number Of Cores: {0}", coreCount);
//Console.WriteLine("Processors: " + System.Environment.ProcessorCount);
Console.WriteLine("Estimated results: " + estimatedAnnOff);
var annotationsAndOffsets = indexedDb.AnnotationsAndOffsetsNoEnzymeParallel(30, 250);
var timeGetAnn = sw.Elapsed;
Console.WriteLine("Read Annotations in " + (timeGetAnn - timeEstimate).TotalSeconds + " Seconds");
/*/Parallel.ForEach(
annotationsAndOffsets,
// new ParallelOptions { MaxDegreeOfParallelism = 2},
annotationAndOffset =>
{
Interlocked.Increment(ref numSequences);
//++numSequences;
}
);/**/
//annotationsAndOffsets.Select(annotationsAndOffset => annotationsAndOffset.)
// Below, original: 110, 109(total) seconds
// Parallelizing AnnotationsAndOffsetsNoEnzyme: 86 seconds
// Parallelizing AnnotationsAndOffsetsNoEnzyme, yield returns: 79.6, 94, 60, 60 seconds
//
// 3MB
// serial:
// Parallel2: 107,
//
// 6MB
// serial:
// Parallel2:
//
// 15MB
// serial:
// Parallel2:
//using (var ofstream = new FileStream(Path.Combine(@"F:\InformedProteomicsTestFiles", Path.GetFileNameWithoutExtension(dbFile) + "_par.txt"), FileMode.Create))
//using (var fout = new StreamWriter(ofstream))
//{
// foreach (var annOff in annotationsAndOffsets)
// {
// numSequences++;
// fout.WriteLine(annOff.Annotation);
// }
//}
numSequences = annotationsAndOffsets.Count();
var timeParForEach = sw.Elapsed;
Console.WriteLine("Parallel ForEach in " + (timeParForEach - timeGetAnn).TotalSeconds + " Seconds");
Console.WriteLine("NumPeptides: {0}", numSequences);
sw.Stop();
Console.WriteLine(@"{0:f4} sec", sw.Elapsed.TotalSeconds);
//Assert.AreEqual(188961836, numSequences);
Assert.AreEqual(expected, numSequences);
}
private void WriteResultsToFile(SortedSet<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\tIcScore"
);
for (var scanNum = _run.MinLcScan; scanNum <= _run.MaxLcScan; scanNum++)
{
if (matches[scanNum] == null) continue;
foreach (var match in matches[scanNum].Reverse())
{
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 scores = _bottomUpScorer.GetScores(match, ion.Composition, ion.Charge, scanNum);
if (ion == null)
{
Console.WriteLine(@"Null ion!");
}
if (scores == null)
{
Console.WriteLine(@"Null scores");
}
// 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}",
scanNum,
match.Pre,
sequence, // Sequence
match.Post,
scores.Modifications, // Modifications
ion.Composition, // Composition
proteinName, // ProteinName
database.GetProteinDescription(match.Offset), // ProteinDescription
protLength, // ProteinLength
start, // Start
end, // End
ion.Charge, // precursorCharge
StringUtilities.DblToString(ion.GetMostAbundantIsotopeMz(), 9, true), // MostAbundantIsotopeMz
StringUtilities.DblToString(ion.Composition.Mass, 9, true), // Mass
match.Score,
scores.Score // Score (re-scored)
);
}
}
}
}
private IEnumerable<AnnotationAndOffset> GetAnnotationsAndOffsets(FastaDatabase database)
{
var indexedDbTarget = new IndexedDatabase(database);
IEnumerable<AnnotationAndOffset> annotationsAndOffsets;
if (NumTolerableTermini == 0)
{
annotationsAndOffsets = indexedDbTarget.AnnotationsAndOffsetsNoEnzyme(MinSequenceLength, MaxSequenceLength);
}
else
{
annotationsAndOffsets = indexedDbTarget.AnnotationsAndOffsets(MinSequenceLength, MaxSequenceLength,
NumTolerableTermini, 2, Enzyme);
}
return annotationsAndOffsets;
}
public bool RunSearch(double corrThreshold)
{
var sw = new Stopwatch();
ErrorMessage = string.Empty;
Console.Write(@"Reading raw file...");
sw.Start();
_run = InMemoryLcMsRun.GetLcMsRun(SpecFilePath, 1.4826, 1.4826);
_bottomUpScorer = new InformedBottomUpScorer(_run, AminoAcidSet, MinProductIonCharge, MaxProductIonCharge, ProductIonTolerance);
sw.Stop();
var sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Elapsed Time: {0:f4} sec", sec);
sw.Reset();
Console.Write(@"Determining precursor masses...");
sw.Start();
var ms1Filter = new Ms1IsotopeAndChargeCorrFilter(_run, PrecursorIonTolerance, MinPrecursorIonCharge, MaxPrecursorIonCharge,
400, 5000, corrThreshold, 0, 0);
sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Elapsed Time: {0:f4} sec", sec);
sw.Reset();
Console.Write(@"Deconvoluting MS2 spectra...");
sw.Start();
_ms2ScorerFactory = new ProductScorerBasedOnDeconvolutedSpectra(
_run,
MinProductIonCharge, MaxProductIonCharge,
new Tolerance(10),
0
);
_ms2ScorerFactory.DeconvoluteAllProductSpectra();
sw.Stop();
sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Elapsed Time: {0:f4} sec", sec);
// Target database
var targetDb = new FastaDatabase(DatabaseFilePath);
// string dirName = OutputDir ?? Path.GetDirectoryName(SpecFilePath);
var baseName = Path.GetFileNameWithoutExtension(SpecFilePath);
var targetOutputFilePath = Path.Combine(OutputDir, baseName + TargetFileExtension);
var decoyOutputFilePath = Path.Combine(OutputDir, baseName + DecoyFileExtension);
var tdaOutputFilePath = Path.Combine(OutputDir, baseName + TdaFileExtension);
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Target))
{
sw.Reset();
Console.Write(@"Reading the target database...");
sw.Start();
targetDb.Read();
sw.Stop();
sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Elapsed Time: {0:f4} sec", sec);
sw.Reset();
Console.WriteLine(@"Searching the target database");
sw.Start();
var targetMatches = RunSearch(GetAnnotationsAndOffsets(targetDb), ms1Filter, false);
sw.Stop();
sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Target database search elapsed time: {0:f4} sec", sec);
sw.Reset();
Console.Write(@"Rescoring and writing target results...");
sw.Start();
WriteResultsToFile(targetMatches, targetOutputFilePath, targetDb);
sw.Stop();
sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Elapsed time: {0:f4} sec", sec);
}
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Decoy))
{
// Decoy database
sw.Reset();
Console.Write(@"Reading the decoy database...");
sw.Start();
var decoyDb = targetDb.Decoy(Enzyme);
decoyDb.Read();
sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Elapsed Time: {0:f4} sec", sec);
sw.Reset();
Console.WriteLine(@"Searching the decoy database");
sw.Start();
var decoyMatches = RunSearch(GetAnnotationsAndOffsets(decoyDb), ms1Filter, true);
sw.Stop();
sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Decoy database search elapsed Time: {0:f4} sec", sec);
sw.Reset();
Console.Write(@"Rescoring and writing decoy results...");
sw.Start();
WriteResultsToFile(decoyMatches, decoyOutputFilePath, decoyDb);
sw.Stop();
sec = sw.ElapsedTicks / (double)Stopwatch.Frequency;
Console.WriteLine(@"Elapsed time: {0:f4} sec", sec);
}
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Both))
{
var fdrCalculator = new FdrCalculator(targetOutputFilePath, decoyOutputFilePath);
if (fdrCalculator.HasError())
{
ErrorMessage = fdrCalculator.ErrorMessage;
Console.WriteLine(@"Error computing FDR: " + fdrCalculator.ErrorMessage);
return false;
}
fdrCalculator.WriteTo(tdaOutputFilePath);
}
Console.WriteLine(@"Done");
return true;
}
private void RunSearch(SortedSet<DatabaseSequenceSpectrumMatch>[] matches, FastaDatabase db, ISequenceFilter sequenceFilter, CancellationToken? cancellationToken = null, IProgress<ProgressData> progress = null)
{
var progData = new ProgressData(progress)
{
Status = "Searching for matches"
};
var sw = new Stopwatch();
long estimatedProteins;
var annotationsAndOffsets = GetAnnotationsAndOffsets(db, out estimatedProteins, cancellationToken);
Console.WriteLine(@"Estimated proteins: " + estimatedProteins);
var numProteins = 0;
var lastUpdate = DateTime.MinValue; // Force original update of 0%
sw.Reset();
sw.Start();
var pfeOptions = new ParallelOptions
{
MaxDegreeOfParallelism = MaxNumThreads,
CancellationToken = cancellationToken ?? CancellationToken.None
};
var maxNumNTermCleavages = SearchMode == InternalCleavageType.NoInternalCleavage ? MaxNumNTermCleavages : 0;
//foreach (var annotationAndOffset in annotationsAndOffsets)
Parallel.ForEach(annotationsAndOffsets, pfeOptions, annotationAndOffset =>
{
if (cancellationToken != null && cancellationToken.Value.IsCancellationRequested)
{
//return matches;
return;
}
SearchProgressReport(ref numProteins, ref lastUpdate, estimatedProteins, sw, progData);
SearchForMatches(annotationAndOffset, sequenceFilter, matches, maxNumNTermCleavages, db.IsDecoy, cancellationToken);
});
Console.WriteLine(@"Collected candidate matches: {0}", GetNumberOfMatches(matches));
progData.StatusInternal = string.Empty;
progData.Report(100.0);
}
public void FindProteins()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string fastaFilePath = @"H:\Research\QCShew_TopDown\Production\ID_002216_235ACCEA.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);
const string tagFilePath = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3_seqtag.tsv";
if (!File.Exists(tagFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, tagFilePath);
}
const string outputFilePath = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3_matchedtag.tsv";
using (var writer = new StreamWriter(outputFilePath))
{
var isHeader = true;
foreach (var line in File.ReadAllLines(tagFilePath))
{
if (isHeader)
{
isHeader = false;
writer.WriteLine(line+"\t"+"Proteins");
continue;
}
var token = line.Split('\t');
if (token.Length != 3) continue;
var tag = token[1];
var matchedProteins =
searchableDb.FindAllMatchedSequenceIndices(tag)
.Select(index => fastaDb.GetProteinName(index))
.Distinct().ToArray();
var matchedProteinStr = string.Join(",", matchedProteins);
writer.WriteLine("{0}\t{1}\t{2}", line, matchedProteins.Length, matchedProteinStr);
}
}
Console.WriteLine(@"Done");
}
[TestCase(3, @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_005133_8491EFA2.fasta", 323719193)] // 3MB
//[TestCase(6, @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004530_B63BD900.fasta", 595227563)] // 6MB
//[TestCase(15, @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004208_295531A4.fasta", 1882434687)] // 15MB
public void TestSequenceEnumerationSerial(double size, string dbFile, int expected)
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName, dbFile);
var sw = new System.Diagnostics.Stopwatch();
sw.Start();
//const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_002216_235ACCEA.fasta"; // 1.5MB
//const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_005133_8491EFA2.fasta"; // 3MB
//const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004530_B63BD900.fasta"; // 6MB
//const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004208_295531A4.fasta"; // 15MB
var db = new FastaDatabase(dbFile);
var indexedDb = new IndexedDatabase(db);
indexedDb.Read();
var numSequences = 0L;
var timeDB = sw.Elapsed;
Console.WriteLine("Read DB in " + timeDB.TotalSeconds + " Seconds");
var estimatedAnnOff = indexedDb.EstimateTotalPeptides(0, 30, 250);
var timeEstimate = sw.Elapsed;
Console.WriteLine("Read Estimate in " + (timeEstimate - timeDB).TotalSeconds + " Seconds");
Console.WriteLine("Estimated results: " + estimatedAnnOff);
var annotationsAndOffsets = indexedDb.AnnotationsAndOffsetsNoEnzyme(30, 250);
var timeGetAnn = sw.Elapsed;
Console.WriteLine("Read Annotations in " + (timeGetAnn - timeEstimate).TotalSeconds + " Seconds");
//foreach (var annotationsAndOffset in annotationsAndOffsets)
//{
// //Interlocked.Increment(ref numSequences);
// ++numSequences;
//}
//using (
// var ofstream =
// new FileStream(
// Path.Combine(@"F:\InformedProteomicsTestFiles",
// Path.GetFileNameWithoutExtension(dbFile) + "_old.txt"), FileMode.Create))
//using (var fout = new StreamWriter(ofstream))
//{
// foreach (var annOff in annotationsAndOffsets)
// {
// numSequences++;
// fout.WriteLine(annOff.Annotation);
// }
//}
numSequences = annotationsAndOffsets.Count();
var timeParForEach = sw.Elapsed;
Console.WriteLine("Parallel ForEach in " + (timeParForEach - timeGetAnn).TotalSeconds + " Seconds");
Console.WriteLine("NumPeptides: {0}", numSequences);
sw.Stop();
Console.WriteLine(@"{0:f4} sec", sw.Elapsed.TotalSeconds);
//Assert.AreEqual(188961836, numSequences);
Assert.AreEqual(expected, numSequences);
}
public void TestTopDownScoringForAllXics()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
// Search parameters
const int numNTermCleavages = 1; // 30
const int minLength = 7;
const int maxLength = 1000;
//const int minCharge = 5; // 3
//const int maxCharge = 15; // 67
const int numMaxModsPerProtein = 0; // 6
var precursorTolerance = new Tolerance(10);
const string dbFilePath = @"..\..\..\TestFiles\sprot.Ecoli.2012_07.fasta";
//const string dbFilePath = @"..\..\..\TestFiles\sprot.Ecoli.2012_07.icdecoy.KR.fasta";
//const string dbFilePath = @"..\..\..\TestFiles\H_sapiens_Uniprot_SPROT_2013-05-01_withContam.fasta";
// const string dbFilePath =
// @"C:\cygwin\home\kims336\Data\TopDown\ID_003558_56D73071.fasta";
var sw = new System.Diagnostics.Stopwatch();
sw.Start();
Console.Write("Reading raw file...");
const string specFilePath = @"C:\workspace\TopDown\E_coli_iscU_60_mock.raw";
var run = InMemoryLcMsRun.GetLcMsRun(specFilePath);
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f4} sec", sw.Elapsed.TotalSeconds);
// Configure amino acid set
// var pyroGluQ = new SearchModification(Modification.PyroGluQ, 'Q', SequenceLocation.ProteinNTerm, false);
var dehydro = new SearchModification(Modification.PyroGluQ, 'C', SequenceLocation.Everywhere, false);
var cysteinylC = new SearchModification(Modification.Cysteinyl, 'C', SequenceLocation.Everywhere, false);
var glutathioneC = new SearchModification(Modification.Glutathione, 'C', SequenceLocation.Everywhere, false);
// var oxM = new SearchModification(Modification.Oxidation, 'M', SequenceLocation.Everywhere, false);
var searchModifications = new List<SearchModification>
{
//pyroGluQ,
dehydro,
cysteinylC,
glutathioneC,
//oxM
};
var aaSet = new AminoAcidSet(searchModifications, numMaxModsPerProtein);
var targetDb = new FastaDatabase(dbFilePath);
// targetDb.CreateDecoyDatabase(Enzyme.Trypsin);
// System.Environment.Exit(1);
var indexedDb = new IndexedDatabase(targetDb);
var numProteins = 0;
long totalProtCompositions = 0;
//long numXics = 0;
TopDownScorer.MaxCharge = 25;
TopDownScorer.MinCharge = 8;
sw.Reset();
sw.Start();
Console.WriteLine("Generating XICs...");
foreach (var protAnnotationAndOffset in indexedDb.IntactSequenceAnnotationsAndOffsets(minLength, maxLength))
{
++numProteins;
//if (numProteins > 2000) break;
if (numProteins % 1000 == 0)
{
Console.WriteLine("Processed {0} proteins", numProteins);
}
//Console.WriteLine(protAnnotation);
var seqGraph = SequenceGraph.CreateGraph(aaSet, protAnnotationAndOffset.Annotation);
//Console.WriteLine(seqGraph.GetSequenceCompositions()[0]);
if (seqGraph == null) continue;
for (var nTermCleavages = 0; nTermCleavages <= numNTermCleavages; nTermCleavages++)
{
if(nTermCleavages > 0) seqGraph.CleaveNTerm();
var protCompositions = seqGraph.GetSequenceCompositions();
foreach (var protComposition in protCompositions)
{
totalProtCompositions++;
// Console.WriteLine(protComposition);
var scorer = new TopDownScorer(protComposition, run, precursorTolerance, null);
var score = scorer.GetScore();
Console.WriteLine(score);
}
}
}
sw.Stop();
Console.WriteLine("NumProteins: {0}", numProteins);
Console.WriteLine("NumProteinCompositions: {0}", totalProtCompositions);
Console.WriteLine(@"Elapsed Time: {0:f4} sec", sw.Elapsed.TotalSeconds);
}
public void TestTagBasedSearchForLewy()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string rawFilePath = @"D:\MassSpecFiles\Lewy\Lewy_AT_AD1_21May15_Bane_14-09-01RZ.pbf";
if (!File.Exists(rawFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(rawFilePath);
const int minTagLength = 4;
var tagFilePath = MassSpecDataReaderFactory.ChangeExtension(rawFilePath, ".seqtag");
//var tagParser = new SequenceTagParser(tagFilePath, minTagLength, 10000);
const string fastaFilePath = @"D:\MassSpecFiles\Lewy\a4_human.fasta";
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fastaDb = new FastaDatabase(fastaFilePath);
var tolerance = new Tolerance(10);
var modsFilePath = @"D:\MassSpecFiles\Lewy\Mods.txt";
if (!File.Exists(modsFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, modsFilePath);
}
var aaSet = new AminoAcidSet(modsFilePath);
TestTagBasedSearch(run, fastaDb, tolerance, aaSet);
}
public void FindProteinDeltaMass()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string folderPath = @"D:\MassSpecFiles\Glyco\";
if (!Directory.Exists(folderPath))
{
Assert.Ignore(@"Skipping test {0} since folder not found: {1}", methodName, folderPath);
}
var fileSet = new string[]
{
"User_sample_test_02252015", "User_sample_test_MWCO_02262016", "User_sample_test_SEC_F3_03022105",
"User_sample_test_SEC_F1_02272015", "User_sample_test_SEC_F2_02282015"
};
const string fastaFilePath = folderPath + "ID_003836_DA9CC1E4.fasta";
for (var i = 0; i < fileSet.Length; i++)
{
var datasetName = fileSet[i];
var tagFilePath = folderPath + datasetName + ".seqtag";
//var outputFilePath = folderPath + datasetName + ".matchedtag";
var outputFilePath = folderPath + datasetName + ".dmass";
var fastaDb = new FastaDatabase(fastaFilePath);
var searchableDb = new SearchableDatabase(fastaDb);
using (var writer = new StreamWriter(outputFilePath))
{
var isHeader = true;
var nReadSeqTag = 0;
Console.WriteLine(@"Reading {0} file", tagFilePath);
var nColumn = 0;
foreach (var line in File.ReadAllLines(tagFilePath))
{
if (isHeader)
{
isHeader = false;
nColumn = line.Split('\t').Length;
writer.WriteLine(line + "\t" + "Protein" + "\t" + "DetectedFlankingMass" + "\t" + "ExpectedFlankingMass" + "\t" + "DeltaMass");
continue;
}
var token = line.Split('\t');
if (token.Length != nColumn) continue;
var tag = token[1];
//var scan = Convert.ToInt32(token[0]);
if (tag.Length < 6) continue;
var nTerminal = token[2].Equals("1");
var detectedFlankingMass = Double.Parse(token[3]);
if (!nTerminal) detectedFlankingMass -= Composition.H2O.Mass;
nReadSeqTag++;
var matchedProteins =
searchableDb.FindAllMatchedSequenceIndices(tag)
.Select(index => fastaDb.GetProteinName(index))
.Distinct().ToArray();
if (matchedProteins.Length < 1) continue;
foreach (var protName in matchedProteins)
{
var seqStr = fastaDb.GetProteinSequence(protName);
var oriSeq = new Sequence(seqStr, AminoAcidSet.GetStandardAminoAcidSet());
var startIdx = 0;
while (true)
{
var idx = seqStr.IndexOf(tag, startIdx);
if (idx < 0) break; //no matching
//var nClv = (nTerminal) ? idx : seqStr.Length - idx - tag.Length;
var nClv = (nTerminal) ? 2 : 1;
for (var j = 0; j < nClv; j++)
{
var flankComposition = (nTerminal)
? oriSeq.GetComposition(j, idx)
: oriSeq.GetComposition(idx + tag.Length, oriSeq.Count - j);
var massDiff = (detectedFlankingMass - flankComposition.Mass);
if (massDiff > -500 && massDiff < 2000)
{
//writer.WriteLine(massDiff);
writer.WriteLine("{0}\t{1}\t{2}\t{3}\t{4}", line, protName, detectedFlankingMass, flankComposition.Mass, massDiff);
}
if (massDiff > 2000) break;
}
startIdx = idx + tag.Length;
}
}
//var matchedProteinStr = string.Join(",", matchedProteins);
//var massDiffStr = string.Join(",", massDiffList);
//writer.WriteLine("{0}\t{1}\t{2}\t{3}", line, matchedProteins.Length, matchedProteinStr, massDiffStr);
}
Console.WriteLine(@"{0} seq tags are processed", nReadSeqTag);
}
Console.WriteLine(@"Done");
}
}
public void CountMatchedScansPerProtein()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.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 TestTagBasedSearchCompRef()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string dataSetPath = @"D:\MassSpecFiles\CompRef";
const string fastaFilePath = @"D:\MassSpecFiles\CompRef\ID_003278_4B4B3CB1.fasta";
const string modsFilePath = @"D:\MassSpecFiles\CompRef\Mods.txt";
if (!Directory.Exists(dataSetPath))
{
Assert.Ignore(@"Skipping test {0} since folder not found: {1}", methodName, dataSetPath);
}
if (!File.Exists(modsFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, modsFilePath);
}
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fileEntries = Directory.GetFiles(dataSetPath);
var dataset = (from fileName in fileEntries where fileName.EndsWith("pbf") select Path.GetFileNameWithoutExtension(fileName)).ToList();
dataset.Sort();
var fastaDb = new FastaDatabase(fastaFilePath);
var tolerance = new Tolerance(10);
var aaSet = new AminoAcidSet(modsFilePath);
for (var i = 0; i < dataset.Count; i++)
{
var rawFile = string.Format(@"{0}\{1}.pbf", dataSetPath, dataset[i]);
var ms1File = string.Format(@"{0}\{1}.ms1ft", dataSetPath, dataset[i]);
var tagFilePath = MassSpecDataReaderFactory.ChangeExtension(rawFile, ".seqtag");
var run = PbfLcMsRun.GetLcMsRun(rawFile);
const int minTagLength = 5;
//var tagParser = new SequenceTagParser(tagFilePath, minTagLength, 100);
Console.WriteLine("-----------------{0}--------------------", rawFile);
TestTagBasedSearch(run, fastaDb, tolerance, aaSet);
Console.WriteLine("-----------------------------------------------------------------------");
}
}
public void TestTagBasedSearch()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
// const string rawFilePath = @"H:\Research\Lewy\raw\Lewy_intact_01.raw";
// const string rawFilePath = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3.raw";
// const string rawFilePath = @"H:\Research\Yufeng\TopDownYufeng\raw\yufeng_column_test2.raw";
// const string rawFilePath = @"H:\Research\Weijun_TopDown\raw\UC4_Intact_plasmaTest_90_6May15_Bane_14-09-01RZ.raw";
// const string rawFilePath = @"H:\Research\Charles\TopDown\raw\SBEP_STM_001_02272012_Aragon.raw";
const string rawFilePath = @"D:\MassSpecFiles\60k\Yufeng_SampleTest1_150614113438.pbf";
//const string rawFilePath = @"D:\MassSpecFiles\60k\NCR_50K_Test_24Jun15_Bane_15-02-02RZ.pbf";
if (!File.Exists(rawFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(rawFilePath);
const int minTagLength = 5;
var tagFilePath = MassSpecDataReaderFactory.ChangeExtension(rawFilePath, ".seqtag");
//var tagParser = new SequenceTagParser(tagFilePath, minTagLength, 100);
const string fastaFilePath = @"D:\MassSpecFiles\60k\ID_003836_DA9CC1E4.fasta";
//const string fastaFilePath = @"D:\MassSpecFiles\60k\ID_004973_9BA6912F.fasta";
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fastaDb = new FastaDatabase(fastaFilePath);
var tolerance = new Tolerance(10);
// var modsFilePath = @"H:\Research\QCShew_TopDown\Production\Mods_Methyl.txt";
var modsFilePath = @"D:\MassSpecFiles\60k\Mods.txt";
// var modsFilePath = "";
if (!File.Exists(modsFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, modsFilePath);
}
var aaSet = new AminoAcidSet(modsFilePath);
TestTagBasedSearch(run, fastaDb, tolerance, aaSet);
}
public void TestNominalMassErrors()
{
const int minLength = 300;
const int maxLength = 400;
var sw = new System.Diagnostics.Stopwatch();
// const string dbFile = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\H_sapiens_Uniprot_SPROT_2013-05-01_withContam.fasta";
const string dbFile = @"C:\cygwin\home\kims336\Data\TopDownJia\database\ID_003962_71E1A1D4.fasta";
//const string dbFile = @"C:\cygwin\home\kims336\Data\TopDownJia\database\TargetProteins.fasta";
var db = new FastaDatabase(dbFile);
db.Read();
var indexedDb = new IndexedDatabase(db);
var numSequences = 0L;
sw.Start();
var hist = new long[11];
var aaSet = new AminoAcidSet();
foreach (var peptideAnnotationAndOffset in indexedDb.AnnotationsAndOffsetsNoEnzyme(minLength, maxLength))
{
++numSequences;
var annotation = peptideAnnotationAndOffset.Annotation;
var sequenceStr = annotation.Substring(2, annotation.Length - 4);
var sequenceComp = aaSet.GetComposition(sequenceStr);
var mass = sequenceComp.Mass;
var nominalMass = sequenceComp.NominalMass;
var error = (int) Math.Round(mass*Constants.RescalingConstant) - nominalMass;
var errorBin = error + hist.Length/2;
if (errorBin < 0) errorBin = 0;
if (errorBin >= hist.Length) errorBin = hist.Length - 1;
hist[errorBin]++;
}
Console.WriteLine("NumSequences: {0}", numSequences);
for (var i = 0; i < hist.Length; i++)
{
Console.WriteLine("{0}\t{1}\t{2}", i - hist.Length/2, hist[i], hist[i]/(double)numSequences);
}
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f4} sec", sw.Elapsed.TotalSeconds);
}
[TestCase(6, @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004530_B63BD900.fasta", 8898)] // 6MB
//[TestCase(15, @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\MSPathFinderT\ID_004208_295531A4.fasta", 6334)] // 15MB
public void TestSequenceEnumerationIntact(double size, string dbFile, int expected)
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName, dbFile);
var sw = new System.Diagnostics.Stopwatch();
sw.Start();
const int numCTermCleavages = 0;
var db = new FastaDatabase(dbFile);
var indexedDb = new IndexedDatabase(db);
var numSequences = 0L;
var timeDB = sw.Elapsed;
Console.WriteLine("Read DB in " + timeDB.TotalSeconds + " Seconds");
var estimatedAnnOff = indexedDb.EstimateTotalPeptides(2, 21, 300, 1, numCTermCleavages);
var timeEstimate = sw.Elapsed;
Console.WriteLine("Read Estimate in " + (timeEstimate - timeDB).TotalSeconds + " Seconds");
Console.WriteLine("Estimated results: " + estimatedAnnOff);
var annotationsAndOffsets = indexedDb.IntactSequenceAnnotationsAndOffsets(21, 300, numCTermCleavages);
var timeGetAnn = sw.Elapsed;
Console.WriteLine("Read Annotations in " + (timeGetAnn - timeEstimate).TotalSeconds + " Seconds");
/*/Parallel.ForEach(
annotationsAndOffsets,
// new ParallelOptions { MaxDegreeOfParallelism = 2},
annotationAndOffset =>
{
Interlocked.Increment(ref numSequences);
//++numSequences;
}
);/**/
//using (var ofstream = new FileStream(Path.Combine(@"F:\InformedProteomicsTestFiles", Path.GetFileNameWithoutExtension(dbFile) + "_par.txt"), FileMode.Create))
//using (var fout = new StreamWriter(ofstream))
//{
// foreach (var annOff in annotationsAndOffsets)
// {
// numSequences++;
// fout.WriteLine(annOff.Annotation);
// }
//}
numSequences = annotationsAndOffsets.Count();
var timeParForEach = sw.Elapsed;
Console.WriteLine("Parallel ForEach in " + (timeParForEach - timeGetAnn).TotalSeconds + " Seconds");
Console.WriteLine("NumPeptides: {0}", numSequences);
sw.Stop();
Console.WriteLine(@"{0:f4} sec", sw.Elapsed.TotalSeconds);
//Assert.AreEqual(188961836, numSequences);
Assert.AreEqual(expected, numSequences);
}
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
);
}
}
}
public void CountMatchedProteins()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const int minTagLength = 3;
var scanToProtein = new Dictionary<int, string>();
var idTag = new Dictionary<int, bool>();
const string resultFilePath = @"H:\Research\ProMex\QC_Shew_Intact_26Sep14_Bane_C2Column3_IcTda.tsv";
if (!File.Exists(resultFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, resultFilePath);
}
var parser = new TsvFileParser(resultFilePath);
var scans = parser.GetData("Scan").Select(s => Convert.ToInt32(s)).ToArray();
var proteinNames = parser.GetData("ProteinName").ToArray();
var qValues = parser.GetData("QValue").Select(Convert.ToDouble).ToArray();
for (var i = 0; i < qValues.Length; i++)
{
if (qValues[i] > 0.01) break;
scanToProtein.Add(scans[i], proteinNames[i]);
idTag.Add(scans[i], false);
}
const string rawFilePath = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3.raw";
if (!File.Exists(rawFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(rawFilePath);
const string fastaFilePath = @"H:\Research\QCShew_TopDown\Production\ID_002216_235ACCEA.fasta";
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
// const string fastaFilePath = @"H:\Research\QCShew_TopDown\Production\ID_002216_235ACCEA.icsfldecoy.fasta";
// const string fastaFilePath =
// @"D:\Research\Data\CommonContaminants\H_sapiens_Uniprot_SPROT_2013-05-01_withContam.fasta";
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";
if (!File.Exists(tagFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, tagFilePath);
}
var hist = new Dictionary<int, int>();
var scanSet = new HashSet<int>();
HashSet<string> proteinSetForThisScan = null;
var prevScan = -1;
var totalNumMatches = 0L;
var isHeader = true;
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 proteinId = scanToProtein.ContainsKey(scan) ? scanToProtein[scan] : null;
if (scan != prevScan)
{
if (proteinSetForThisScan != null)
{
var numMatches = proteinSetForThisScan.Count;
int numOcc;
if (hist.TryGetValue(numMatches, out numOcc)) hist[numMatches] = numOcc + 1;
else hist.Add(numMatches, 1);
}
prevScan = scan;
proteinSetForThisScan = new HashSet<string>();
}
scanSet.Add(scan);
var tag = token[1];
if (tag.Length < minTagLength) continue;
if (proteinSetForThisScan == null) continue;
var numMatchesForThisTag = 0;
foreach (var matchedProtein in searchableDb.FindAllMatchedSequenceIndices(tag)
.Select(index => fastaDb.GetProteinName(index)))
{
proteinSetForThisScan.Add(matchedProtein);
++numMatchesForThisTag;
if (proteinId != null && matchedProtein.Equals(proteinId))
{
idTag[scan] = true;
}
}
totalNumMatches += numMatchesForThisTag;
// if (numMatchesForThisTag > 10)
// {
// Console.WriteLine("{0}\t{1}", tag, numMatchesForThisTag);
// }
}
if (proteinSetForThisScan != null)
{
var numMatches = proteinSetForThisScan.Count;
int numOcc;
if (hist.TryGetValue(numMatches, out numOcc)) hist[numMatches] = numOcc + 1;
else hist.Add(numMatches, 1);
}
Console.WriteLine("AvgNumMatches: {0}", totalNumMatches/(float)scanSet.Count);
Console.WriteLine("Histogram:");
foreach (var entry in hist.OrderBy(e => e.Key))
{
Console.WriteLine("{0}\t{1}", entry.Key, entry.Value);
}
Console.WriteLine("NumId: {0}", idTag.Count);
Console.WriteLine("NumIdByTag: {0}", idTag.Select(e => e.Value).Count(v => v));
}
private void TestTagBasedSearch(LcMsRun run,
FastaDatabase fastaDb, Tolerance tolerance, AminoAcidSet aaSet)
{
var engine = new ScanBasedTagSearchEngine(run, new SequenceTagGenerator(run, new Tolerance(8)), new LcMsPeakMatrix(run), fastaDb, tolerance, aaSet);
// engine.MinScan = 3400;
// engine.MaxScan = 3900;
engine.RunSearch();
}
