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 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 TestGettingProteinLengthAndPosition()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
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);
db.Read();
var indexedDb = new IndexedDatabase(db);
foreach (var peptideAnnotationAndOffset in indexedDb.AnnotationsAndOffsets(6, 20, 2, 0, Enzyme.Trypsin))
{
var annotation = peptideAnnotationAndOffset.Annotation;
var offset = peptideAnnotationAndOffset.Offset;
Console.WriteLine("{0}\t{1}\t{2}\t{3}\t{4}",
annotation,
offset,
db.GetProteinName(offset),
db.GetProteinLength(db.GetProteinName(offset)),
db.GetOneBasedPositionInProtein(offset)+1);
}
}
public void GenerateAbrfSpecCountAllProteins()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string dir = @"H:\Research\IPRG2015";
if (!Directory.Exists(dir))
{
Assert.Ignore(@"Skipping test {0} since folder not found: {1}", methodName, dir);
}
const double qValueThreshold = 0.01;
//var names = new[] { "ENO1_YEAST", "ADH1_YEAST", "CYC_BOVIN", "ALBU_BOVIN" };
//var accessions = new[] { "P00924", "P00330", "P62894", "P02769" };
const string resultDir = dir + @"\NTT1";
if (!Directory.Exists(resultDir))
{
Assert.Ignore(@"Skipping test {0} since folder not found: {1}", methodName, resultDir);
}
var msgfResultFiles = Directory.GetFiles(resultDir, "*.tsv").ToArray();
var specCount = new Dictionary<string, int[]>(); // protein name => array of counts
for (var i = 0; i < msgfResultFiles.Length; i++)
{
var msgfResultFile = msgfResultFiles[i];
MsGfPlusHeaderInformation headerInfo = null;
var prevScanNum = -1;
foreach (var line in File.ReadLines(msgfResultFile))
{
if (line.StartsWith("#"))
{
headerInfo = new MsGfPlusHeaderInformation(line);
continue;
}
var match = new MsGfMatch(line, headerInfo);
if (match.ScanNum == prevScanNum) continue;
prevScanNum = match.ScanNum;
if (!match.IsValid || match.Protein.StartsWith(FastaDatabase.DecoyProteinPrefix)) continue;
if (match.QValue > qValueThreshold) continue;
var proteins = match.Protein.Split(';');
foreach (var protein in proteins)
{
var proteinName = protein.Substring(0, protein.LastIndexOf("(pre=", StringComparison.Ordinal));
int[] countArr;
if (!specCount.TryGetValue(proteinName, out countArr)) specCount[proteinName] = new int[msgfResultFiles.Length];
specCount[proteinName][i]++;
}
}
}
// Writing
const string databaseFilePath = dir + @"\database\iPRG2015.fasta";
var database = new FastaDatabase(databaseFilePath);
database.Read();
// var spikeInAccessions = new[] { "STANDARD_Alpha-Casein", "STANDARD_Beta-Lactoglobulin", "STANDARD_Carbonic-Anhydrase", "P02769"};
const string outputFilePath = dir + @"\SpecCountAllProteins.tsv";
using (var writer = new StreamWriter(outputFilePath))
{
var fileIds = msgfResultFiles.Select(f => f.Substring(f.IndexOf("_sample", StringComparison.Ordinal) + 1,
f.LastIndexOf('.') - f.IndexOf("_sample", StringComparison.Ordinal) - 1));
writer.WriteLine("Protein\tLength\t" + string.Join("\t", fileIds) + "\tSpikeIn");
foreach (var entry in specCount)
{
var proteinId = entry.Key;
var length = database.GetProteinLength(proteinId);
Assert.True(length > 0);
var counts = entry.Value;
Assert.True(counts.Length == msgfResultFiles.Length);
var spikeIn = 0;
//if (spikeInAccessions.Any(spikeInAccession => proteinId.StartsWith("sp|" + spikeInAccession)))
if (proteinId.StartsWith("sp|"))
{
spikeIn = 1;
}
writer.WriteLine("{0}\t{1}\t{2}\t{3}", proteinId, length, string.Join("\t", counts), spikeIn);
}
}
}
public void ProcessIprg2015PreStudy()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string dir = @"H:\Research\IPRG2015";
const string databaseFilePath = dir + @"\database\yeast6proteaprotein.fasta";
if (!File.Exists(databaseFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, databaseFilePath);
}
var database = new FastaDatabase(databaseFilePath);
database.Read();
const string jobFilePath = dir + @"\Jobs.tsv";
if (!File.Exists(jobFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, jobFilePath);
}
var jobParser = new TsvFileParser(jobFilePath);
var jobs = jobParser.GetData("Jobs").Select(j => Convert.ToInt32(j)).ToArray();
var experiments = jobParser.GetData("Experiments").Select(e => e.Split('_')[2]).ToArray();
//const string resultFilePath = dir + @"\AMT_Proteins_NA.tsv";
//const string outputFilePath = dir + @"\AMT_Proteins.tsv";
const string resultFilePath = dir + @"\AMT_Peptides_NA.tsv";
const string outputFilePath = dir + @"\AMT_Peptides.tsv";
var parser = new TsvFileParser(resultFilePath);
var headers = parser.GetHeaders();
var jobColNum = new int[jobs.Length];
for (var i = 0; i < jobs.Length; i++)
{
for (var j = 0; j < headers.Count; j++)
{
if (headers[j].Contains("" + jobs[i]))
{
jobColNum[i] = j;
break;
}
}
}
for (var i = 0; i < jobs.Length; i++)
{
Console.WriteLine("{0}\t{1}\t{2}", jobs[i], jobColNum[i], experiments[i]);
}
using (var writer = new StreamWriter(outputFilePath))
{
var peptides = parser.GetData("Peptide"); // Peptides
var proteins = parser.GetData("Reference"); // Proteins
var abundances = new string[jobs.Length][];
for (var i = 0; i < jobs.Length; i++)
{
abundances[i] = parser.GetData(headers[jobColNum[i]]).ToArray();
}
if (peptides != null) writer.Write("Peptide\t");
writer.Write("Protein\tLength");
for (var i = 0; i < jobs.Length; i++)
{
writer.Write("\t" + experiments[i]);
}
writer.WriteLine("\tSpikeIn");
for (var i = 0; i < proteins.Count; i++)
{
var protein = proteins[i];
if (protein.StartsWith("XXX") || protein.StartsWith("Contaminant")) continue;
var length = database.GetProteinLength(protein);
//if (length <= 0)
//{
// Console.WriteLine("Shit!");
// return;
//}
if (peptides != null) writer.Write(peptides[i] + "\t");
writer.Write(protein + "\t" + length);
for (var j = 0; j < jobs.Length; j++)
{
writer.Write("\t" + abundances[j][i]);
}
writer.WriteLine("\t" + (protein.StartsWith("STANDARD") ? 1 : 0));
}
}
}
public void AddProteinLengths()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string databaseFilePath = @"H:\Research\IPRG2015\database\yeast6proteaprotein.fasta";
if (!File.Exists(databaseFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, databaseFilePath);
}
var database = new FastaDatabase(databaseFilePath);
database.Read();
const string resultPath = @"H:\Research\IPRG2015\AMT_Peptides_NA.tsv";
if (!File.Exists(resultPath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, resultPath);
}
const string outputFilePath = @"H:\Research\IPRG2015\AMT_Peptides.tsv";
using (var writer = new StreamWriter(outputFilePath))
{
foreach (var line in File.ReadLines(resultPath))
{
var data = line.Split(null);
if (data.Length != 14) continue;
var peptide = data[0];
if (peptide.Equals("Peptide"))
{
writer.WriteLine("Peptide\tProtein\tLength\t{0}", string.Join("\t", data.Skip(2)));
continue;
}
var protein = data[1];
var length = database.GetProteinLength(protein);
writer.WriteLine("{0}\t{1}\t{2}\t{3}", peptide, protein, length, string.Join("\t", data.Skip(2)));
}
}
}
