public static ResultInstance OverlapShortInstances(this ResultInstance shortInstance1, ResultInstance shortInstance2)
{
if (shortInstance1.start > shortInstance2.end || shortInstance2.start > shortInstance1.end)
{
return(null);
}
return(new ResultInstance
{
featureDetails = new List <FeatureDetails>()
{
new FeatureDetails()
{
featureType = shortInstance1.featureDetails[0].featureType,
info = shortInstance1.featureDetails[0].info
},
new FeatureDetails()
{
featureType = shortInstance2.featureDetails[0].featureType,
info = shortInstance2.featureDetails[0].info
},
},
start = new[] { shortInstance1.start, shortInstance2.start }.Max(),
end = new[] { shortInstance1.end, shortInstance2.end }.Min(),
});
}
public static List <ResultInstance> UnionShortInstances(this ResultInstance shortInstance1, ResultInstance shortInstance2)
{
if (shortInstance1.start > shortInstance2.end || shortInstance2.start > shortInstance1.end)
{
return(new List <ResultInstance> {
shortInstance1, shortInstance2
});
}
return(new List <ResultInstance>
{
new ResultInstance {
//FeatureType = shortInstance1.FeatureType + shortInstance2.FeatureType,
start = new[] { shortInstance1.start, shortInstance2.start }.Min(),
end = new[] { shortInstance1.end, shortInstance2.end }.Max(),
}
});
}
/// <summary>
/// add spectra labels to ScanRanker metrics file
/// identified spectra ids stored in idpXML file by idqQonvert
/// </summary>
public void AddSpectraLabel()
{
// get pepXML files and run idpQonvert for all, so only read database once
idpCfg.PepXMLFiles = "";
foreach (FileInfo file in inFileList)
{
string fileBaseName = Path.GetFileNameWithoutExtension(file.Name);
string pepXMLFile = idpCfg.PepXMLFileDir + "\\" + fileBaseName + ".pepXML"; // pepXML file has to be the same basename
idpCfg.PepXMLFiles += "\"" + pepXMLFile + "\"" + " "; // each file enclosed by "..." and separated by a space
string idpDBFilename = fileBaseName + ".idpDB"; // pepXML file has to be the same basename as spectra filename
if (File.Exists(idpDBFilename))
{
File.Delete(idpDBFilename);
Workspace.SetText("\r\nRemoved existing idpDB file: " + idpDBFilename + "\n");
}
}
//Workspace.SetText("\r\npepXMLFies\r\n" + idpCfg.PepXMLFiles + "\r\n\r\n");
Workspace.SetText("\r\nRunning idpQonvert for peptide validation\r\n");
try
{
runIdpQonvert(idpCfg, outDir);
}
catch (Exception exc)
{
Workspace.SetText("\r\nError in running idpQonvert\r\n");
Workspace.ChangeButtonTo("Close");
throw new Exception(exc.Message);
}
// use IDPickerWorkspace.dll to parse idpXML, get spectra->peptides->proteins info, work for IDP2
// use System.Data.SQLite to read idpDB for IDP3, get spectra->peptides->proteins info
// add labels, peptide and proteins to metrics file
foreach (FileInfo file in inFileList)
{
string fileBaseName = Path.GetFileNameWithoutExtension(file.Name);
//string idpXMLFilename = fileBaseName + ".idpXML"; // pepXML file has to be the same basename as spectra filename
string idpDBFilename = fileBaseName + ".idpDB"; // pepXML file has to be the same basename as spectra filename
if (!File.Exists(idpDBFilename))
{
Workspace.SetText("\r\nError: Cannot find idpDB file: " + idpDBFilename + " in output directory!");
Workspace.SetText("\r\nPlease check IDPicker configurations and the database file");
Workspace.ChangeButtonTo("Close");
return;
}
// parse idpDB file
Workspace.SetText("\r\nReading idpDB file: " + idpDBFilename);
Map <string, ResultInstance> resultsMap = new Map <string, ResultInstance>();
// Connect to database
string dbConnectionString = "Data Source=" + idpDBFilename;
using (SQLiteConnection sqliteCon = new SQLiteConnection(dbConnectionString))
{
// Execute query on database
string querySQL = @"SELECT s.NativeID,
REPLACE(IFNULL(GROUP_CONCAT(DISTINCT pm.Offset || '=' || mod.MonoMassDelta),''),',',';') as Mods,
IFNULL(IFNULL(SUBSTR(pd.Sequence, pi.Offset+1, pi.Length),
(SELECT DecoySequence FROM Peptide p WHERE p.Id = pi.Peptide)),
(SELECT SUBSTR(pd.Sequence, pi.Offset+1, pi.Length)
FROM PeptideInstance pi
JOIN Protein pro ON pi.Protein = pro.Id AND pro.IsDecoy = 0
LEFT JOIN ProteinData pd ON pi.Protein=pd.Id
WHERE psm.Peptide = pi.Peptide
UNION
SELECT p.DecoySequence
FROM Peptide p
JOIN PeptideInstance pi ON p.Id = pi.Peptide
JOIN Protein pro ON pi.Protein = pro.Id AND pro.IsDecoy = 1
WHERE psm.Peptide = pi.Peptide AND p.DecoySequence is not null)) as Peptide,
GROUP_CONCAT(pro.Accession) as Protein
FROM PeptideSpectrumMatch psm
JOIN Spectrum s ON s.Id = psm.Spectrum
JOIN SpectrumSource source ON s.Source = source.Id
JOIN PeptideInstance pi ON psm.Peptide = pi.Peptide
JOIN Protein pro ON pi.Protein = pro.Id
LEFT JOIN ProteinData pd ON pi.Protein=pd.Id
LEFT JOIN PeptideModification pm ON psm.Id = pm.PeptideSpectrumMatch
LEFT JOIN Modification mod ON pm.Modification = mod.Id
Where psm.Rank = 1 and psm.Qvalue < " + idpCfg.MaxFDR.ToString() +
" GROUP BY psm.Id";
Workspace.SetText("\r\nExtracting identified spectra, peptides and proteins from " + idpDBFilename);
using (SQLiteCommand queryCommand = new SQLiteCommand(querySQL, sqliteCon))
{
sqliteCon.Open();
using (SQLiteDataReader dataReader = queryCommand.ExecuteReader())
{
// Iterate every record
while (dataReader.Read())
{
Console.WriteLine(dataReader.GetString(0) + " " + dataReader.GetString(1) + " " + dataReader.GetString(2));
ResultInstance ri = new ResultInstance();
string nativeID = dataReader.GetString(0);
ri.Mods = dataReader.GetString(1);
ri.Peptide = dataReader.GetString(2);
ri.Protein = dataReader.GetString(3);
resultsMap.Add(nativeID, ri);
}
}
}
}
//dataReader.Close();
//sqliteCon.Close();
//IDPicker.Workspace ws = new IDPicker.Workspace();
//using (StreamReader idpXMLStream = new StreamReader(idpDBFilename))
//{
// ws.readPeptidesXml(idpXMLStream, "", (float)idpCfg.MaxFDR, 1);
//}
//foreach (SourceGroupInfo group in ws.groups.Values)
// foreach (SourceInfo source in group.getSources())
// foreach (SpectrumInfo spectrum in source.spectra.Values)
// {
// ResultInstance ri = spectrum.results[1];
// resultsMap.Add(spectrum.nativeID, ri);
// //Workspace.SetText("\r\n" + spectrum.nativeID);
// }
// remove idpDB file
if (File.Exists(idpDBFilename))
{
File.Delete(idpDBFilename);
}
Workspace.SetText("\r\nRemoved idpDB file: " + idpDBFilename + "\n");
// read idpxml, extract spectra id, save to a dictionary
/*Dictionary<string, int> idtScanDict = new Dictionary<string, int>();
*
* try
* {
* using (XmlTextReader reader = new XmlTextReader(idpXMLFilename))
* {
* while (reader.Read())
* {
* if (reader.NodeType.Equals(XmlNodeType.Element) && reader.Name.Equals("spectrum"))
* {
* // Read the spectrum tag
* // <spectrum id="scan=3601" index="2834" z="2" mass="1466.6829557734375" time="33.856500000000004" targets="45327" decoys="45504" results="1">
* string nativeID = getAttributeAs<string>(reader, "id", false); // id in idpXML = nativeID in DirecTag
* Match m = Regex.Match(nativeID, @"scan=(\d+)");
* if (m.Success)
* {
* nativeID = m.Groups[1].Value;
* }
* //int z = getAttributeAs<int>(reader, "z", true);
* //int index = getAttributeAs<int>(reader, "index", true);
* //string idtScan = nativeID + "." + Convert.ToString(z);
* if (!idtScanDict.ContainsKey(nativeID))
* idtScanDict.Add(nativeID, 1); // use only scan number as identification, no charge info
* }
* }
* }
* }
* catch (Exception exc)
* {
* Workspace.SetText("\r\nError in reading idpXML file, please check IDPicker configuration and try again\r\n");
* Workspace.ChangeButtonTo("Close");
* throw new Exception(exc.Message);
* }*/
// open metrics file, if identified, add label, write to a new file
string metricsFileName = fileBaseName + metricsFileSuffix + ".txt";
string outFileName = fileBaseName + metricsFileSuffix + outFileSuffix + ".txt";
Workspace.SetText("\r\nWriting labbled metrics file: " + outFileName);
//List<int> unidentifiedSpectra = new List<int>();
List <string> unidentifiedSpectra = new List <string>();
int numSpectra = 0;
int cumsum = 0;
try
{
if (File.Exists(outFileName))
{
File.Delete(outFileName);
}
using (TextReader r = File.OpenText(metricsFileName))
{
using (TextWriter w = File.CreateText(outFileName))
{
w.WriteLine(r.ReadLine()); // read and write the 1st header line
w.WriteLine(r.ReadLine()); // read and write the 1st header line
string header = r.ReadLine(); // read the 3rd header line
w.WriteLine(header + "\t" + "Label" + "\t" + "CumsumLabel" + "\t" + "Peptide" + "\t" + "Protein"); // write the 3rd header line
string line = string.Empty;
while ((line = r.ReadLine()) != null)
{
numSpectra++;
string[] items = line.Split('\t');
//string scanNativeID = items[2]; // nativeID
string scanNativeID = items[1]; // nativeID
//int index = Convert.ToInt32(items[1]); //index
//Match m = Regex.Match(scanNativeID, @"scan=(\d+)"); // extract scan number in nativeID
//if (m.Success)
//{
// scanNativeID = m.Groups[1].Value;
//}
if (resultsMap.Contains(scanNativeID))
{
cumsum += 1;
ResultInstance ri = resultsMap[scanNativeID];
if (ri.Mods.Equals(string.Empty))
{
w.WriteLine(line + "\t1\t" + cumsum + "\t" + ri.Peptide.ToString() + "\t" + ri.Protein.ToString());
}
else
{
w.WriteLine(line + "\t1\t" + cumsum + "\t" + ri.Peptide.ToString() + '{' + ri.Mods.ToString() + '}' + "\t" + ri.Protein.ToString());
}
}
else
{
w.WriteLine(line + "\t0\t" + cumsum);
//unidentifiedSpectra.Add(index);
unidentifiedSpectra.Add(scanNativeID);
}
//scanNativeID = scanNativeID + "." + items[2]; // use nativeID scanNumber.charge as scanID
}
}
}
}
catch (Exception exc)
{
//throw new Exception("Error in creating spectra lable file\r\n", exc);
Workspace.SetText("\r\nError in creating a file with spectra labels, please check the ScanRanker metrics file\r\n");
Workspace.ChangeButtonTo("Close");
throw new Exception(exc.Message);
}
Workspace.SetText("\r\nFinished adding spectral labels for file: " + metricsFileName + " \r\n\r\n");
// write out unidentfied high quality spectra
if (unidentifiedSpectra.Count == numSpectra)
{
Workspace.SetText("\r\nNo spectra were identified. This might because the spectrumNativeID attribute is missing in pepXML files");
Workspace.ChangeButtonTo("Close");
return;
}
if (writeOutUnidentifiedSpectra)
{
writeUnidentifiedSpectra(file.Name, outFileSuffix, unidentifiedSpectra, recoveryCutoff, recoveryOutFormat);
}
Workspace.SetText("\r\n" + file.Name
+ ":\r\n\tTotal number of spectra: " + numSpectra.ToString()
+ "\r\n\tNumber of identified spectra: " + cumsum.ToString()
+ "\r\n\tNumber of unidentified spectra in output: " + (Convert.ToInt32(unidentifiedSpectra.Count * recoveryCutoff)).ToString()
+ "\r\n");
//delete old metrics file
//if (File.Exists(metricsFileName))
//{
// File.Delete(metricsFileName);
//}
} // end of foreach file
Workspace.SetText("\r\nFinished adding spectral labels!");
Workspace.ChangeButtonTo("Close");
}// end of addSpectraLabel()
