private bool AMTTagPassesFilter(MassTagLight item, int minimumObservationCount)
{
if (item.ObservationCount < minimumObservationCount)
{
return(false);
}
if (_options.AMTTagFilterNETMax > _options.AMTTagFilterNETMin)
{
if (item.Net < _options.AMTTagFilterNETMin || item.Net > _options.AMTTagFilterNETMax)
{
return(false);
}
}
if (_options.AMTTagFilterMassMax > _options.AMTTagFilterMassMin)
{
if (item.MassMonoisotopic < _options.AMTTagFilterMassMin ||
item.MassMonoisotopic > _options.AMTTagFilterMassMax)
{
return(false);
}
}
return(true);
}
public MassTagDatabase LoadDatabase()
{
var database = new MassTagDatabase();
var massTags = new List<MassTagLight>();
var masstagMap = new Dictionary<int, List<Protein>>();
//TODO: Put into base table file reader
var lines = File.ReadAllLines(DatabasePath);
var delimiters = new[] {"\t"};
for (var i = 1; i < lines.Length; i++)
{
var data = lines[i].Split(delimiters, StringSplitOptions.RemoveEmptyEntries);
var tag = new MassTagLight();
if (data.Length > 4)
{
tag.Id = Convert.ToInt32(data[0]);
tag.PeptideSequence = data[1];
tag.MassMonoisotopic = Convert.ToDouble(data[2]);
tag.Net = Convert.ToDouble(data[3]);
tag.DriftTime = Convert.ToDouble(data[4]);
}
massTags.Add(tag);
}
database.AddMassTagsAndProteins(massTags, masstagMap);
return database;
}
public MassTagDatabase LoadDatabase()
{
var database = new MassTagDatabase();
var massTags = new List <MassTagLight>();
var masstagMap = new Dictionary <int, List <Protein> >();
//TODO: Put into base table file reader
var lines = File.ReadAllLines(DatabasePath);
var delimiters = new[] { "\t" };
for (var i = 1; i < lines.Length; i++)
{
var data = lines[i].Split(delimiters, StringSplitOptions.RemoveEmptyEntries);
var tag = new MassTagLight();
if (data.Length > 4)
{
tag.Id = Convert.ToInt32(data[0]);
tag.PeptideSequence = data[1];
tag.MassMonoisotopic = Convert.ToDouble(data[2]);
tag.Net = Convert.ToDouble(data[3]);
tag.DriftTime = Convert.ToDouble(data[4]);
}
massTags.Add(tag);
}
database.AddMassTagsAndProteins(massTags, masstagMap);
return(database);
}
/// <summary>
/// Downloads the mass tags
/// </summary>
/// <returns></returns>
protected virtual List<MassTagLight> LoadMassTags()
{
var massTags = new List<MassTagLight>();
using (var connection = CreateConnection(CreateConnectionString()))
{
connection.Open();
using (var command = connection.CreateCommand())
{
SetupMassTagCommand(command);
try
{
using (var reader = command.ExecuteReader())
{
while (reader.Read())
{
var massTag = new MassTagLight();
if (reader["Mass_Tag_ID"] != DBNull.Value)
{
var id = Convert.ToInt32(reader["Mass_Tag_ID"]);
var peptide = "";
float ganet = -1;
float xcorr_max = 0;
float stdNet = 0;
var monoMass = 0.0;
float highDiscriminant = 0;
var numObservations = 0;
var modification = "";
var modCount = 0;
short cleaveageState = 2;
float driftTime = 0;
var charge = 0;
var conformerID = 0;
float highPeptideProphetProbability = 0;
double msgf = 0;
if (reader["Peptide"] != DBNull.Value) peptide = reader["Peptide"].ToString();
if (reader["Net_Value_to_Use"] != DBNull.Value)
ganet = Convert.ToSingle(reader["Net_Value_to_Use"]);
if (reader["High_Normalized_Score"] != DBNull.Value)
xcorr_max = Convert.ToSingle(reader["High_Normalized_Score"]);
if (reader["StD_GANET"] != DBNull.Value)
stdNet = Convert.ToSingle(reader["StD_GANET"]);
if (reader["Monoisotopic_Mass"] != DBNull.Value)
monoMass = Convert.ToDouble(reader["Monoisotopic_Mass"]);
if (reader["Min_MSGF_SpecProb"] != DBNull.Value)
msgf = Convert.ToDouble(reader["Min_MSGF_SpecProb"]);
if (reader["Peptide_Obs_Count_Passing_Filter"] != DBNull.Value)
numObservations = Convert.ToInt32(reader["Peptide_Obs_Count_Passing_Filter"]);
if (reader["Mod_Count"] != DBNull.Value)
modCount = Convert.ToInt32(reader["Mod_Count"]);
if (reader["Mod_Description"] != DBNull.Value)
modification = reader["Mod_Description"].ToString();
if (reader["High_Peptide_Prophet_Probability"] != DBNull.Value)
highPeptideProphetProbability =
Convert.ToSingle(reader["High_Peptide_Prophet_Probability"]);
if (reader["Cleavage_State"] != DBNull.Value)
cleaveageState = Convert.ToInt16(reader["Cleavage_State"]);
if (reader["Drift_Time_Avg"] != DBNull.Value)
driftTime = Convert.ToSingle(reader["Drift_Time_Avg"]);
if (reader["Conformer_Charge"] != DBNull.Value)
charge = Convert.ToInt32(reader["Conformer_Charge"]);
if (reader["Conformer_ID"] != DBNull.Value)
conformerID = Convert.ToInt32(reader["Conformer_ID"]);
/// Make sure the mass tag has been seen enough times
if (numObservations >= Options.MinimumObservationCountFilter)
{
var molecule = new Molecule();
molecule.Name = peptide;
massTag.Id = id;
massTag.Molecule = molecule;
massTag.Net = ganet;
massTag.NetAverage = ganet;
massTag.XCorr = xcorr_max;
massTag.DiscriminantMax = highDiscriminant;
massTag.MassMonoisotopic = monoMass;
massTag.ConformationId = conformerID;
massTag.NetStandardDeviation = stdNet;
massTag.ObservationCount = numObservations;
massTag.DriftTime = driftTime;
massTag.PriorProbability = highPeptideProphetProbability;
massTag.CleavageState = cleaveageState;
massTag.ModificationCount = modCount;
massTag.MsgfSpecProbMax = msgf;
massTag.PeptideSequence = peptide;
massTag.ChargeState = charge;
if (massTag.NetAverage != -1)
{
var shouldAdd = false;
// If we are using drift time, then we should only
// use mass tags that have drift time.
if (Options.OnlyLoadTagsWithDriftTime)
{
if (driftTime > 0)
{
shouldAdd = true;
}
}
else
{
shouldAdd = true;
}
if (shouldAdd)
{
massTags.Add(massTag);
}
}
}
}
}
reader.Close();
}
}
catch (Exception)
{
throw;
}
}
connection.Close();
}
return massTags;
}
public MassTagDatabase LoadDatabase()
{
var database = new MassTagDatabase();
//Implement the loading / reading of the MTDB Framework objects.
var reader = new SqLiteTargetDatabaseReader();
var mtdbDatabase = reader.ReadDb(m_path);
var massTags = new List<MassTagLight>();
// Mapping objects to create unique proteins and for the mass tag database to load only those proteins for
// the given consensus targets/mass tags.
var proteinMap = new Dictionary<int, Protein>();
var massTagProteinMap = new Dictionary<int, List<Protein>>();
if (mtdbDatabase == null)
return database;
foreach (var target in mtdbDatabase.ConsensusTargets)
{
// Copy the consensus data into a mass tag light.
var tag = new MassTagLight
{
Id = target.Id,
MassMonoisotopic = target.TheoreticalMonoIsotopicMass,
NetAverage = target.AverageNet,
NetPredicted = target.PredictedNet,
PeptideSequence = target.Sequence,
NetStandardDeviation = target.StdevNet
};
// Here we create unique proteins for the mass tag copying information from the consensus target proteins.
var proteinsForMassTag = new List<Protein>();
foreach (var targetProtein in target.Proteins)
{
if (!proteinMap.ContainsKey(targetProtein.Id))
{
var newProtein = new Protein
{
Id = targetProtein.Id,
Name = targetProtein.ProteinName,
ResidueStartPosition = targetProtein.ResidueStart,
ResidueEndPosition = targetProtein.ResidueEnd
};
//TODO: Do something about the cleavage state and terminus state of a protein loaded from MTDBCreator database.
//protein.CleavageState = ??
//protein.TerminusState = ??
proteinMap.Add(newProtein.Id, newProtein);
}
var protein = proteinMap[targetProtein.Id];
proteinsForMassTag.Add(protein);
}
massTagProteinMap.Add(tag.Id, proteinsForMassTag);
massTags.Add(tag);
}
database.AddMassTagsAndProteins(massTags, massTagProteinMap);
return database;
}
/// <summary>
/// Handles converting the rows to factor objects.
/// </summary>
/// <param name="sender"></param>
/// <param name="args"></param>
public void HandleDataRow(object sender, MageDataEventArgs args)
{
if (args == null)
{
throw new NullReferenceException("The mass tags are invalid.");
}
if (args.Fields == null)
{
return;
throw new NullReferenceException("The mass tag database rows are invalid.");
}
if (args.Fields.Length < 11)
{
return;
throw new ArgumentException("The number of columns for the mass tags are invalid.");
}
var tag = new MassTagLight();
if (m_columnMapping.ContainsKey("Mass"))
{
tag.MassMonoisotopic = Convert.ToDouble(args.Fields[m_columnMapping["Mass"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("NET"))
{
tag.NetAverage = Convert.ToDouble(args.Fields[m_columnMapping["NET"]]);
tag.Net = tag.NetAverage;
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Dataset_Member_Count"))
{
tag.ObservationCount = Convert.ToInt32(args.Fields[m_columnMapping["Dataset_Member_Count"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Cluster_ID"))
{
tag.Id = Convert.ToInt32(args.Fields[m_columnMapping["Cluster_ID"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Drift_Time"))
{
tag.DriftTime = Convert.ToDouble(args.Fields[m_columnMapping["Drift_Time"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Charge"))
{
tag.ChargeState = Convert.ToInt32(args.Fields[m_columnMapping["Charge"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Score"))
{
tag.Score = Convert.ToDouble(args.Fields[m_columnMapping["Score"]]);
}
else
{
return;
}
MassTags.Add(tag);
}
/// <summary>
/// Performs STAC against the mass tag database.
/// </summary>
public List <MultiAlignCore.Data.MassTags.FeatureMatchLight <T, MassTagLight> > PerformPeakMatching(List <T> clusters, MassTagDatabase database)
{
var clusterMap = new Dictionary <int, T>();
var tagMap = new Dictionary <int, Dictionary <int, MassTagLight> >();
var massTags = new List <MassTagLight>();
var i = 0;
foreach (var tag in database.MassTags)
{
var mt = new MassTagLight
{
Abundance = Convert.ToInt32(tag.Abundance),
ChargeState = tag.ChargeState,
CleavageState = tag.CleavageState,
ConformationId = tag.ConformationId,
ConformationObservationCount = tag.ConformationObservationCount,
DiscriminantMax = tag.DiscriminantMax,
DriftTime = Convert.ToSingle(tag.DriftTime),
DriftTimePredicted = tag.DriftTimePredicted,
Id = tag.Id,
MassMonoisotopic = tag.MassMonoisotopic,
ModificationCount = tag.ModificationCount,
Modifications = tag.Modifications,
Molecule = tag.Molecule,
MsgfSpecProbMax = tag.MsgfSpecProbMax,
Net = tag.NetAverage,
NetAverage = tag.NetAverage,
NetPredicted = tag.NetPredicted,
NetStandardDeviation = tag.NetStandardDeviation,
ObservationCount = tag.ObservationCount,
PeptideSequence = tag.PeptideSequence,
PeptideSequenceEx = tag.PeptideSequenceEx,
PriorProbability = tag.PriorProbability,
QualityScore = tag.QualityScore,
XCorr = tag.XCorr
};
mt.Index = i++;
massTags.Add(mt);
if (!tagMap.ContainsKey(tag.Id))
{
tagMap.Add(tag.Id, new Dictionary <int, MassTagLight>());
}
tagMap[tag.Id].Add(tag.ConformationId, tag);
}
// convert data needed by the algorithm.
var features = new List <UMCClusterLight>();
foreach (var cluster in clusters)
{
var feature = new UMCClusterLight
{
Id = cluster.Id,
MassMonoisotopicAligned = cluster.MassMonoisotopic,
MassMonoisotopic = cluster.MassMonoisotopic
};
feature.Net = cluster.Net;
feature.NetAligned = cluster.Net;
feature.ChargeState = cluster.ChargeState;
feature.DriftTime = Convert.ToSingle(cluster.DriftTime);
feature.DriftTimeAligned = Convert.ToDouble(cluster.DriftTime);
features.Add(feature);
clusterMap.Add(cluster.Id, cluster);
}
// create a stac manager and run.
var matcher = new FeatureMatcher <UMCClusterLight, MassTagLight>(features, massTags, Options);
matcher.MessageEvent += StatusHandler;
matcher.ProcessingCompleteEvent += StatusHandler;
matcher.MatchFeatures();
Matcher = matcher;
Matcher.PopulateStacfdrTable(matcher.MatchList);
var matches = new List <MultiAlignCore.Data.MassTags.FeatureMatchLight <T, MassTagLight> >();
foreach (var match in matcher.MatchList)
{
var matched = new MultiAlignCore.Data.MassTags.FeatureMatchLight <T, MassTagLight>
{
Observed = clusterMap[match.ObservedFeature.Id],
Target = tagMap[match.TargetFeature.Id][match.TargetFeature.ConformationId],
Confidence = match.STACScore,
Uniqueness = match.STACSpecificity
};
matches.Add(matched);
}
matcher.MessageEvent -= StatusHandler;
matcher.ProcessingCompleteEvent -= StatusHandler;
return(matches);
}
/// <summary>
/// Builds an ID for mapping and caching features
/// </summary>
/// <param name="tag"></param>
/// <returns></returns>
public static object BuildId(this MassTagLight tag)
{
return(string.Format("{0}-{1}", tag.Id, tag.ConformationId));
}
public MassTagDatabase LoadDatabase()
{
var database = new MassTagDatabase();
//Implement the loading / reading of the MTDB Framework objects.
var reader = new SqLiteTargetDatabaseReader();
var mtdbDatabase = reader.ReadDb(m_path);
var massTags = new List <MassTagLight>();
// Mapping objects to create unique proteins and for the mass tag database to load only those proteins for
// the given consensus targets/mass tags.
var proteinMap = new Dictionary <int, Protein>();
var massTagProteinMap = new Dictionary <int, List <Protein> >();
if (mtdbDatabase == null)
{
return(database);
}
foreach (var target in mtdbDatabase.ConsensusTargets)
{
// Copy the consensus data into a mass tag light.
var tag = new MassTagLight
{
Id = target.Id,
MassMonoisotopic = target.TheoreticalMonoIsotopicMass,
NetAverage = target.AverageNet,
NetPredicted = target.PredictedNet,
PeptideSequence = target.Sequence,
NetStandardDeviation = target.StdevNet
};
// Here we create unique proteins for the mass tag copying information from the consensus target proteins.
var proteinsForMassTag = new List <Protein>();
foreach (var targetProtein in target.Proteins)
{
if (!proteinMap.ContainsKey(targetProtein.Id))
{
var newProtein = new Protein
{
Id = targetProtein.Id,
Name = targetProtein.ProteinName,
ResidueStartPosition = targetProtein.ResidueStart,
ResidueEndPosition = targetProtein.ResidueEnd
};
//TODO: Do something about the cleavage state and terminus state of a protein loaded from MTDBCreator database.
//protein.CleavageState = ??
//protein.TerminusState = ??
proteinMap.Add(newProtein.Id, newProtein);
}
var protein = proteinMap[targetProtein.Id];
proteinsForMassTag.Add(protein);
}
massTagProteinMap.Add(tag.Id, proteinsForMassTag);
massTags.Add(tag);
}
database.AddMassTagsAndProteins(massTags, massTagProteinMap);
return(database);
}
/// <summary>
/// Handles converting the rows to factor objects.
/// </summary>
/// <param name="sender"></param>
/// <param name="args"></param>
public void HandleDataRow(object sender, MageDataEventArgs args)
{
if (args == null)
{
throw new NullReferenceException("The mass tags are invalid.");
}
if (args.Fields == null)
{
return;
throw new NullReferenceException("The mass tag database rows are invalid.");
}
if (args.Fields.Length < 11)
{
return;
throw new ArgumentException("The number of columns for the mass tags are invalid.");
}
var tag = new MassTagLight();
if (m_columnMapping.ContainsKey("Mass"))
{
tag.MassMonoisotopic = Convert.ToDouble(args.Fields[m_columnMapping["Mass"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("NET"))
{
tag.NetAverage = Convert.ToDouble(args.Fields[m_columnMapping["NET"]]);
tag.Net = tag.NetAverage;
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Dataset_Member_Count"))
{
tag.ObservationCount = Convert.ToInt32(args.Fields[m_columnMapping["Dataset_Member_Count"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Cluster_ID"))
{
tag.Id = Convert.ToInt32(args.Fields[m_columnMapping["Cluster_ID"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Drift_Time"))
{
tag.DriftTime = Convert.ToDouble(args.Fields[m_columnMapping["Drift_Time"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Charge"))
{
tag.ChargeState = Convert.ToInt32(args.Fields[m_columnMapping["Charge"]]);
}
else
{
return;
}
if (m_columnMapping.ContainsKey("Score"))
{
tag.Score = Convert.ToDouble(args.Fields[m_columnMapping["Score"]]);
}
else
{
return;
}
MassTags.Add(tag);
}
/// <summary>
/// Downloads the mass tags
/// </summary>
/// <returns></returns>
protected virtual List <MassTagLight> LoadMassTags()
{
var massTags = new List <MassTagLight>();
using (var connection = CreateConnection(CreateConnectionString()))
{
connection.Open();
using (var command = connection.CreateCommand())
{
SetupMassTagCommand(command);
try
{
using (var reader = command.ExecuteReader())
{
while (reader.Read())
{
var massTag = new MassTagLight();
if (reader["Mass_Tag_ID"] != DBNull.Value)
{
var id = Convert.ToInt32(reader["Mass_Tag_ID"]);
var peptide = string.Empty;
float ganet = -1;
var netDefined = false;
float xcorr_max = 0;
float stdNet = 0;
var monoMass = 0.0;
float highDiscriminant = 0;
var numObservations = 0;
var modCount = 0;
var modDescription = string.Empty;
short cleaveageState = 2;
float driftTime = 0;
var charge = 0;
var conformerID = 0;
float highPeptideProphetProbability = 0;
double msgf = 0;
if (reader["Peptide"] != DBNull.Value)
{
peptide = reader["Peptide"].ToString();
}
if (reader["Net_Value_to_Use"] != DBNull.Value)
{
ganet = Convert.ToSingle(reader["Net_Value_to_Use"]);
netDefined = true;
}
if (reader["High_Normalized_Score"] != DBNull.Value)
{
xcorr_max = Convert.ToSingle(reader["High_Normalized_Score"]);
}
if (netDefined && reader["StD_GANET"] != DBNull.Value)
{
stdNet = Convert.ToSingle(reader["StD_GANET"]);
}
if (reader["Monoisotopic_Mass"] != DBNull.Value)
{
monoMass = Convert.ToDouble(reader["Monoisotopic_Mass"]);
}
if (reader["Min_MSGF_SpecProb"] != DBNull.Value)
{
msgf = Convert.ToDouble(reader["Min_MSGF_SpecProb"]);
}
if (reader["Peptide_Obs_Count_Passing_Filter"] != DBNull.Value)
{
numObservations = Convert.ToInt32(reader["Peptide_Obs_Count_Passing_Filter"]);
}
if (reader["Mod_Count"] != DBNull.Value)
{
modCount = Convert.ToInt32(reader["Mod_Count"]);
}
if (reader["Mod_Description"] != DBNull.Value)
{
modDescription = reader["Mod_Description"].ToString();
}
// Note: for AMT tags from MSGF+ identifications, this value is actually 1 minus MSGFPlus_PValue
// In other words, if the PValue reported by MSGF+ is 2.469814E-07 the Peptide_Prophet_Probability is 0.9999998
if (reader["High_Peptide_Prophet_Probability"] != DBNull.Value)
{
highPeptideProphetProbability = Convert.ToSingle(reader["High_Peptide_Prophet_Probability"]);
}
if (reader["Cleavage_State"] != DBNull.Value)
{
cleaveageState = Convert.ToInt16(reader["Cleavage_State"]);
}
if (reader["Drift_Time_Avg"] != DBNull.Value)
{
driftTime = Convert.ToSingle(reader["Drift_Time_Avg"]);
}
if (reader["Conformer_Charge"] != DBNull.Value)
{
charge = Convert.ToInt32(reader["Conformer_Charge"]);
}
if (reader["Conformer_ID"] != DBNull.Value)
{
conformerID = Convert.ToInt32(reader["Conformer_ID"]);
}
// Make sure the mass tag has been seen enough times
if (numObservations >= Options.MinimumObservationCountFilter)
{
var molecule = new Molecule
{
Name = peptide
};
massTag.Id = id;
massTag.Molecule = molecule;
massTag.Net = ganet;
massTag.NetAverage = ganet;
massTag.NetStandardDeviation = stdNet;
massTag.XCorr = xcorr_max;
massTag.DiscriminantMax = highDiscriminant;
massTag.MassMonoisotopic = monoMass;
massTag.ConformationId = conformerID;
massTag.ObservationCount = numObservations;
massTag.DriftTime = driftTime;
massTag.PriorProbability = highPeptideProphetProbability;
massTag.CleavageState = cleaveageState;
massTag.ModificationCount = modCount;
massTag.Modifications = modDescription;
massTag.MsgfSpecProbMax = msgf;
massTag.PeptideSequence = peptide;
massTag.ChargeState = charge;
if (!netDefined)
{
continue;
}
var shouldAdd = false;
// If we are using drift time, then we should only
// use mass tags that have drift time.
if (massTag.Net >= Options.MinimumNet &&
massTag.Net <= Options.MaximumNet &&
massTag.MassMonoisotopic >= Options.MinimumMass &&
massTag.MassMonoisotopic <= Options.MaximumMass)
{
if (Options.OnlyLoadTagsWithDriftTime)
{
if (driftTime > 0)
{
shouldAdd = true;
}
}
else
{
shouldAdd = true;
}
}
if (shouldAdd)
{
massTags.Add(massTag);
}
}
}
}
reader.Close();
}
}
catch (Exception)
{
throw;
}
}
connection.Close();
}
return(massTags);
}
