/// <summary>
/// Starts a multi-Align analysis job.
/// </summary>
public void StartAnalysis(AnalysisConfig config)
{
m_config = config;
if (m_algorithms == null)
{
throw new NullReferenceException("The algorithms have not been set for this analysis.");
}
if (config.Analysis == null)
{
throw new NullReferenceException("The analysis data storage cannot be null.");
}
if (config.Analysis.DataProviders == null)
{
throw new NullReferenceException("The data cache providers have not been set for this analysis.");
}
Logger.LogPath = AnalysisPathUtils.BuildLogPath(config.AnalysisPath, config.AnalysisName);
// Make sure we start with a fresh analysis.
AbortAnalysisThread(m_analysisThread);
var threadStart = new ThreadStart(PerformAnalysis);
m_analysisThread = new Thread(threadStart);
m_analysisThread.Start();
}
public static FeatureDataAccessProviders CreateDataAccessProviders(MultiAlignAnalysis analysis,
bool deleteIfExists)
{
var path = AnalysisPathUtils.BuildAnalysisName(analysis.MetaData.AnalysisPath,
analysis.MetaData.AnalysisName);
return(CreateDataAccessProviders(path, deleteIfExists));
}
/// <summary>
/// The main entry point for the application.
/// </summary>
static int Main(string [] args)
{
var handle = System.Diagnostics.Process.GetCurrentProcess().MainWindowHandle;
SetConsoleMode(handle, ENABLE_EXTENDED_FLAGS);
try
{
if (args.Length < 2)
{
Console.WriteLine(@"MultiAlignChargeStateProcessor databasePath chargeState crossTabPath [dataset List]");
Console.WriteLine(@"\tThe cross-tab file will be placed in the same directory as the database path");
return(1);
}
// Setup the analysis processing
var databasePath = args[0];
var databaseName = Path.GetFileNameWithoutExtension(databasePath);
var path = Path.GetDirectoryName(databasePath);
var crossPath = args[2];
var chargeState = Convert.ToInt32(args[1]);
List <string> datasetList = null;
if (args.Length == 4)
{
datasetList = File.ReadAllLines(args[3]).ToList();
}
if (path == null)
{
Console.WriteLine(@"The directory path is invalid");
return(1);
}
NHibernateUtil.ConnectToDatabase(databasePath, false);
IDatasetDAO datasetCache = new DatasetDAOHibernate();
var dateSuffix = AnalysisPathUtils.BuildDateSuffix();
Logger.LogPath = Path.Combine(path, string.Format("{0}_charge_{2}_{1}.txt", databaseName, dateSuffix, chargeState));
Logger.PrintMessage("Find all datasets", true);
var datasets = datasetCache.FindAll();
Logger.PrintMessage(string.Format("Found {0} datasets", datasets.Count), true);
// Create the clustering algorithm - average linkage
IClusterer <UMCLight, UMCClusterLight> clusterer = new UMCAverageLinkageClusterer <UMCLight, UMCClusterLight>();
// Create the DAO object to extract the features
var database = new UmcAdoDAO {
DatabasePath = databasePath
};
IUmcDAO featureDao = database;
Logger.PrintMessage(string.Format("Extracting Features"), true);
var tempFeatures = featureDao.FindByCharge(chargeState);
Logger.PrintMessage(string.Format("Found {0} features", tempFeatures.Count), true);
var features = new List <UMCLight>();
if (datasetList != null)
{
var featuremap = datasets.ToDictionary(info => info.DatasetName.ToLower());
var focusedDatasetList = new Dictionary <int, DatasetInformation>();
foreach (var name in datasetList)
{
var key = name.ToLower();
if (featuremap.ContainsKey(key))
{
Logger.PrintMessage("Using dataset: " + name);
focusedDatasetList.Add(featuremap[key].DatasetId, featuremap[key]);
}
else
{
throw new Exception("Didn't find the dataset required..." + name);
}
}
features.AddRange(from feature in tempFeatures let use = focusedDatasetList.ContainsKey(feature.GroupId) where use select feature);
Logger.PrintMessage(string.Format("Found {0} filtered features for dataset list", features.Count), true);
}
else
{
features = tempFeatures;
}
// Handle logging progress.
clusterer.Progress += clusterer_Progress;
clusterer.Parameters.Tolerances.DriftTime = .3;
clusterer.Parameters.Tolerances.Mass = 16;
clusterer.Parameters.Tolerances.Net = .014;
clusterer.Parameters.OnlyClusterSameChargeStates = true;
clusterer.Parameters.CentroidRepresentation = ClusterCentroidRepresentation.Mean;
clusterer.Parameters.DistanceFunction = DistanceFactory <UMCLight> .CreateDistanceFunction(DistanceMetric.WeightedEuclidean);
// Then cluster
var clusterWriter = new UmcClusterWriter();
IClusterWriter <UMCClusterLight> writer = clusterWriter; //new UMCClusterDummyWriter();
try
{
clusterWriter.Open(crossPath);
clusterWriter.WriteHeader(datasets);
clusterer.ClusterAndProcess(features, writer);
Logger.PrintMessage("", true);
Logger.PrintMessage("ANALYSIS SUCCESS", true);
return(0);
}
catch (Exception ex)
{
Logger.PrintMessage("Unhandled Error: " + ex.Message);
var innerEx = ex.InnerException;
while (innerEx != null)
{
Logger.PrintMessage("Inner Exception: " + innerEx.Message);
innerEx = innerEx.InnerException;
}
Logger.PrintMessage("Stack: " + ex.StackTrace);
Logger.PrintMessage("");
Logger.PrintMessage("ANALYSIS FAILED");
return(1);
}
finally
{
clusterWriter.Close();
}
}
catch (Exception ex)
{
Logger.PrintMessage("Unhandled Error: " + ex.Message, true);
var innerEx = ex.InnerException;
while (innerEx != null)
{
Logger.PrintMessage("Inner Exception: " + innerEx.Message);
innerEx = innerEx.InnerException;
}
Logger.PrintMessage("Stack: " + ex.StackTrace, true);
Logger.PrintMessage("");
Logger.PrintMessage("ANALYSIS FAILED");
return(1);
}
}
/// <summary>
/// The main entry point for the application.
/// </summary>
static int Main(string[] args)
{
var handle = System.Diagnostics.Process.GetCurrentProcess().MainWindowHandle;
SetConsoleMode(handle, ENABLE_EXTENDED_FLAGS);
var clusterIdMap = new Dictionary <int, UMCClusterLight>();
try
{
if (args.Length < 2)
{
Console.WriteLine("MultiAlignSTACRunner crosstabDirectory databasePath outputName");
Console.WriteLine("\tThe cross-tab file will be named similar to the database path");
return(1);
}
// Setup the analysis processing
Logger.PrintMessage("Find all datasets", true);
var directoryPath = args[0].Replace("\r", "").Replace("\n", "");
var databasePath = args[1].Replace("\r", "").Replace("\n", "");
var name = args[2].Replace("\r", "").Replace("\n", "");
var files = Directory.GetFiles(directoryPath, "*.csv");
Logger.PrintMessage("Creating Log File");
var loggerPath = AnalysisPathUtils.BuildLogPath(Path.GetDirectoryName(name), Path.GetFileNameWithoutExtension(name));
Logger.LogPath = loggerPath;
Logger.PrintMessage("Saving Log Data to: " + loggerPath);
Logger.PrintMessage(string.Format("Creating STAC"), true);
// Hardcode bad, but per discussions with OHSU
var stac = new STACAdapter <UMCClusterLight>
{
Options = new FeatureMatcherParameters
{
UserTolerances =
{
MassTolerancePPM = 25,
NETTolerance = .035,
DriftTimeTolerance = 3
},
UseDriftTime = true
}
};
var clusterFilteringOptions = new FilteringOptions();
WriteOptionsToLogFile(clusterFilteringOptions);
WriteOptionsToLogFile(stac.Options);
// Read the cluster data
var allClusters = new List <UMCClusterLight>();
var clusterId = 0;
foreach (var file in files)
{
var filename = Path.GetFileName(file);
Logger.PrintMessage(string.Format("Reading {0}", filename));
const string chargeString = "charge";
if (filename == null)
{
continue;
}
if (!filename.Contains(chargeString))
{
continue;
}
// Read each file.
var start = DateTime.Now;
var xname = filename.Replace("_", "");
// ReSharper disable once StringIndexOfIsCultureSpecific.1
var indexOfChargeString = xname.IndexOf(chargeString);
var charge = Convert.ToInt32(xname.Substring(indexOfChargeString + chargeString.Length, 1));
var clusters = ReadClusters(file, clusterFilteringOptions);
var end = DateTime.Now;
Logger.PrintMessage(string.Format("\tReading Took {0:.00} seconds", end.Subtract(start).TotalSeconds));
foreach (var cluster in clusters)
{
clusterIdMap.Add(clusterId, cluster);
cluster.Id = clusterId++;
cluster.ChargeState = charge;
}
allClusters.AddRange(clusters);
}
// Load the database
Logger.PrintMessage(string.Format("Loading Mass Tag Database: {0}", Path.GetFileNameWithoutExtension(databasePath)));
var options = new MassTagDatabaseOptions();
var databaseDefinition = new InputDatabase
{
DatabaseFormat = MassTagDatabaseFormat.Sqlite,
LocalPath = databasePath
};
var database = MtdbLoaderFactory.LoadMassTagDatabase(databaseDefinition, options);
// Run stac
try
{
Logger.PrintMessage("Matching clusters to peptides in mass tag database.");
var matches = stac.PerformPeakMatching(allClusters, database);
Logger.PrintMessage(string.Format("Writing Results To {0}", name));
var duplicateMatches = new Dictionary <int, Dictionary <int, FeatureMatchLight <UMCClusterLight, MassTagLight> > >();
foreach (var match in matches)
{
if (!duplicateMatches.ContainsKey(match.Observed.Id))
{
duplicateMatches.Add(match.Observed.Id, new Dictionary <int, FeatureMatchLight <UMCClusterLight, MassTagLight> >());
}
if (!duplicateMatches[match.Observed.Id].ContainsKey(match.Target.Id))
{
duplicateMatches[match.Observed.Id].Add(match.Target.Id, match);
}
}
WriteClusterData(name, duplicateMatches, clusterIdMap);
Logger.PrintMessage("ANALYSIS SUCCESS");
return(0);
}
catch (Exception ex)
{
Logger.PrintMessage("Unhandled Error: " + ex.Message);
var innerEx = ex.InnerException;
while (innerEx != null)
{
Logger.PrintMessage("Inner Exception: " + innerEx.Message);
innerEx = innerEx.InnerException;
}
Logger.PrintMessage("Stack: " + ex.StackTrace);
Logger.PrintMessage("");
Logger.PrintMessage("ANALYSIS FAILED");
return(1);
}
}
catch (Exception ex)
{
Logger.PrintMessage("Unhandled Error: " + ex.Message, true);
var innerEx = ex.InnerException;
while (innerEx != null)
{
Logger.PrintMessage("Inner Exception: " + innerEx.Message);
innerEx = innerEx.InnerException;
}
Logger.PrintMessage("Stack: " + ex.StackTrace, true);
Logger.PrintMessage("");
Logger.PrintMessage("ANALYSIS FAILED");
return(1);
}
}
