public void TestRestrictiveBoxMethod(string path, DistanceMetric dist, bool useBoxMethod)
{
var features = ReadFeatures(path);
var clusterer = new UMCAverageLinkageClusterer <UMCLight, UMCClusterLight>
{
ShouldTestClustersWithinTolerance = useBoxMethod,
Parameters =
{
CentroidRepresentation = ClusterCentroidRepresentation.Mean,
DistanceFunction = DistanceFactory <UMCLight> .CreateDistanceFunction(dist),
OnlyClusterSameChargeStates = true,
Tolerances =
{
Mass = 10,
DriftTime = .3,
Net = .03
}
}
};
var clusters = clusterer.Cluster(features);
var i = 0;
clusters.ForEach(x => x.Id = i++);
WriteClusters(clusters);
}
public IClusterer <T, U> Create(GenericClusteringAlgorithmType clusterType)
{
IClusterer <T, U> clusterer = null;
switch (clusterType)
{
case GenericClusteringAlgorithmType.AverageLinkage:
clusterer = new UMCAverageLinkageClusterer <T, U>();
break;
case GenericClusteringAlgorithmType.Centroid:
clusterer = new UMCCentroidClusterer <T, U>();
break;
case GenericClusteringAlgorithmType.SingleLinkage:
clusterer = new UMCSingleLinkageClusterer <T, U>();
break;
case GenericClusteringAlgorithmType.Prims:
clusterer = new UMCPrimsClustering <T, U>();
break;
case GenericClusteringAlgorithmType.BinarySearchTree:
clusterer = new Clustering.MsFeatureTreeClusterer <T, U>();
break;
default:
throw new ArgumentOutOfRangeException(string.Format("Cannot create generic {0} clusterer.", clusterType));
}
return(clusterer);
}
public void TestDistanceDistributions(string path, DistanceMetric dist)
{
var features = ReadFeatures(path);
var clusterer = new UMCAverageLinkageClusterer <UMCLight, UMCClusterLight>
{
ShouldTestClustersWithinTolerance = false,
Parameters =
{
CentroidRepresentation = ClusterCentroidRepresentation.Mean,
DistanceFunction = DistanceFactory <UMCLight> .CreateDistanceFunction(dist),
OnlyClusterSameChargeStates = true,
Tolerances =
{
Mass = 10,
DriftTime = .3,
Net = .03
}
}
};
var clusters = clusterer.Cluster(features);
var distances = new List <double>();
foreach (var cluster in clusters)
{
var centroid = new UMCLight();
centroid.MassMonoisotopicAligned = cluster.MassMonoisotopic;
centroid.Net = cluster.Net;
centroid.DriftTime = cluster.DriftTime;
var func = clusterer.Parameters.DistanceFunction;
foreach (var feature in cluster.Features)
{
var distance = func(feature, centroid);
distances.Add(distance);
}
distances.Sort();
var sum = 0;
foreach (var distance in distances)
{
sum++;
Console.WriteLine("{0},{1}", distance, sum);
}
}
}
public static IClusterer<UMCLight, UMCClusterLight> Create(LcmsFeatureClusteringAlgorithmType clusterType)
{
IClusterer<UMCLight, UMCClusterLight> clusterer = null;
switch (clusterType)
{
case LcmsFeatureClusteringAlgorithmType.AverageLinkage:
clusterer = new UMCAverageLinkageClusterer<UMCLight, UMCClusterLight>();
break;
case LcmsFeatureClusteringAlgorithmType.Centroid:
clusterer = new UMCCentroidClusterer<UMCLight, UMCClusterLight>();
break;
case LcmsFeatureClusteringAlgorithmType.SingleLinkage:
clusterer = new UMCSingleLinkageClusterer<UMCLight, UMCClusterLight>();
break;
case LcmsFeatureClusteringAlgorithmType.Prims:
clusterer = new UMCPrimsClustering<UMCLight, UMCClusterLight>();
break;
}
return clusterer;
}
//[TestCase(@"ClusterData\clusterData-merged-nodelin.txt")]
public void TestWeightedAverageLinkage(string path)
{
Console.WriteLine("Test: " + path);
var features = GetClusterData(Path.Combine(TestPaths.TestFilesDirectory, path));
Assert.IsNotEmpty(features);
var cluster = new UMCClusterLight();
cluster.Id = features[0].Id;
features.ForEach(x => cluster.AddChildFeature(x));
var maps = new Dictionary <int, UMCClusterLight>();
var average = new UMCAverageLinkageClusterer <UMCLight, UMCClusterLight>();
average.Parameters = new FeatureClusterParameters <UMCLight>();
average.Parameters.CentroidRepresentation = ClusterCentroidRepresentation.Mean;
average.Parameters.Tolerances = new Algorithms.FeatureTolerances();
var distance = new WeightedEuclideanDistance <UMCLight>();
average.Parameters.DistanceFunction = distance.EuclideanDistance;
var clusters = average.Cluster(features);
Console.WriteLine("dataset\tfeature\tmass\tnet\tdrift");
foreach (var newCluster in clusters)
{
foreach (var feature in newCluster.Features)
{
Console.WriteLine("{0},{1},{2},{3},{4}", feature.GroupId,
feature.Id,
feature.Net,
feature.MassMonoisotopicAligned,
feature.DriftTime);
}
}
}
/// <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);
}
}
//[TestCase(@"ClusterData\clusterData-single-1500.txt")]
public void TestAverageLinkage(string path)
{
Console.WriteLine("Average Linkage Test: " + path);
var features = GetClusterData(Path.Combine(TestPaths.TestFilesDirectory, path));
Assert.IsNotEmpty(features);
var cluster = new UMCClusterLight();
cluster.Id = features[0].Id;
features.ForEach(x => cluster.AddChildFeature(x));
var maps = new Dictionary <int, UMCClusterLight>();
var average = new UMCAverageLinkageClusterer <UMCLight, UMCClusterLight>();
average.Parameters = new FeatureClusterParameters <UMCLight>();
average.Parameters.CentroidRepresentation = ClusterCentroidRepresentation.Median;
average.Parameters.Tolerances = new Algorithms.FeatureTolerances();
average.Parameters.Tolerances.Net = .02;
average.Parameters.Tolerances.Mass = 6;
average.Parameters.Tolerances.DriftTime = .3;
var distance = new WeightedEuclideanDistance <UMCLight>();
average.Parameters.DistanceFunction = distance.EuclideanDistance;
var euclid = new EuclideanDistanceMetric <UMCLight>();
average.Parameters.DistanceFunction = euclid.EuclideanDistance;
var clusters = average.Cluster(features);
Console.WriteLine("Clusters = {0}", clusters.Count);
var id = 1;
foreach (var testCluster in clusters)
{
testCluster.CalculateStatistics(ClusterCentroidRepresentation.Mean);
var distances = new List <double>();
// Show a sampling of 5 results
var threshold = (int)(testCluster.Features.Count / (double)5);
if (threshold < 1)
{
threshold = 1;
}
testCluster.Id = id++;
var featureID = 0;
foreach (var feature in testCluster.Features)
{
featureID++;
if (featureID % threshold == 0)
{
Console.WriteLine("{0},{1},{2},{3}",
feature.Net,
feature.MassMonoisotopicAligned,
feature.DriftTime,
testCluster.Id);
}
var newDistance = distance.EuclideanDistance(feature, testCluster);
distances.Add(newDistance);
}
//Console.WriteLine();
//Console.WriteLine("Distances");
//distances.ForEach(x => Console.WriteLine(x));
//Console.WriteLine();
}
}
public void TestRestrictiveBoxMethod(string path, DistanceMetric dist, bool useBoxMethod)
{
var features = ReadFeatures(path);
var clusterer = new UMCAverageLinkageClusterer<UMCLight, UMCClusterLight>
{
ShouldTestClustersWithinTolerance = useBoxMethod,
Parameters =
{
CentroidRepresentation = ClusterCentroidRepresentation.Mean,
DistanceFunction = DistanceFactory<UMCLight>.CreateDistanceFunction(dist),
OnlyClusterSameChargeStates = true,
Tolerances =
{
Mass = 10,
DriftTime = .3,
Net = .03
}
}
};
var clusters = clusterer.Cluster(features);
var i = 0;
clusters.ForEach(x => x.Id = i++);
WriteClusters(clusters);
}
public void TestDistanceDistributions(string path, DistanceMetric dist)
{
var features = ReadFeatures(path);
var clusterer = new UMCAverageLinkageClusterer<UMCLight, UMCClusterLight>
{
ShouldTestClustersWithinTolerance = false,
Parameters =
{
CentroidRepresentation = ClusterCentroidRepresentation.Mean,
DistanceFunction = DistanceFactory<UMCLight>.CreateDistanceFunction(dist),
OnlyClusterSameChargeStates = true,
Tolerances =
{
Mass = 10,
DriftTime = .3,
Net = .03
}
}
};
var clusters = clusterer.Cluster(features);
var distances = new List<double>();
foreach (var cluster in clusters)
{
var centroid = new UMCLight();
centroid.MassMonoisotopicAligned = cluster.MassMonoisotopic;
centroid.Net = cluster.Net;
centroid.DriftTime = cluster.DriftTime;
var func = clusterer.Parameters.DistanceFunction;
foreach (var feature in cluster.Features)
{
var distance = func(feature, centroid);
distances.Add(distance);
}
distances.Sort();
var sum = 0;
foreach (var distance in distances)
{
sum++;
Console.WriteLine("{0},{1}", distance, sum);
}
}
}
/// <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 = PNNLOmics.Algorithms.Distance.DistanceFactory<UMCLight>.CreateDistanceFunction(PNNLOmics.Algorithms.Distance.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;
}
}