util.bio.cluster
Class HierarchicalClustering

java.lang.Object
  util.bio.cluster.HierarchicalClustering

public class HierarchicalClustering

extends java.lang.Object

Calculates a Pearson correlation coefficient between all pairs of serialized float[]s, uses hierarchical clustering to group like arrays. Displays on screen and saves as a png.

Constructor Summary

HierarchicalClustering(double minRValue, int maxSizeToFlagCluster, java.lang.StringBuffer results, java.io.File pngResultFile, boolean displayClusterPlot)
For integration with CelFileQualityControl App.

HierarchicalClustering(java.io.File[] floatArrayFiles)
For integration with Correlate App.

HierarchicalClustering(java.io.File[] floatArrayFiles, java.io.File pngResultFile)
For integration with CelProcessor.

HierarchicalClustering(java.lang.String[] args)
For stand alone use.

Method Summary

void	breakOutClusters()
void	cluster()
java.io.File[]	clusterCelaFiles(java.io.File[] virtualCelFiles, int[][] pmCoordinates) Clusters cela files, extracts PM intensities, median normalizes.
void	clusterNormalizedFloatArrayFiles() Clusters float[] files, does not median normalize
java.io.File[]	clusterVirtualCelFiles(java.io.File[] virtualCelFiles, int[][] pmCoordinates) Performs actual clustering for CelFileQualityControl app.
boolean	displayClusterPlot()
double[]	findBestPair() Takes an ArrayList of VCluster and finds the best pair based on a PearsonCorrelationCoefficient.
java.util.ArrayList	getClusters()
Cluster	getMegaCluster()
double	getMinimalCorrelationCoefficient()
java.io.File	getPngResultFile()
java.lang.String	getTitle()
static void	main(java.lang.String[] args)
void	mergeClusters(double[] ccBestPair) Merges Clusters from best pair notes whether they have a bad cc.
static void	printDocs()
void	processArgs(java.lang.String[] args) This method will process each argument and assign new varibles
boolean	savePNG()
void	setDisplayClusterPlot(boolean displayClusterPlot)
void	setPngResultFile(java.io.File pngResultFile)
void	setSavePNG(boolean savePNG)
void	setTitle(java.lang.String panelTitle)

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

HierarchicalClustering

public HierarchicalClustering(java.lang.String[] args)

For stand alone use.

HierarchicalClustering

public HierarchicalClustering(double minRValue,
                              int maxSizeToFlagCluster,
                              java.lang.StringBuffer results,
                              java.io.File pngResultFile,
                              boolean displayClusterPlot)

For integration with CelFileQualityControl App.

HierarchicalClustering

public HierarchicalClustering(java.io.File[] floatArrayFiles)

For integration with Correlate App. Fires stand alone settings using the provided serialized float[] files.

HierarchicalClustering

public HierarchicalClustering(java.io.File[] floatArrayFiles,
                              java.io.File pngResultFile)

For integration with CelProcessor.

Method Detail

cluster

public void cluster()

clusterVirtualCelFiles

public java.io.File[] clusterVirtualCelFiles(java.io.File[] virtualCelFiles,
                                             int[][] pmCoordinates)

Performs actual clustering for CelFileQualityControl app.

clusterCelaFiles

public java.io.File[] clusterCelaFiles(java.io.File[] virtualCelFiles,
                                       int[][] pmCoordinates)

Clusters cela files, extracts PM intensities, median normalizes. Returns the names of virtual cel files with bad correlation coefficients.

clusterNormalizedFloatArrayFiles

public void clusterNormalizedFloatArrayFiles()

Clusters float[] files, does not median normalize

breakOutClusters

public void breakOutClusters()

mergeClusters

public void mergeClusters(double[] ccBestPair)

Merges Clusters from best pair notes whether they have a bad cc.

findBestPair

public double[] findBestPair()

Takes an ArrayList of VCluster and finds the best pair based on a PearsonCorrelationCoefficient. Returns a double[3]{cc, indexOne, indexTwo}. The HashMap contains calculated ccs. These will be used if the same comparison is made again.

processArgs

public void processArgs(java.lang.String[] args)

This method will process each argument and assign new varibles

printDocs

public static void printDocs()

main

public static void main(java.lang.String[] args)

getClusters

public java.util.ArrayList getClusters()

getMegaCluster

public Cluster getMegaCluster()

getMinimalCorrelationCoefficient

public double getMinimalCorrelationCoefficient()

savePNG

public boolean savePNG()

setSavePNG

public void setSavePNG(boolean savePNG)

getPngResultFile

public java.io.File getPngResultFile()

setPngResultFile

public void setPngResultFile(java.io.File pngResultFile)

getTitle

public java.lang.String getTitle()

setTitle

public void setTitle(java.lang.String panelTitle)

displayClusterPlot

public boolean displayClusterPlot()

setDisplayClusterPlot

public void setDisplayClusterPlot(boolean displayClusterPlot)