Package ubic.gemma.core.analysis.expression.coexpression.links

Class AbstractMatrixRowPairAnalysis

java.lang.Object

ubic.gemma.core.analysis.expression.coexpression.links.AbstractMatrixRowPairAnalysis

All Implemented Interfaces:

MatrixRowPairAnalysis

Direct Known Subclasses:

PearsonMetrics

public abstract class AbstractMatrixRowPairAnalysis
extends Object
implements MatrixRowPairAnalysis

Author:

paul

Field Summary

Fields

Modifier and Type	Field	Description
static int	HARD_LIMIT_MIN_NUM_USED	Absolute lower limit to minNumUsed.

Fields inherited from interface ubic.gemma.core.analysis.expression.coexpression.links.MatrixRowPairAnalysis

NUM_BINS

Constructor Summary

Constructors

Constructor	Description
AbstractMatrixRowPairAnalysis()

Method Summary

Modifier and Type	Method	Description
long	getCrossHybridizationRejections()	Use after analysis.
DoubleArrayList	getHistogramArrayList()	Read back the histogram as a DoubleArrayList of counts.
ObjectArrayList	getKeepers()	Identify the correlations that are above the set thresholds.
Matrix2D<ExpressionDataMatrixRowElement,ExpressionDataMatrixRowElement,Double>	getMatrix()
double	getNumUniqueGenes()
CompositeSequence	getProbeForRow(ExpressionDataMatrixRowElement rowEl)
double	getScoreInBin(int i)
boolean	isUsePvalueThreshold()
double	kurtosis()
void	nullMatrix()	Flag the correlation matrix as un-fillable.
int	numCached()
protected abstract void	rowStatistics()
void	setDuplicateMap(Map<CompositeSequence,Set<Gene>> probeToGeneMap)
void	setLowerTailThreshold(double k)	Set the threshold, below which correlations are kept (e.g., negative values)
void	setMinNumpresent(int minSamplesToKeepCorrelation)	Set the number of mutually present values in a pairwise comparison that must be attained before the correlation is stored.
void	setOmitNegativeCorrelationLinks(boolean omitNegativeCorrelationLinks)
void	setPValueThreshold(double k)
void	setUpperTailThreshold(double k)	Set the threshold, above which correlations are kept.
void	setUseAbsoluteValue(boolean k)	If set to true, then the absolute value of the correlation is used for histograms and choosing correlations to keep.
void	setUsePvalueThreshold(boolean usePvalueThreshold)	Default is true; set to false to disable use of the pvalue threshold, in which case only the upper and lower tail thresholds will be used.
int	size()
String	toString()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface ubic.gemma.core.analysis.expression.coexpression.links.MatrixRowPairAnalysis

calculateMetrics, getMetricType

Field Detail

HARD_LIMIT_MIN_NUM_USED

public static final int HARD_LIMIT_MIN_NUM_USED

Absolute lower limit to minNumUsed. (This used to be 3, and then 5). It doesn't make much sense to set this higher than ExpressionExperimentFilter.MIN_NUMBER_OF_SAMPLES_PRESENT

See Also:

Constant Field Values

Constructor Detail

AbstractMatrixRowPairAnalysis

public AbstractMatrixRowPairAnalysis()

Method Detail

getHistogramArrayList

public DoubleArrayList getHistogramArrayList()

Read back the histogram as a DoubleArrayList of counts.

Specified by:

getHistogramArrayList in interface MatrixRowPairAnalysis

Returns:

cern.colt.list.DoubleArrayList

getKeepers

public ObjectArrayList getKeepers()

Identify the correlations that are above the set thresholds.

Specified by:

getKeepers in interface MatrixRowPairAnalysis

Returns:

cern.colt.list.ObjectArrayList

getProbeForRow

public CompositeSequence getProbeForRow(ExpressionDataMatrixRowElement rowEl)

Specified by:

getProbeForRow in interface MatrixRowPairAnalysis

getScoreInBin

public double getScoreInBin(int i)

Specified by:

getScoreInBin in interface MatrixRowPairAnalysis

numCached

public int numCached()

Specified by:

numCached in interface MatrixRowPairAnalysis

Returns:

The number of values stored in the correlation matrix.

setDuplicateMap

public void setDuplicateMap(Map<CompositeSequence,Set<Gene>> probeToGeneMap)

Specified by:

setDuplicateMap in interface MatrixRowPairAnalysis

setLowerTailThreshold

public void setLowerTailThreshold(double k)

Set the threshold, below which correlations are kept (e.g., negative values)

Specified by:

setLowerTailThreshold in interface MatrixRowPairAnalysis

Parameters:

k - double

setMinNumpresent

public void setMinNumpresent(int minSamplesToKeepCorrelation)

Set the number of mutually present values in a pairwise comparison that must be attained before the correlation is stored. Note that you cannot set the value less than HARD_LIMIT_MIN_NUM_USED.

Specified by:

setMinNumpresent in interface MatrixRowPairAnalysis

setOmitNegativeCorrelationLinks

public void setOmitNegativeCorrelationLinks(boolean omitNegativeCorrelationLinks)

Specified by:

setOmitNegativeCorrelationLinks in interface MatrixRowPairAnalysis

Parameters:

omitNegativeCorrelationLinks - the omitNegativeCorrelationLinks to set

setPValueThreshold

public void setPValueThreshold(double k)

Specified by:

setPValueThreshold in interface MatrixRowPairAnalysis

Parameters:

k - double

setUpperTailThreshold

public void setUpperTailThreshold(double k)

Set the threshold, above which correlations are kept.

Specified by:

setUpperTailThreshold in interface MatrixRowPairAnalysis

Parameters:

k - double

setUseAbsoluteValue

public void setUseAbsoluteValue(boolean k)

If set to true, then the absolute value of the correlation is used for histograms and choosing correlations to keep. The correlation matrix, if actually used to store all the values, maintains the actual number.

Specified by:

setUseAbsoluteValue in interface MatrixRowPairAnalysis

Parameters:

k - boolean

getMatrix

public Matrix2D<ExpressionDataMatrixRowElement,ExpressionDataMatrixRowElement,Double> getMatrix()

Returns:

baseCode.dataStructure.NamedMatrix

isUsePvalueThreshold

public boolean isUsePvalueThreshold()

Returns:

the usePvalueThreshold

setUsePvalueThreshold

public void setUsePvalueThreshold(boolean usePvalueThreshold)

Description copied from interface: MatrixRowPairAnalysis

Default is true; set to false to disable use of the pvalue threshold, in which case only the upper and lower tail thresholds will be used.

Specified by:

setUsePvalueThreshold in interface MatrixRowPairAnalysis

Parameters:

usePvalueThreshold - the usePvalueThreshold to set

getCrossHybridizationRejections

public long getCrossHybridizationRejections()

Description copied from interface: MatrixRowPairAnalysis

Use after analysis.

Specified by:

getCrossHybridizationRejections in interface MatrixRowPairAnalysis

Returns:

how many pairs were rejected due to cross-hybridization potential

getNumUniqueGenes

public double getNumUniqueGenes()

Specified by:

getNumUniqueGenes in interface MatrixRowPairAnalysis

size

public int size()

Specified by:

size in interface MatrixRowPairAnalysis

Returns:

how many rows/columns the matrix has.

kurtosis

public double kurtosis()

Returns:

double

nullMatrix

public void nullMatrix()

Flag the correlation matrix as un-fillable. This means that when PearsonMatrix is called, only the histogram will be filled in. Also trashes any values that might have been stored there.

toString

public String toString()

Overrides:

toString in class Object

Returns:

java.lang.String

rowStatistics

protected abstract void rowStatistics()