de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous
Class TwoPointEvaluater

java.lang.Object
  de.jstacs.sequenceScores.statisticalModels.trainable.discrete.inhomogeneous.TwoPointEvaluater

public class TwoPointEvaluater
extends Object

This class is for visualizing two point dependency between sequence positions.

Author:

Jens Keilwagen

Constructor Summary

TwoPointEvaluater()

Method Summary

static BufferedImage	getImage(DataSet d, double[] weights, REnvironment r, double alpha, int... borders)
static BufferedImage	getImage(double[][] mi, REnvironment r) This method can be used to create an image of a mutual information matrix.
static double	getMax(double[][] mi) This method can be used to determine the maximal value of the matrix of mutual informations.
static double[][]	getMI(DataSet s, double[] weights) This method computes the pairwise mutual information between the sequence positions.
static double[][]	getMIInBits(DataSet s, double[] weights) This method computes the pairwise mutual information (in bits) between the sequence positions.

Methods inherited from class java.lang.Object

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

Constructor Detail

TwoPointEvaluater

public TwoPointEvaluater()

Method Detail

getMI

public static double[][] getMI(DataSet s,
                               double[] weights)
                        throws IllegalArgumentException

This method computes the pairwise mutual information between the sequence positions.

Parameters:

s - the DataSet

weights - the weights for each sequence of the DataSet or null

Returns:

a matrix containing all pairwise mutual informations

Throws:

IllegalArgumentException - if something went wrong (e.g. the AlphabetContainer is not discrete, the length is not matching with the AlphabetContainer, ...)

getMIInBits

public static double[][] getMIInBits(DataSet s,
                                     double[] weights)
                              throws IllegalArgumentException

This method computes the pairwise mutual information (in bits) between the sequence positions.

Parameters:

s - the DataSet

weights - the weights for each sequence of the DataSet or null

Returns:

a matrix containing all pairwise mutual informations (in bits)

Throws:

IllegalArgumentException - if something went wrong (e.g. the AlphabetContainer is not discrete, the length is not matching with the AlphabetContainer, ...)

getImage

public static BufferedImage getImage(double[][] mi,
                                     REnvironment r)
                              throws Exception

This method can be used to create an image of a mutual information matrix.

Parameters:

mi - the matrix of mutual information

r - the R environment

Returns:

an image of the matrix

Throws:

Exception - if an Exception is thrown from REnvironment

See Also:

getMIInBits(DataSet, double[])

getImage

public static BufferedImage getImage(DataSet d,
                                     double[] weights,
                                     REnvironment r,
                                     double alpha,
                                     int... borders)
                              throws Exception

Throws:

Exception

getMax

public static double getMax(double[][] mi)

This method can be used to determine the maximal value of the matrix of mutual informations.

Parameters:

mi - the matrix of mutual informations

Returns:

the maximal value