umcg.genetica.math.stats

Class FDRUtils

java.lang.Object

umcg.genetica.math.stats.FDRUtils

public class FDRUtils
extends Object

Author:

juha

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	FDRUtils.AbsoluteValues
static class	FDRUtils.CloneArrays
static class	FDRUtils.SortArrays

Method Summary

Modifier and Type	Method and Description
static double	getGlobalThreshold(double[][] real, double[][][] permuted, double fdr, FDRUtils.CloneArrays c, FDRUtils.SortArrays s, FDRUtils.AbsoluteValues a) Returns the threshold value in the whole of given real data matrix that corresponds to the given false discovery rate against the null distribution from given permuted data matrices.
static double	getGlobalThreshold(double[][] real, double[][] permuted, double fdr, FDRUtils.CloneArrays c, FDRUtils.SortArrays s, FDRUtils.AbsoluteValues a) Returns the threshold value in the whole of given real data matrix that corresponds to the given false discovery rate against the null distribution from given permuted data matrix.
static int[]	getNrSignificant(double[][] data, double[] thresholds, FDRUtils.CloneArrays c, FDRUtils.SortArrays s, FDRUtils.AbsoluteValues a)
static int[]	getNrSignificant(double[][] data, double threshold, FDRUtils.SortArrays s, FDRUtils.AbsoluteValues a) Calculates the number of significant Z-scores per each row in the given Z-score matrix.
static double[]	getThresholdsPerRow(double[][] real, double[][][] permuted, double fdr, FDRUtils.CloneArrays c, FDRUtils.SortArrays s, FDRUtils.AbsoluteValues a) Returns the threshold values per row in the given real data matrix that corresponds to the given false discovery rate against the null distribution from rows of given permuted data matrices.
static double[]	getThresholdsPerRow(double[][] real, double[][] permuted, double fdr, FDRUtils.CloneArrays c, FDRUtils.SortArrays s, FDRUtils.AbsoluteValues a) Returns the threshold values per row in the given real data matrix that corresponds to the given false discovery rate against the null distribution from rows of given permuted data matrix.
static double[]	getThresholdsPerRow(double[][] real, String[] permutedFiles, double fdr, double mvp, FDRUtils.CloneArrays c, FDRUtils.SortArrays s, FDRUtils.AbsoluteValues a)
static double	zToP(double z) Returns the p-value corresponding to the given Z-score assuming a two tailed-test.

Methods inherited from class java.lang.Object

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

Method Detail

getGlobalThreshold

public static double getGlobalThreshold(double[][] real,
                                        double[][] permuted,
                                        double fdr,
                                        FDRUtils.CloneArrays c,
                                        FDRUtils.SortArrays s,
                                        FDRUtils.AbsoluteValues a)

Returns the threshold value in the whole of given real data matrix that corresponds to the given false discovery rate against the null distribution from given permuted data matrix.

Parameters:

real - real data [x][y]

permuted - permuted data [x][y]

fdr - false discovery rate to find

c - should the data be cloned or not (because the arrays are modified)

s - should the data be sorted first or are they already sorted

a - should absolute values be used

Returns:

largest threshold value in real data so that FDR is not larger than the given one

getNrSignificant

public static int[] getNrSignificant(double[][] data,
                                     double threshold,
                                     FDRUtils.SortArrays s,
                                     FDRUtils.AbsoluteValues a)

Calculates the number of significant Z-scores per each row in the given Z-score matrix. Z-scores are converted to p-values and the given p-value threshold is used as cutoff for significance. A two-tailed test is assumed.

Parameters:

data - a Z-score matrix

threshold - p-value threshold to use

s - should the data be sorted first or are they already sorted

a - should absolute values be used

Returns:

an array with the number of significant (according to the given p-value threshold) values in the matrix per row

zToP

public static double zToP(double z)

Returns the p-value corresponding to the given Z-score assuming a two tailed-test.

Parameters:

z - A Z-score to convert

Returns:

The p-value corresponding to the Z-score

getGlobalThreshold

public static double getGlobalThreshold(double[][] real,
                                        double[][][] permuted,
                                        double fdr,
                                        FDRUtils.CloneArrays c,
                                        FDRUtils.SortArrays s,
                                        FDRUtils.AbsoluteValues a)

Returns the threshold value in the whole of given real data matrix that corresponds to the given false discovery rate against the null distribution from given permuted data matrices.

Parameters:

real - real data [x][y]

permuted - permuted data [permutation][x][y]

fdr - false discovery rate to find

c - should the data be cloned or not (because the arrays are modified)

s - should the data be sorted first or are they already sorted

a - should absolute values be used

Returns:

largest threshold value in real data so that FDR is not larger than the given one

getThresholdsPerRow

public static double[] getThresholdsPerRow(double[][] real,
                                           double[][] permuted,
                                           double fdr,
                                           FDRUtils.CloneArrays c,
                                           FDRUtils.SortArrays s,
                                           FDRUtils.AbsoluteValues a)

Returns the threshold values per row in the given real data matrix that corresponds to the given false discovery rate against the null distribution from rows of given permuted data matrix.

Parameters:

real - real data [x][y]

permuted - permuted data [x][y]

fdr - false discovery rate to find

c - should the data be cloned or not (because the arrays are modified)

s - should the data be sorted first or are they already sorted

a - should absolute values be used

Returns:

largest threshold values per row in real data so that FDR is not larger than the given one

getThresholdsPerRow

public static double[] getThresholdsPerRow(double[][] real,
                                           double[][][] permuted,
                                           double fdr,
                                           FDRUtils.CloneArrays c,
                                           FDRUtils.SortArrays s,
                                           FDRUtils.AbsoluteValues a)

Returns the threshold values per row in the given real data matrix that corresponds to the given false discovery rate against the null distribution from rows of given permuted data matrices.

Parameters:

real - real data [x][y]

permuted - permuted data [permutation][x][y]

fdr - false discovery rate to find

c - should the data be cloned or not (because the arrays are modified)

s - should the data be sorted first or are they already sorted

a - should absolute values be used

Returns:

largest threshold values per row in real data so that FDR is not larger than the given one

getNrSignificant

public static int[] getNrSignificant(double[][] data,
                                     double[] thresholds,
                                     FDRUtils.CloneArrays c,
                                     FDRUtils.SortArrays s,
                                     FDRUtils.AbsoluteValues a)

getThresholdsPerRow

public static double[] getThresholdsPerRow(double[][] real,
                                           String[] permutedFiles,
                                           double fdr,
                                           double mvp,
                                           FDRUtils.CloneArrays c,
                                           FDRUtils.SortArrays s,
                                           FDRUtils.AbsoluteValues a)
                                    throws IOException

Throws:

IOException