Computes Fis, Fst and Fit from diploid data. More...

#include <FStatistics.hpp>

Public Member Functions
	FStatistics ()
	Constructor. More...

virtual	~FStatistics ()
	Destructor. More...

void	reserve (unsigned int numberOfIndividuals)
	Reserve sufficient memory for a given number of individuals. More...

void	loadIndividual (unsigned int genotype1, unsigned int genotype2, unsigned int populationLabel)
	Loads the data for one individual. More...

unsigned int	populationLabel (unsigned int populationIndex)
	Label of a population. More...

unsigned int	alleleValue (unsigned int alleleIndex)
	Value of an allele. More...

unsigned int	firstAllele (unsigned int individualIndex) const
	First allele of a given individual (no checking)

unsigned int	secondAllele (unsigned int individualIndex) const
	Second allele of a given individual (no checking)

unsigned int	individualLabel (unsigned int individualIndex) const
	Population label of a given individual (no checking)

unsigned int	numberOfAlleles ()
	Number of alleles. More...

unsigned int	numberOfPopulations ()
	Number of populations. More...

unsigned int	numberOfGenotypes () const
	Number of loaded genotypes. More...

unsigned int	alleleFrequencyTotal (unsigned int alleleIndex)
	Absolute total allele frequency. More...

unsigned int	alleleFrequencyPerPopulation (unsigned int populationIndex, unsigned int alleleIndex)
	Absolute allele frequency in a population. More...

unsigned int	genotypeFrequencyTotal (unsigned int alleleIndex1, unsigned int alleleIndex2)
	Absolute genotype frequency. More...

unsigned int	genotypeFrequencyPerPopulation (unsigned int populationIndex, unsigned int alleleIndex1, unsigned int alleleIndex2)
	Absolute genotype frequency in a population. More...

unsigned int	populationFrequency (unsigned int populationIndex)
	Sample size of a population. More...

double	F ()
	Weir-Cockerham F-statistic. More...

double	theta ()
	Weir-Cockerham theta-statistic. More...

double	f ()
	Weir-Cockerham f-statistic. More...

double	Vpopulation ()
	Between-population component of variance. More...

double	Vindividual ()
	Within-population, between-individual component of variance. More...

double	Vallele ()
	Within-individual component of variance. More...

Protected Member Functions
void	d_init ()

void	d_clear ()

void	s_init ()

void	s_clear ()

void	s_compute ()

void	processPopulations ()

void	processAlleles ()

unsigned int	getPopulationIndex (unsigned int) const

unsigned int	getAlleleIndex (unsigned int) const

void	w_init ()

void	w_clear ()

void	w_compute ()

Protected Attributes
bool	d_flag

unsigned int	d_reserved

unsigned int	d_numberOfGenotypes

unsigned int *	d_genotypes

unsigned int *	d_populationLabels

bool	s_flag

unsigned int	s_numberOfAlleles

unsigned int *	s_alleleValueMapping

unsigned int	s_numberOfPopulations

unsigned int *	s_populationLabelMapping

unsigned int *	s_populationFrequencies

unsigned int *	s_alleleFrequenciesTotal

unsigned int **	s_alleleFrequenciesPerPopulation

unsigned int **	s_genotypeFrequenciesTotal

unsigned int ***	s_genotypeFrequenciesPerPopulation

bool	w_flag

double	w_F

double	w_T

double	w_f

double *	w_a

double *	w_b

double *	w_c

double	w_nbar

double	w_nc

double *	w_pbar

double *	w_ssquare

double *	w_hbar

double	w_sum_a

double	w_sum_b

double	w_sum_c

double	w_sum_abc

double	w_sum_bc

Detailed Description

Computes Fis, Fst and Fit from diploid data.

The class requires loading data. Data are loaded by individual (two genotypes per individual). The analyses are cached: they are performed upon the first call to statistics accessors. The cache is emptied whenever a datum is loaded.

The computations are performed after Weir and Cockerham. The statistics F, theta and f are generalized for multiple alleles. To allow computation of multi-locus statistics, variance components are also available. The three components of the variance are Vpopulation (between-population), Vindividual (within-population, between-individual) and Vallele (within- individual). The formulas to compute the F-statistics are as follows:

1-F = Vallele/(Vpopulation+Vindividual+Vallele)
theta = Vpopulation/(Vpopulation+Vindividual+Vallele)
1-f = Vallele/(Vindividual+Vallele).

Constructor & Destructor Documentation

FStatistics ( )

Constructor.

~FStatistics ( )

virtual

Destructor.

Member Function Documentation

unsigned int alleleFrequencyPerPopulation	(	unsigned int	populationIndex,
		unsigned int	alleleIndex
	)

Absolute allele frequency in a population.

unsigned int alleleFrequencyTotal ( unsigned int alleleIndex )

Absolute total allele frequency.

unsigned int alleleValue ( unsigned int alleleIndex )

Value of an allele.

The index corresponds to the local mapping of alleles regardless of the ranking of allele values. (No out of bound checking.)

double F ( )

Weir-Cockerham F-statistic.

Note: equivalent to Fit.

double f ( )

Weir-Cockerham f-statistic.

Note: equivalent to Fis.

unsigned int genotypeFrequencyPerPopulation	(	unsigned int	populationIndex,
		unsigned int	alleleIndex1,
		unsigned int	alleleIndex2
	)

Absolute genotype frequency in a population.

Note that allele AB is considered different to BA (this means that values can be accessed both sides of the diagonal.

unsigned int genotypeFrequencyTotal	(	unsigned int	alleleIndex1,
		unsigned int	alleleIndex2
	)

Absolute genotype frequency.

Note that allele AB is considered different to BA (this means that values can be accessed both sides of the diagonal.

void loadIndividual	(	unsigned int	genotype1,
		unsigned int	genotype2,
		unsigned int	populationLabel
	)

Loads the data for one individual.

Parameters

genotype1	an integer giving the first allele.
genotype2	an integer giving the second allele.
populationLabel	an integer indication belonging to a population.

Genotypes and population labels are not required to be consecutive (both are labels, not indices). They are internally mapped to indices (the mapping can be obtained by accessors populationLabel and allele).

All genotypes are considered to be valid (no missing data). If statistics were computed previous to call to this function, all data will be erase.

unsigned int numberOfAlleles ( )

Number of alleles.

unsigned int numberOfGenotypes ( ) const

Number of loaded genotypes.

unsigned int numberOfPopulations ( )

Number of populations.

unsigned int populationFrequency ( unsigned int populationIndex )

Sample size of a population.

unsigned int populationLabel ( unsigned int populationIndex )

Label of a population.

The index corresponds to the local mapping of populations regardless of the ranking of population labels. (No out of bound checking.)

void reserve ( unsigned int numberOfIndividuals )

Reserve sufficient memory for a given number of individuals.

This method makes the load function faster by allocating all required memory at once.

Parameters

numberOfIndividuals a strictly positive integer.

double theta ( )

Weir-Cockerham theta-statistic.

Note: equivalent to Fst.

double Vallele ( )

Within-individual component of variance.

double Vindividual ( )

Within-population, between-individual component of variance.

double Vpopulation ( )

Between-population component of variance.

The documentation for this class was generated from the following files:

/home/stephane/Documents/devel/egglib2/egglib-cpp/FStatistics.hpp
/home/stephane/Documents/devel/egglib2/egglib-cpp/FStatistics.cpp

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation