full_dosages_results Class Reference

Stores full and leave-one-out (LOO) dosages for imputed variants. More...

#include <hidden_markov_model.hpp>

Public Member Functions
void	resize (std::size_t n_rows, std::size_t n_loo_rows, std::size_t n_columns)
	Resizes the dosage matrices to the specified dimensions.
void	fill_eov ()
	Fills all dosage matrices with the end-of-vector sentinel value.
std::array< std::size_t, 2 >	dimensions () const
	Returns the dimensions of the main dosages matrix.
std::array< std::size_t, 2 >	dimensions_loo () const
	Returns the dimensions of the leave-one-out dosages matrix.
float &	dosage (std::size_t i, std::size_t j)
	Accesses an element of the main dosages matrix (modifiable).
const float &	dosage (std::size_t i, std::size_t j) const
	Accesses an element of the main dosages matrix (const version).
float &	loo_dosage (std::size_t i, std::size_t j)
	Accesses an element of the leave-one-out dosages matrix (modifiable).
const float &	loo_dosage (std::size_t i, std::size_t j) const
	Accesses an element of the leave-one-out dosages matrix (const version).

Public Attributes
std::vector< std::vector< float > >	dosages_
	Matrix of imputed dosages.
std::vector< std::vector< float > >	loo_dosages_
	Matrix of leave-one-out (LOO) dosages.

Detailed Description

Stores full and leave-one-out (LOO) dosages for imputed variants.

This class holds the imputed genotype probabilities for all variants and haplotypes in the dataset. It provides storage for both standard dosages and leave-one-out dosages, which are used for model validation and cross-checking accuracy.

Member Function Documentation

dimensions()

std::array< std::size_t, 2 > full_dosages_results::dimensions ( ) const

inline

Returns the dimensions of the main dosages matrix.

Returns

A std::array of size 2:

• [0]: number of rows (dosages_.size())
• [1]: number of columns (dosages_[0].size() if not empty, 0 otherwise)

Here is the caller graph for this function:

dimensions_loo()

std::array< std::size_t, 2 > full_dosages_results::dimensions_loo ( ) const

inline

Returns the dimensions of the leave-one-out dosages matrix.

Returns

A std::array of size 2:

• [0]: number of rows (loo_dosages_.size())
• [1]: number of columns (loo_dosages_[0].size() if not empty, 0 otherwise)

dosage() [1/2]

float & full_dosages_results::dosage ( std::size_t i, std::size_t j )

inline

Accesses an element of the main dosages matrix (modifiable).

Parameters

i	Row index.
j	Column index.

Returns

Reference to the element at position (i, j) in dosages_.

dosage() [2/2]

const float & full_dosages_results::dosage ( std::size_t i, std::size_t j ) const

inline

Accesses an element of the main dosages matrix (const version).

Parameters

i	Row index.
j	Column index.

Returns

Const reference to the element at position (i, j) in dosages_.

fill_eov()

void full_dosages_results::fill_eov ( )

inline

Fills all dosage matrices with the end-of-vector sentinel value.

This function sets every element in the dosages_ and loo_dosages_ 2D vectors to a sentinel value returned by savvy::typed_value::end_of_vector_value<float>().

This is useful for reinitializing matrices before recalculation or ensuring that all entries are in a known "empty" state.

After calling this function:

• All elements in dosages_ are set to the sentinel value.
• All elements in loo_dosages_ are set to the sentinel value.

Here is the caller graph for this function:

loo_dosage() [1/2]

float & full_dosages_results::loo_dosage ( std::size_t i, std::size_t j )

inline

Accesses an element of the leave-one-out dosages matrix (modifiable).

Parameters

i	Row index.
j	Column index.

Returns

Reference to the element at position (i, j) in loo_dosages_.

loo_dosage() [2/2]

const float & full_dosages_results::loo_dosage ( std::size_t i, std::size_t j ) const

inline

Accesses an element of the leave-one-out dosages matrix (const version).

Parameters

i	Row index.
j	Column index.

Returns

Const reference to the element at position (i, j) in loo_dosages_.

resize()

void full_dosages_results::resize ( std::size_t n_rows, std::size_t n_loo_rows, std::size_t n_columns )

inline

Resizes the dosage matrices to the specified dimensions.

This function resizes two internal 2D vectors: dosages_ and loo_dosages_. Each element is initialized to a sentinel value returned by savvy::typed_value::end_of_vector_value<float>().

Parameters

n_rows	Number of rows for the main dosages_ matrix.
n_loo_rows	Number of rows for the leave-one-out loo_dosages_ matrix.
n_columns	Number of columns for both matrices.

After calling this function:

• dosages_ will have dimensions n_rows x n_columns.
• loo_dosages_ will have dimensions n_loo_rows x n_columns.
• All elements in both matrices are initialized to a sentinel value indicating the end of a vector, typically used for safe iteration or checking uninitialized entries.

Here is the caller graph for this function:

Member Data Documentation

dosages_

std::vector<std::vector<float> > full_dosages_results::dosages_

Matrix of imputed dosages.

Each row corresponds to a variant, and each column corresponds to a target haplotype or sample. Values are floating-point dosages in [0,1].

loo_dosages_

std::vector<std::vector<float> > full_dosages_results::loo_dosages_

Matrix of leave-one-out (LOO) dosages.

These dosages are computed excluding the current target haplotype from the reference panel, useful for cross-validation of imputation.

---

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

• src/hidden_markov_model.hpp