Class for storing parameters of assembling statistical model. Note: event with index 0 (zero) is "null" event and always has zero probability. Vector is ordered so for VJ or VDJ recombination specific family events stored in specific order. More...

#include <modelparametervector.h>

Public Member Functions
	ModelParameterVector (Recombination vec_type, const vector< prob_t > &param_vec, const vector< event_ind_t > &lens_vec, const vector< event_ind_t > &event_classes, const vector< seq_len_t > &event_family_col_numbers, const vector< prob_t > &laplace_vec=vector< prob_t >(), prob_t err_prob=0, bool do_normalise=true, const vector< seq_len_t > &d_genes_min_len=vector< seq_len_t >())

bool	operator== (const ModelParameterVector &other) const

Recombination	recombination () const

event_ind_t	eventFamilySize (event_ind_t event_family) const

event_ind_t	eventFamilySize (event_ind_t event_class, event_ind_t event_family) const

event_ind_t	eventClassSize (event_ind_t event_class) const

vector< prob_t >::const_iterator	get_iterator (event_ind_t i) const

event_ind_t	event_index (EventClass event_class, event_ind_t event_family, event_ind_t event_index) const

event_ind_t	event_index (EventClass event_class, event_ind_t event_family, event_ind_t event_row, event_ind_t event_column) const

prob_t	event_prob (EventClass event_class, event_ind_t event_family, event_ind_t event_index) const

prob_t	event_prob (EventClass event_class, event_ind_t event_family, event_ind_t event_row, event_ind_t event_column) const

void	normaliseEventFamilies ()
	Normalise each event family to have sum equal to 1.

void	normaliseEventClass (EventClass event_class)

event_ind_t	max_VJ_ins_len () const

event_ind_t	max_VD_ins_len () const

event_ind_t	max_DJ_ins_len () const

seq_len_t	D_min_len (event_ind_t d_index) const

prob_t	error_prob () const

void	set_error_prob (prob_t val)

void	fill (prob_t val=0)
	Fill the vector with the given value.

void	familyFill (event_ind_t event_family, prob_t val=0)

size_t	families () const

size_t	size () const

seq_len_t	n_columns (EventClass event_class, event_ind_t event_family=0) const


prob_t &	operator[] (event_ind_t gl_event_index)
	Get a probability of an event with the given global index. More...

prob_t	operator[] (event_ind_t gl_event_index) const

prob_t	operator() (event_ind_t event_family, event_ind_t loc_event_index) const

Detailed Description

Class for storing parameters of assembling statistical model. Note: event with index 0 (zero) is "null" event and always has zero probability. Vector is ordered so for VJ or VDJ recombination specific family events stored in specific order.

Vector with event's probabilities stored in specific order: VJ - recombination: [0] - null family [1] - Variable gene segment probability [2] - J prob [3 : 3 + [2] - [1]] - V deletions [4] - J deletions [5] - VJ insertion length [6] - VJ insertion markov chain

VDJ - recombination: [0] - null family [1] - Variable gene segment probability [2] - J prob [3] - V deletions [4] - J deletions [5] - VJ insertion length [6] - VJ insertion markov chain

Hypermutations added as the last element in the vector. Note: gene segment deletion probabilities stored in order of gene segment probabilities.

Constructor & Destructor Documentation

ymir::ModelParameterVector::ModelParameterVector	(	Recombination	vec_type,
		const vector< prob_t > &	param_vec,
		const vector< event_ind_t > &	lens_vec,
		const vector< event_ind_t > &	event_classes,
		const vector< seq_len_t > &	event_family_col_numbers,
		const vector< prob_t > &	laplace_vec = `vector<prob_t>()`,
		prob_t	err_prob = `0`,
		bool	do_normalise = `true`,
		const vector< seq_len_t > &	d_genes_min_len = `vector<seq_len_t>()`
	)

inline

Event family - family of specific events like deletions of V4 segment or DJ insertions. Event class - class of events like V deletions, J deletions, D deletions, VJ insertions.

Parameters

param_vec	Vector of probabilities of events.
lens_vec	Vector of lengths for each family of events (V1 deletions, V2 deletions, etc.)
event_classes	Vector of start indices in lens_vec of each class of events (all V deletions, all J deletions, VJ insertions, etc.)
event_family_col_numbers	Number of rows in each event family (zero means there is no matrix, but only a vector). Vector with the same length as lens_vec.
laplace_vec	Vector of pseudo counts for each event family.

Member Function Documentation

event_ind_t ymir::ModelParameterVector::event_index	(	EventClass	event_class,
		event_ind_t	event_family,
		event_ind_t	event_index
	)		const

inline

0-based indices for families and events! All segment indices must be passed as (segindex - 1), segment families as (segindex_deletion_index - 1), but deletions and insertions should be passed as it is (#deletions).

prob_t& ymir::ModelParameterVector::operator[] ( event_ind_t gl_event_index )

inline

Get a probability of an event with the given global index.

Parameters

event_family	Event family's index.
gl_event_index	Global index of the event.
loc_event_index	Local index of the event, i.e., index of the event in the given event family.

Returns: Probability of the event.

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

Ymir/src/modelparametervector.h

Public Member Functions

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation