handlegraph::PathMetadata Class Reference

#include <path_metadata.hpp>

Inheritance diagram for handlegraph::PathMetadata:

Public Member Functions
virtual	~PathMetadata ()=default
virtual PathSense	get_sense (const path_handle_t &handle) const
	What is the given path meant to be representing? More...
virtual std::string	get_sample_name (const path_handle_t &handle) const
virtual std::string	get_locus_name (const path_handle_t &handle) const
virtual size_t	get_haplotype (const path_handle_t &handle) const
virtual size_t	get_phase_block (const path_handle_t &handle) const
virtual subrange_t	get_subrange (const path_handle_t &handle) const
template<typename Iteratee >
bool	for_each_path_of_sense (const PathSense &sense, const Iteratee &iteratee) const
template<typename Iteratee >
bool	for_each_path_of_sample (const std::string &sample, const Iteratee &iteratee) const
template<typename Iteratee >
bool	for_each_path_matching (const std::unordered_set< PathSense > *senses, const std::unordered_set< std::string > *samples, const std::unordered_set< std::string > *loci, const Iteratee &iteratee) const
template<typename Iteratee >
bool	for_each_path_matching (const std::unordered_set< PathSense > &senses, const std::unordered_set< std::string > &samples, const std::unordered_set< std::string > &loci, const Iteratee &iteratee) const
template<typename Iteratee >
bool	for_each_step_of_sense (const handle_t &visited, const PathSense &sense, const Iteratee &iteratee) const

Static Public Member Functions
static PathSense	parse_sense (const std::string &path_name)
static std::string	parse_sample_name (const std::string &path_name)
static std::string	parse_locus_name (const std::string &path_name)
static size_t	parse_haplotype (const std::string &path_name)
static size_t	parse_phase_block (const std::string &path_name)
static subrange_t	parse_subrange (const std::string &path_name)
static void	parse_path_name (const std::string &path_name, PathSense &sense, std::string &sample, std::string &locus, size_t &haplotype, size_t &phase_block, subrange_t &subrange)
	Decompose a formatted path name into metadata. More...
static std::string	create_path_name (const PathSense &sense, const std::string &sample, const std::string &locus, const size_t &haplotype, const size_t &phase_block, const subrange_t &subrange)

Static Public Attributes
static const std::string	NO_SAMPLE_NAME = ""
static const std::string	NO_LOCUS_NAME = ""
static const size_t	NO_HAPLOTYPE = std::numeric_limits<size_t>::max()
static const size_t	NO_PHASE_BLOCK = std::numeric_limits<size_t>::max()
static const subrange_t	NO_SUBRANGE {PathMetadata::NO_END_POSITION, PathMetadata::NO_END_POSITION}
static const offset_t	NO_END_POSITION = std::numeric_limits<offset_t>::max()

Protected Member Functions
virtual bool	for_each_path_matching_impl (const std::unordered_set< PathSense > *senses, const std::unordered_set< std::string > *samples, const std::unordered_set< std::string > *loci, const std::function< bool(const path_handle_t &)> &iteratee) const
virtual bool	for_each_step_of_sense_impl (const handle_t &visited, const PathSense &sense, const std::function< bool(const step_handle_t &)> &iteratee) const
virtual std::string	get_path_name (const path_handle_t &path_handle) const =0
	Look up the name of a path from a handle to it. More...
virtual path_handle_t	get_path_handle_of_step (const step_handle_t &step_handle) const =0
	Returns a handle to the path that an step is on. More...
virtual bool	for_each_path_handle_impl (const std::function< bool(const path_handle_t &)> &iteratee) const =0
virtual bool	for_each_step_on_handle_impl (const handle_t &handle, const std::function< bool(const step_handle_t &)> &iteratee) const =0

Static Private Attributes
static const std::regex	FORMAT
static const size_t	ASSEMBLY_OR_NAME_MATCH = 1
static const size_t	LOCUS_MATCH_NUMERICAL_WITHOUT_HAPLOTYPE = 2
static const size_t	HAPLOTYPE_MATCH = 2
static const size_t	LOCUS_MATCH_ANY = 3
static const size_t	PHASE_BLOCK_MATCH = 4
static const size_t	RANGE_START_MATCH = 5
static const size_t	RANGE_END_MATCH = 6
static const char	SEPARATOR = '#'
	Separator used to separate path name components. More...
static const char	RANGE_START_SEPARATOR = '['
static const char	RANGE_END_SEPARATOR = '-'
static const char	RANGE_TERMINATOR = ']'

Detailed Description

This is the interface for embedded path and haplotype thread metadata.

Comes with a default implementation of this interface, based on a get_path_name() and special path name formatting.

Our model is that paths come in different "senses":

• SENSE_GENERIC: a generic named path. Has a "locus" name.
• SENSE_REFERENCE: a part of a reference assembly. Has a "sample" name, a "locus" name, and a haplotype number.
• SENSE_HAPLOTYPE: a haplotype from a particular individual. Has a "sample" name, a "locus" name, a haplotype number, and a phase block identifier.

Paths of all sneses can represent subpaths, with bounds.

Depending on sense, a path might have:

• Sample: sample or assembly name.
• Locus: contig, scaffold, or gene name path either represents in its assembly or is an allele of in its sample.
• Haplotype number: number identifying which haplotype of a locus is being represented. GFA uses a convention where the presence of a haplotype 0 implies that only one haplotype is present.
• Phase block identifier: Distinguishes fragments of a haplotype that are phased but not necessarily part of a single self-consistent scaffold (often due to self-contradictory VCF information). Must be unique within a sample, locus, and haplotype. May be a number or a start coordinate.
• Bounds, for when a path as stored gives only a sub-range of a conceptually longer path. Multiple items can be stored with identical metadata in the other fields if their bounds are non-overlapping. TODO: Interaction with phase block in GBWT???

Constructor & Destructor Documentation

~PathMetadata()

virtual handlegraph::PathMetadata::~PathMetadata ( )

virtualdefault

Member Function Documentation

create_path_name()

std::string handlegraph::PathMetadata::create_path_name ( const PathSense &  sense, const std::string &  sample, const std::string &  locus, const size_t &  haplotype, const size_t &  phase_block, const subrange_t &  subrange  )

static

Compose a formatted path name for the given metadata. Any item can be the corresponding unset sentinel (PathMetadata::NO_LOCUS_NAME, PathMetadata::NO_PHASE_BLOCK, etc.).

for_each_path_handle_impl()

virtual bool handlegraph::PathMetadata::for_each_path_handle_impl ( const std::function< bool(const path_handle_t &)> &  iteratee ) const

protectedpure virtual

Execute a function on each path in the graph. If it returns false, stop iteration. Returns true if we finished and false if we stopped early.

If the graph contains compressed haplotype paths and properly implements for_each_path_of_sense to retrieve them, they should not be visible here. Only reference or generic named paths should be visible.

Implemented in handlegraph::PathHandleGraph, xg::XG, vg::PathSubgraphOverlay, and vg::MemoizingGraph.

for_each_path_matching() [1/2]

template<typename Iteratee >

bool handlegraph::PathMetadata::for_each_path_matching	(	const std::unordered_set< PathSense > &	senses,
		const std::unordered_set< std::string > &	samples,
		const std::unordered_set< std::string > &	loci,
		const Iteratee &	iteratee
	)		const

Loop through all the paths matching the given query. Query elements which are empty match everything. Returns false and stops if the iteratee returns false.

for_each_path_matching() [2/2]

template<typename Iteratee >

bool handlegraph::PathMetadata::for_each_path_matching	(	const std::unordered_set< PathSense > *	senses,
		const std::unordered_set< std::string > *	samples,
		const std::unordered_set< std::string > *	loci,
		const Iteratee &	iteratee
	)		const

Loop through all the paths matching the given query. Query elements which are null match everything. Returns false and stops if the iteratee returns false.

for_each_path_matching_impl()

bool handlegraph::PathMetadata::for_each_path_matching_impl ( const std::unordered_set< PathSense > *  senses, const std::unordered_set< std::string > *  samples, const std::unordered_set< std::string > *  loci, const std::function< bool(const path_handle_t &)> &  iteratee  ) const

protectedvirtual

Loop through all the paths matching the given query. Query elements which are null match everything. Returns false and stops if the iteratee returns false.

for_each_path_of_sample()

template<typename Iteratee >

bool handlegraph::PathMetadata::for_each_path_of_sample	(	const std::string &	sample,
		const Iteratee &	iteratee
	)		const

Loop through all the paths with the given sample name. Returns false and stops if the iteratee returns false.

for_each_path_of_sense()

template<typename Iteratee >

bool handlegraph::PathMetadata::for_each_path_of_sense	(	const PathSense &	sense,
		const Iteratee &	iteratee
	)		const

Loop through all the paths with the given sense. Returns false and stops if the iteratee returns false.

for_each_step_of_sense()

template<typename Iteratee >

bool handlegraph::PathMetadata::for_each_step_of_sense	(	const handle_t &	visited,
		const PathSense &	sense,
		const Iteratee &	iteratee
	)		const

Loop through all steps on the given handle for paths with the given sense. Returns false and stops if the iteratee returns false. TODO: Take the opportunity to make steps track orientation better?

for_each_step_of_sense_impl()

bool handlegraph::PathMetadata::for_each_step_of_sense_impl ( const handle_t &  visited, const PathSense &  sense, const std::function< bool(const step_handle_t &)> &  iteratee  ) const

protectedvirtual

Loop through all steps on the given handle for paths with the given sense. Returns false and stops if the iteratee returns false.

for_each_step_on_handle_impl()

virtual bool handlegraph::PathMetadata::for_each_step_on_handle_impl ( const handle_t &  handle, const std::function< bool(const step_handle_t &)> &  iteratee  ) const

protectedpure virtual

Execute a function on each step of a handle in any path. If it returns false, stop iteration. Returns true if we finished and false if we stopped early.

If the graph contains compressed haplotype paths and properly implements for_each_step_of_sense to find them, they should not be visible here. Only reference or generic named paths should be visible.

Implemented in handlegraph::PathHandleGraph, xg::XG, vg::PathSubgraphOverlay, and vg::MemoizingGraph.

get_haplotype()

size_t handlegraph::PathMetadata::get_haplotype ( const path_handle_t &  handle ) const

virtual

Get the haplotype number (0 or 1, for diploid) of the path-or-thread, or NO_HAPLOTYPE if it does not belong to one.

get_locus_name()

std::string handlegraph::PathMetadata::get_locus_name ( const path_handle_t &  handle ) const

virtual

Get the name of the contig or gene asociated with the path-or-thread, or NO_LOCUS_NAME if it does not belong to one.

get_path_handle_of_step()

virtual path_handle_t handlegraph::PathMetadata::get_path_handle_of_step ( const step_handle_t &  step_handle ) const

protectedpure virtual

Returns a handle to the path that an step is on.

Implemented in handlegraph::PathHandleGraph, xg::XG, vg::VG, vg::PathSubgraphOverlay, and vg::MemoizingGraph.

get_path_name()

virtual std::string handlegraph::PathMetadata::get_path_name ( const path_handle_t &  path_handle ) const

protectedpure virtual

Look up the name of a path from a handle to it.

Implemented in handlegraph::PathHandleGraph, xg::XG, vg::VG, vg::PathSubgraphOverlay, and vg::MemoizingGraph.

get_phase_block()

size_t handlegraph::PathMetadata::get_phase_block ( const path_handle_t &  handle ) const

virtual

Get the phase block number (contiguously phased region of a sample, contig, and haplotype) of the path-or-thread, or NO_PHASE_BLOCK if it does not belong to one.

get_sample_name()

std::string handlegraph::PathMetadata::get_sample_name ( const path_handle_t &  handle ) const

virtual

Get the name of the sample or assembly asociated with the path-or-thread, or NO_SAMPLE_NAME if it does not belong to one.

get_sense()

PathSense handlegraph::PathMetadata::get_sense ( const path_handle_t &  handle ) const

virtual

What is the given path meant to be representing?

get_subrange()

subrange_t handlegraph::PathMetadata::get_subrange ( const path_handle_t &  handle ) const

virtual

Get the bounds of the path-or-thread that are actually represented here. Should be NO_SUBRANGE if the entirety is represented here, and 0-based inclusive start and exclusive end positions of the stored region on the full path-or-thread if a subregion is stored.

If no end position is stored, NO_END_POSITION may be returned for the end position.

parse_haplotype()

size_t handlegraph::PathMetadata::parse_haplotype ( const std::string &  path_name )

static

Get the haplotype number (0 or 1, for diploid) embedded in the given formatted path name, or NO_HAPLOTYPE if it does not belong to one.

parse_locus_name()

std::string handlegraph::PathMetadata::parse_locus_name ( const std::string &  path_name )

static

Get the name of the contig or gene embedded in the given formatted path name, or NO_LOCUS_NAME if it does not belong to one.

parse_path_name()

void handlegraph::PathMetadata::parse_path_name ( const std::string &  path_name, PathSense &  sense, std::string &  sample, std::string &  locus, size_t &  haplotype, size_t &  phase_block, subrange_t &  subrange  )

static

Decompose a formatted path name into metadata.

parse_phase_block()

size_t handlegraph::PathMetadata::parse_phase_block ( const std::string &  path_name )

static

Get the phase block number (contiguously phased region of a sample, contig, and haplotype) embedded in the given formatted path name, or NO_PHASE_BLOCK if it does not belong to one.

parse_sample_name()

std::string handlegraph::PathMetadata::parse_sample_name ( const std::string &  path_name )

static

Get the name of the sample or assembly embedded in the given formatted path name, or NO_SAMPLE_NAME if it does not belong to one.

parse_sense()

PathSense handlegraph::PathMetadata::parse_sense ( const std::string &  path_name )

static

Extract the sense of a path from the given formatted path name, if possible. If not possible, return SENSE_GENERIC.

parse_subrange()

subrange_t handlegraph::PathMetadata::parse_subrange ( const std::string &  path_name )

static

Get the bounds embedded in the given formatted path name, or NO_SUBRANGE if they are absent. If no end position is stored, NO_END_POSITION may be returned for the end position.

Member Data Documentation

ASSEMBLY_OR_NAME_MATCH

const size_t handlegraph::PathMetadata::ASSEMBLY_OR_NAME_MATCH = 1

staticprivate

FORMAT

const std::regex handlegraph::PathMetadata::FORMAT

staticprivate

HAPLOTYPE_MATCH

const size_t handlegraph::PathMetadata::HAPLOTYPE_MATCH = 2

staticprivate

LOCUS_MATCH_ANY

const size_t handlegraph::PathMetadata::LOCUS_MATCH_ANY = 3

staticprivate

LOCUS_MATCH_NUMERICAL_WITHOUT_HAPLOTYPE

const size_t handlegraph::PathMetadata::LOCUS_MATCH_NUMERICAL_WITHOUT_HAPLOTYPE = 2

staticprivate

NO_END_POSITION

const offset_t handlegraph::PathMetadata::NO_END_POSITION = std::numeric_limits<offset_t>::max()

static

NO_HAPLOTYPE

const size_t handlegraph::PathMetadata::NO_HAPLOTYPE = std::numeric_limits<size_t>::max()

static

NO_LOCUS_NAME

const std::string handlegraph::PathMetadata::NO_LOCUS_NAME = ""

static

NO_PHASE_BLOCK

const size_t handlegraph::PathMetadata::NO_PHASE_BLOCK = std::numeric_limits<size_t>::max()

static

NO_SAMPLE_NAME

const std::string handlegraph::PathMetadata::NO_SAMPLE_NAME = ""

static

NO_SUBRANGE

const subrange_t handlegraph::PathMetadata::NO_SUBRANGE {PathMetadata::NO_END_POSITION, PathMetadata::NO_END_POSITION}

static

PHASE_BLOCK_MATCH

const size_t handlegraph::PathMetadata::PHASE_BLOCK_MATCH = 4

staticprivate

RANGE_END_MATCH

const size_t handlegraph::PathMetadata::RANGE_END_MATCH = 6

staticprivate

RANGE_END_SEPARATOR

const char handlegraph::PathMetadata::RANGE_END_SEPARATOR = '-'

staticprivate

RANGE_START_MATCH

const size_t handlegraph::PathMetadata::RANGE_START_MATCH = 5

staticprivate

RANGE_START_SEPARATOR

const char handlegraph::PathMetadata::RANGE_START_SEPARATOR = '['

staticprivate

RANGE_TERMINATOR

const char handlegraph::PathMetadata::RANGE_TERMINATOR = ']'

staticprivate

SEPARATOR

const char handlegraph::PathMetadata::SEPARATOR = '#'

staticprivate

Separator used to separate path name components.

---

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

• deps/libhandlegraph/src/include/handlegraph/path_metadata.hpp
• deps/libhandlegraph/src/path_metadata.cpp