Represent a Colored and Compacted de Bruijn graph. More...

Inherits CompactedDBG< DataAccessor< void >, DataStorage< void > >.

Public Member Functions
	ColoredCDBG (int kmer_length=31, int minimizer_length=-1)
	Constructor (set up an empty colored cdBG). More...

	ColoredCDBG (const ColoredCDBG &o)
	Copy constructor (copy a colored cdBG). More...

	ColoredCDBG (ColoredCDBG &&o)
	Move constructor (move a colored cdBG). More...

ColoredCDBG &	operator= (const ColoredCDBG &o)
	Copy assignment operator (copy a colored cdBG). More...

ColoredCDBG &	operator= (ColoredCDBG &&o)
	Move assignment operator (move a colored cdBG). More...

bool	operator== (const ColoredCDBG &o) const
	Equality operator. More...

bool	operator!= (const ColoredCDBG &o) const
	Inequality operator. More...

ColoredCDBG &	operator+= (const ColoredCDBG &o)
	Addition assignment operator (merge a colored cdBG). More...

void	clear ()
	Clear the graph: remove unitigs, user data and colors + reset its parameters.

bool	buildGraph (const CCDBG_Build_opt &opt)
	Build the Colored and compacted de Bruijn graph (only the unitigs). More...

bool	buildColors (const CCDBG_Build_opt &opt)
	Map the colors to the unitigs. More...

bool	write (const string &prefix_output_fn, const size_t nb_threads=1, const bool write_index_file=true, const bool compress_output=false, const bool verbose=false) const
	Write a colored and compacted de Bruijn graph to disk. More...

bool	read (const string &input_graph_fn, const string &input_colors_fn, const size_t nb_threads=1, const bool verbose=false)
	Read a colored and compacted de Bruijn graph from disk. More...

bool	read (const string &input_graph_fn, const string &input_index_fn, const string &input_colors_fn, const size_t nb_threads=1, const bool verbose=false)
	Read a colored and compacted de Bruijn graph from disk using an index file. More...

bool	merge (const ColoredCDBG &o, const size_t nb_threads=1, const bool verbose=false)
	Merge a colored and compacted de Bruijn graph. More...

bool	merge (ColoredCDBG &&o, const size_t nb_threads=1, const bool verbose=false)
	Merge and clear a colored and compacted de Bruijn graph. More...

bool	merge (const vector< ColoredCDBG > &v, const size_t nb_threads=1, const bool verbose=false)
	Merge multiple colored and compacted de Bruijn graphs. More...

bool	merge (vector< ColoredCDBG > &&v, const size_t nb_threads=1, const bool verbose=false)
	Merge and clear multiple colored and compacted de Bruijn graphs. More...

string	getColorName (const size_t color_id) const
	Get the name of a color. More...

vector< string >	getColorNames () const
	Get the names of all colors. More...

size_t	getNbColors () const
	Get the number of colors in the graph. More...

Public Member Functions inherited from CompactedDBG< DataAccessor< void >, DataStorage< void > >
	CompactedDBG (const int kmer_length=31, const int minimizer_length=-1)
	Constructor (set up an empty compacted dBG). More...

	CompactedDBG (const CompactedDBG< U, G > &o)
	Copy constructor (copy a compacted de Bruijn graph). More...

	CompactedDBG (CompactedDBG< U, G > &&o)
	Move constructor (move a compacted de Bruijn graph). More...

virtual	~CompactedDBG ()
	Destructor.

CompactedDBG< U, G > &	operator= (const CompactedDBG< U, G > &o)
	Copy assignment operator (copy a compacted de Bruijn graph). More...

CompactedDBG< U, G > &	operator= (CompactedDBG< U, G > &&o)
	Move assignment operator (move a compacted de Bruijn graph). More...

CompactedDBG< U, G > &	operator+= (const CompactedDBG< U, G > &o)
	Addition assignment operator (merge a compacted de Bruijn graph). More...

bool	operator== (const CompactedDBG< U, G > &o) const
	Equality operator. More...

bool	operator!= (const CompactedDBG< U, G > &o) const
	Inequality operator. More...

void	clear ()
	Clear the graph: empty the graph and reset its parameters.

bool	build (CDBG_Build_opt &opt)
	Build the Compacted de Bruijn graph. More...

bool	simplify (const bool delete_short_isolated_unitigs=true, const bool clip_short_tips=true, const bool verbose=false)
	Simplify the Compacted de Bruijn graph: clip short (< 2k length) tips and/or delete short (< 2k length) isolated unitigs. More...

bool	write (const string &output_fn, const size_t nb_threads=1, const bool GFA_output=true, const bool FASTA_output=false, const bool BFG_output=false, const bool write_index_file=true, const bool compressed_output=false, const bool verbose=false) const
	Write the Compacted de Bruijn graph to disk (GFA1 format). More...

bool	read (const string &input_graph_fn, const size_t nb_threads=1, const bool verbose=false)
	Load a Compacted de Bruijn graph from disk (GFA1 or FASTA format). More...

bool	read (const string &input_graph_fn, const string &input_index_fn, const size_t nb_threads=1, const bool verbose=false)
	Read a Compacted de Bruijn graph from disk (GFA1, FASTA or BFG format) using an index file (BFI format). More...

UnitigMap< U, G >	find (const Kmer &km, const bool extremities_only=false)
	Find the unitig containing the queried k-mer in the Compacted de Bruijn graph. More...

const_UnitigMap< U, G >	find (const Kmer &km, const bool extremities_only=false) const
	Find the unitig containing the queried k-mer in the Compacted de Bruijn graph. More...

UnitigMap< U, G >	findUnitig (const char *s, const size_t pos, const size_t len)
	Find the unitig containing the k-mer starting at a given position in a query sequence and extends the mapping (if the k-mer is found, the function extends the mapping from the k-mer as long as the query sequence and the unitig matches). More...

const_UnitigMap< U, G >	findUnitig (const char *s, const size_t pos, const size_t len) const
	Find the unitig containing the k-mer starting at a given position in a query sequence and extends the mapping (if the k-mer is found, the function extends the mapping from the k-mer as long as the query sequence and the unitig matches). More...

vector< pair< size_t, UnitigMap< U, G > > >	searchSequence (const string &s, const bool exact, const bool insertion, const bool deletion, const bool substitution, const bool or_exclusive_match=false)
	Performs exact and/or inexact search of the k-mers of a sequence query in the Compacted de Bruijn graph. More...

vector< pair< size_t, const_UnitigMap< U, G > > >	searchSequence (const string &s, const bool exact, const bool insertion, const bool deletion, const bool substitution, const bool or_exclusive_match=false) const
	Performs exact and/or inexact search of the k-mers of a sequence query in the Compacted de Bruijn graph. More...

bool	add (const string &seq, const bool verbose=false)
	Add a sequence to the Compacted de Bruijn graph. More...

bool	remove (const const_UnitigMap< U, G > &um, const bool verbose=false)
	Remove a unitig from the Compacted de Bruijn graph. More...

bool	merge (const CompactedDBG &o, const size_t nb_threads=1, const bool verbose=false)
	Merge a compacted de Bruijn graph. More...

bool	merge (const vector< CompactedDBG > &v, const size_t nb_threads=1, const bool verbose=false)
	Merge multiple compacted de Bruijn graphs. More...

iterator	begin ()
	Create an iterator to the first unitig of the Compacted de Bruijn graph (unitigs are NOT sorted lexicographically). More...

const_iterator	begin () const
	Create an constant iterator to the first unitig of the Compacted de Bruijn graph (unitigs are NOT sorted lexicographically). More...

iterator	end ()
	Create an iterator to the "past-the-last" unitig of the Compacted de Bruijn graph (unitigs are NOT sorted lexicographically). More...

const_iterator	end () const
	Create a constant iterator to the "past-the-last" unitig of the Compacted de Bruijn graph (unitigs are NOT sorted lexicographically). More...

size_t	length () const
	Return the sum of the unitigs length. More...

size_t	nbKmers () const
	Return the number of k-mers in the graph. More...

bool	isInvalid () const
	Return a boolean indicating if the graph is invalid (wrong input parameters/files, error occurring during a method, etc.). More...

int	getK () const
	Return the length of k-mers of the graph. More...

int	getG () const
	Return the length of minimizers of the graph. More...

size_t	size () const
	Return the number of unitigs in the graph. More...

G *	getData ()
	Return a pointer to the graph data. More...

const G *	getData () const
	Return a constant pointer to the graph data. More...

Additional Inherited Members
Public Types inherited from CompactedDBG< DataAccessor< void >, DataStorage< void > >
typedef unitigIterator< U, G, false >	iterator
	An iterator for the unitigs of the graph. More...

typedef unitigIterator< U, G, true >	const_iterator
	A constant iterator for the unitigs of the graph. More...

Detailed Description

template<typename Unitig_data_t = void>
class ColoredCDBG< Unitig_data_t >

Represent a Colored and Compacted de Bruijn graph.

The class inherits from CompactedDBG which means that all public functions available with CompactedDBG are also available with ColoredCDBG.

ColoredCDBG<> ccdbg_1; // No unitig data
ColoredCDBG<void> ccdbg_2; // Equivalent to previous notation
ColoredCDBG<MyUnitigData> ccdbg_3; // An object of type MyUnitigData will be associated with each unitig (along the colors)

If data are to be associated with the unitigs, these data must be wrapped into a class that inherits from the abstract class CCDBG_Data_t, such as in:

class MyUnitigData : public CCDBG_Data_t<MyUnitigData> { ... };

CompactedDBG<MyUnitigData> cdbg;

Because CCDBG_Data_t is an abstract class, all the methods of the base class (CCDBG_Data_t) must be implemented in your wrapper (the derived class, aka MyUnitigData in this example). IMPORTANT: If you do not implement those methods, default ones that have no effects will be applied.

Constructor & Destructor Documentation

◆ ColoredCDBG() [1/3]

template<typename Unitig_data_t = void>

ColoredCDBG< Unitig_data_t >::ColoredCDBG	(	int	kmer_length = `31`,
		int	minimizer_length = `-1`
	)

Constructor (set up an empty colored cdBG).

Parameters

kmer_length	is the length k of k-mers used in the graph (each unitig is of length at least k).
minimizer_length	is the length g of minimizers (g < k) used in the graph.

◆ ColoredCDBG() [2/3]

template<typename Unitig_data_t = void>

ColoredCDBG< Unitig_data_t >::ColoredCDBG ( const ColoredCDBG< Unitig_data_t > & o )

Copy constructor (copy a colored cdBG).

This function is expensive in terms of time and memory as the content of a colored and compacted de Bruijn graph is copied. After the call to this function, the same graph exists twice in memory.

Parameters

o	is a constant reference to the colored and compacted de Bruijn graph to copy.

◆ ColoredCDBG() [3/3]

template<typename Unitig_data_t = void>

ColoredCDBG< Unitig_data_t >::ColoredCDBG ( ColoredCDBG< Unitig_data_t > && o )

Move constructor (move a colored cdBG).

The content of o is moved ("transfered") to a new colored and compacted de Bruijn graph. The colored and compacted de Bruijn graph referenced by o will be empty after the call to this constructor.

Parameters

o	is a reference on a reference to the colored and compacted de Bruijn graph to move.

Member Function Documentation

◆ buildColors()

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::buildColors ( const CCDBG_Build_opt & opt )

Map the colors to the unitigs.

This is done by reading the input files and querying the graph. If a color filename is provided in opt.filename_colors_in, colors are loaded from that file instead.

Parameters

opt	is a structure from which the members are parameters of this function. See CCDBG_Build_opt.

Returns: boolean indicating if the colors have been mapped successfully.

◆ buildGraph()

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::buildGraph ( const CCDBG_Build_opt & opt )

Build the Colored and compacted de Bruijn graph (only the unitigs).

A call to ColoredCDBG::mapColors is required afterwards to map colors to unitigs.

Parameters

opt	is a structure from which the members are parameters of this function. See CCDBG_Build_opt.

Returns: boolean indicating if the graph has been built successfully.

◆ getColorName()

template<typename Unitig_data_t = void>

string ColoredCDBG< Unitig_data_t >::getColorName ( const size_t color_id ) const

Get the name of a color.

As colors match the input files, the color names match the input filenames.

Returns: a string which is either a color name or an empty string if the color ID is invalid or if the colors have not yet been mapped to the unitigs.

◆ getColorNames()

template<typename Unitig_data_t = void>

vector<string> ColoredCDBG< Unitig_data_t >::getColorNames ( ) const

Get the names of all colors.

As colors match the input files, the color names match the input filenames.

Returns: a vector of strings for which each string is either a color name or an empty string if the color ID is invalid or if the colors have not yet been mapped to the unitigs.

◆ getNbColors()

template<typename Unitig_data_t = void>

size_t ColoredCDBG< Unitig_data_t >::getNbColors ( ) const

inline

Get the number of colors in the graph.

Returns: the number of colors in the graph.

◆ merge() [1/4]

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::merge	(	ColoredCDBG< Unitig_data_t > &&	o,
		const size_t	nb_threads = `1`,
		const bool	verbose = `false`
	)

Merge and clear a colored and compacted de Bruijn graph.

After merging, all unitigs and colors of the input graph have been added to and compacted with the current colored and compacted de Bruijn graph (this). The input graph is cleared before the function returns. If the unitigs of the input graph had data of type "MyUnitigData" associated, they have been added to the current colored and compacted de Bruijn graph using the functions of the class MyUnitigData which are also present in its base class CCDBG_Data_t<MyUnitigData>. Note that if multiple colored and compacted de Bruijn graphs have to be merged, it is more efficient to call ColoredCDBG::merge with a vector of ColoredCDBG as input.

Parameters

o	is a reference on a reference to the colored and compacted de Bruijn graph to merge. It can be obtained using std::move(). After merging, the graph pointed by o is cleared.
nb_threads	is an integer indicating how many threads can be used during the merging.
verbose	is a boolean indicating if information messages must be printed during the execution of the function.

Returns: a boolean indicating if the graph has been successfully merged.

◆ merge() [2/4]

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::merge	(	const ColoredCDBG< Unitig_data_t > &	o,
		const size_t	nb_threads = `1`,
		const bool	verbose = `false`
	)

Merge a colored and compacted de Bruijn graph.

After merging, all unitigs and colors of the input graph have been added to and compacted with the current colored and compacted de Bruijn graph (this). If the unitigs of the input graph had data of type "MyUnitigData" associated, they have been added to the current colored and compacted de Bruijn graph using the functions of the class MyUnitigData which are also present in its base class CCDBG_Data_t<MyUnitigData>. Note that if multiple colored and compacted de Bruijn graphs have to be merged, it is more efficient to call ColoredCDBG::merge with a vector of ColoredCDBG as input.

Parameters

o	is a constant reference to the colored and compacted de Bruijn graph to merge.
nb_threads	is an integer indicating how many threads can be used during the merging.
verbose	is a boolean indicating if information messages must be printed during the execution of the function.

Returns: a boolean indicating if the graph has been successfully merged.

◆ merge() [3/4]

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::merge	(	const vector< ColoredCDBG< Unitig_data_t > > &	v,
		const size_t	nb_threads = `1`,
		const bool	verbose = `false`
	)

Merge multiple colored and compacted de Bruijn graphs.

After merging, all unitigs and colors of the input colored and compacted de Bruijn graphs have been added to and compacted with the current colored and compacted de Bruijn graph (this). If the unitigs had data of type "MyUnitigData" associated, they have been added to the current colored and compacted de Bruijn graph using the functions of the class MyUnitigData which are also present in its base class CCDBG_Data_t<MyUnitigData>.

Parameters

v	is a constant reference to a vector of colored and compacted de Bruijn graphs to merge.
nb_threads	is an integer indicating how many threads can be used during the merging.
verbose	is a boolean indicating if information messages must be printed during the execution of the function.

Returns: a boolean indicating if the graphs have been successfully merged.

◆ merge() [4/4]

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::merge	(	vector< ColoredCDBG< Unitig_data_t > > &&	v,
		const size_t	nb_threads = `1`,
		const bool	verbose = `false`
	)

Merge and clear multiple colored and compacted de Bruijn graphs.

After merging, all unitigs and colors of the input colored and compacted de Bruijn graphs have been added to and compacted with the current colored and compacted de Bruijn graph (this). The input graphs are cleared before the function returns. If the input unitigs had data of type "MyUnitigData" associated, they have been added to the current colored and compacted de Bruijn graph using the functions of the class MyUnitigData which are also present in its base class CCDBG_Data_t<MyUnitigData>.

Parameters

v	is a reference on a reference to a vector of colored and compacted de Bruijn graphs to merge. It can be obtained using std::move(). After merging, the graphs in v are cleared.
nb_threads	is an integer indicating how many threads can be used during the merging.
verbose	is a boolean indicating if information messages must be printed during the execution of the function.

Returns: a boolean indicating if the graphs have been successfully merged.

◆ operator!=()

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::operator!= ( const ColoredCDBG< Unitig_data_t > & o ) const

inline

Inequality operator.

Returns: a boolean indicating if two compacted de Bruijn graphs have different colored unitigs (does not compare the data associated with the unitigs).

◆ operator+=()

template<typename Unitig_data_t = void>

ColoredCDBG& ColoredCDBG< Unitig_data_t >::operator+= ( const ColoredCDBG< Unitig_data_t > & o )

Addition assignment operator (merge a colored cdBG).

After merging, all unitigs and colors of o have been added to and compacted with the current colored and compacted de Bruijn graph (this). If the unitigs of o had data of type "MyUnitigData" associated, they have been added to the current colored and compacted de Bruijn graph using the functions of the class MyUnitigData which are in base class CCDBG_Data_t<MyUnitigData>. This function is similar to ColoredCDBG::merge except that it uses only one thread while ColoredCDBG::merge can work with multiple threads (number of threads provided as a parameter). Note that if multiple colored and compacted de Bruijn graphs have to be merged, it is more efficient to call ColoredCDBG::merge with a vector of ColoredCDBG as input.

Parameters

o	is a constant reference to the colored and compacted de Bruijn graph to merge.

Returns: a reference to the current colored and compacted de Bruijn after merging.

◆ operator=() [1/2]

template<typename Unitig_data_t = void>

ColoredCDBG& ColoredCDBG< Unitig_data_t >::operator= ( ColoredCDBG< Unitig_data_t > && o )

Move assignment operator (move a colored cdBG).

The content of o is moved ("transfered") to a new colored and compacted de Bruijn graph. The colored and compacted de Bruijn graph referenced by o will be empty after the call to this operator.

Parameters

o	is a reference on a reference to the colored and compacted de Bruijn graph to move.

Returns: a reference to the colored and compacted de Bruijn which has (and owns) the content of o.

◆ operator=() [2/2]

template<typename Unitig_data_t = void>

ColoredCDBG& ColoredCDBG< Unitig_data_t >::operator= ( const ColoredCDBG< Unitig_data_t > & o )

Copy assignment operator (copy a colored cdBG).

This function is expensive in terms of time and memory as the content of a colored and compacted de Bruijn graph is copied. After the call to this function, the same graph exists twice in memory.

Parameters

o	is a constant reference to the colored and compacted de Bruijn graph to copy.

Returns: a reference to the colored and compacted de Bruijn which is the copy.

◆ operator==()

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::operator== ( const ColoredCDBG< Unitig_data_t > & o ) const

Equality operator.

Returns: a boolean indicating if two compacted de Bruijn graphs have the same colored unitigs (does not compare the data associated with the unitigs).

◆ read() [1/2]

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::read	(	const string &	input_graph_fn,
		const string &	input_colors_fn,
		const size_t	nb_threads = `1`,
		const bool	verbose = `false`
	)

Read a colored and compacted de Bruijn graph from disk.

The graph (in GFA, FASTA or BFG format) must have been produced by Bifrost. By default, the function detects if an index file (BFI format) exists for the input graph and will use it to load the graph. Otherwise, reading the graph will be much slower than function read() with the index filename in input parameter.

Parameters

input_graph_fn	is a string which is the prefix of the graph filename to read
input_colors_fn	is a string which is the prefix of the color filename to read
nb_threads	is the number of threads that can be used to read the graph and its colors from disk.
verbose	is a boolean indicating if information messages are printed during reading (true) or not (false).

Returns: a boolean indicating if the graph was successfully read.

◆ read() [2/2]

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::read	(	const string &	input_graph_fn,
		const string &	input_index_fn,
		const string &	input_colors_fn,
		const size_t	nb_threads = `1`,
		const bool	verbose = `false`
	)

Read a colored and compacted de Bruijn graph from disk using an index file.

The graph (in GFA, FASTA or BFG format) must have been produced by Bifrost.

Parameters

input_graph_fn	is a string which is the prefix of the graph filename to read
input_index_fn	is a string which is the prefix of the index filename to read
input_colors_fn	is a string which is the prefix of the color filename to read
nb_threads	is the number of threads that can be used to read the graph and its colors from disk.
verbose	is a boolean indicating if information messages are printed during reading (true) or not (false).

Returns: a boolean indicating if the graph was successfully read.

◆ write()

template<typename Unitig_data_t = void>

bool ColoredCDBG< Unitig_data_t >::write	(	const string &	prefix_output_fn,
		const size_t	nb_threads = `1`,
		const bool	write_index_file = `true`,
		const bool	compress_output = `false`,
		const bool	verbose = `false`
	)		const

Write a colored and compacted de Bruijn graph to disk.

Parameters

prefix_output_fn	is a string which is the prefix of the filename for the two files that are going to be written to disk. Assuming the prefix is "XXX", two files "XXX.gfa" and "XXX.color.bfg" will be written to disk.
nb_threads	is the number of threads that can be used to write the graph to disk.
write_meta_file	indicates if a graph meta file is written to disk. Graph meta files enable faster graph loading.
compressed_output	indicates if the output file is compressed.
verbose	is a boolean indicating if information message are printed during writing (true) or not (false).

Returns: a boolean indicating if the graph was successfully written.

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

src/ColoredCDBG.hpp

Public Member Functions

Additional Inherited Members

Detailed Description

template<typename Unitig_data_t = void> class ColoredCDBG< Unitig_data_t >

Constructor & Destructor Documentation

◆ ColoredCDBG() [1/3]

◆ ColoredCDBG() [2/3]

◆ ColoredCDBG() [3/3]

Member Function Documentation

◆ buildColors()

◆ buildGraph()

◆ getColorName()

◆ getColorNames()

◆ getNbColors()

◆ merge() [1/4]

◆ merge() [2/4]

◆ merge() [3/4]

◆ merge() [4/4]

◆ operator!=()

◆ operator+=()

◆ operator=() [1/2]

◆ operator=() [2/2]

◆ operator==()

◆ read() [1/2]

◆ read() [2/2]

◆ write()

template<typename Unitig_data_t = void>
class ColoredCDBG< Unitig_data_t >